sysmocom
/
asterisk
11
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
asterisk/res/res_sorcery_realtime.c

383 lines
12 KiB
C
Raw Permalink Normal View History

Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
/*
* Asterisk -- An open source telephony toolkit.
*
* Copyright (C) 2013, Digium, Inc.
*
* Joshua Colp <jcolp@digium.com>
*
* See http://www.asterisk.org for more information about
* the Asterisk project. Please do not directly contact
* any of the maintainers of this project for assistance;
* the project provides a web site, mailing lists and IRC
* channels for your use.
*
* This program is free software, distributed under the terms of
* the GNU General Public License Version 2. See the LICENSE file
* at the top of the source tree.
*/
/*!
* \file
*
* \brief Sorcery Realtime Object Wizard
*
* \author Joshua Colp <jcolp@digium.com>
*/
/*** MODULEINFO
<support_level>core</support_level>
***/
#include "asterisk.h"
#include <regex.h>
#include "asterisk/module.h"
#include "asterisk/sorcery.h"
/*! \brief They key field used to store the unique identifier for the object */
#define UUID_FIELD "id"
sorcery/res_pjsip: Refactor for realtime performance There were a number of places in the res_pjsip stack that were getting all endpoints or all aors, and then filtering them locally. A good example is pjsip_options which, on startup, retrieves all endpoints, then the aors for those endpoints, then tests the aors to see if the qualify_frequency is > 0. One issue was that it never did anything with the endpoints other than retrieve the aors so we probably could have skipped a step and just retrieved all aors. But nevermind. This worked reasonably well with local config files but with a realtime backend and thousands of objects, this was a nightmare. The issue really boiled down to the fact that while realtime supports predicates that are passed to the database engine, the non-realtime sorcery backends didn't. They do now. The realtime engines have a scheme for doing simple comparisons. They take in an ast_variable (or list) for matching, and the name of each variable can contain an operator. For instance, a name of "qualify_frequency >" and a value of "0" would create a SQL predicate that looks like "where qualify_frequency > '0'". If there's no operator after the name, the engines add an '=' so a simple name of "qualify_frequency" and a value of "10" would return exact matches. The non-realtime backends decide whether to include an object in a result set by calling ast_sorcery_changeset_create on every object in the internal container. However, ast_sorcery_changeset_create only does exact string matches though so a name of "qualify_frequency >" and a value of "0" returns nothing because the literal "qualify_frequency >" doesn't match any name in the objset set. So, the real task was to create a generic string matcher that can take a left value, operator and a right value and perform the match. To that end, strings.c has a new ast_strings_match(left, operator, right) function. Left and right are the strings to operate on and the operator can be a string containing any of the following: = (or NULL or ""), !=, >, >=, <, <=, like or regex. If the operator is like or regex, the right string should be a %-pattern or a regex expression. If both left and right can be converted to float, then a numeric comparison is performed, otherwise a string comparison is performed. To use this new function on ast_variables, 2 new functions were added to config.c. One that compares 2 ast_variables, and one that compares 2 ast_variable lists. The former is useful when you want to compare 2 ast_variables that happen to be in a list but don't want to traverse the list. The latter will traverse the right list and return true if all the variables in it match the left list. Now, the backends' fields_cmp functions call ast_variable_lists_match instead of ast_sorcery_changeset_create and they can now process the same syntax as the realtime engines. The realtime backend just passes the variable list unaltered to the engine. The only gotcha is that there's no common realtime engine support for regex so that's been noted in the api docs for ast_sorcery_retrieve_by_fields. Only one more change to sorcery was done... A new config flag "allow_unqualified_fetch" was added to reg_sorcery_realtime. "no": ignore fetches if no predicate fields were supplied. "error": same as no but emit an error. (good for testing) "yes": allow (the default); "warn": allow but emit a warning. (good for testing) Now on to res_pjsip... pjsip_options was modified to retrieve aors with qualify_frequency > 0 rather than all endpoints then all aors. Not only was this a big improvement in realtime retrieval but even for config files there's an improvement because we're not going through endpoints anymore. res_pjsip_mwi was modified to retieve only endpoints with something in the mailboxes field instead of all endpoints then testing mailboxes. res_pjsip_registrar_expire was completely refactored. It was retrieving all contacts then setting up scheduler entries to check for expiration. Now, it's a single thread (like keepalive) that periodically retrieves only contacts whose expiration time is < now and deletes them. A new contact_expiration_check_interval was added to global with a default of 30 seconds. Ross Beer reports that with this patch, his Asterisk startup time dropped from around an hour to under 30 seconds. There are still objects that can't be filtered at the database like identifies, transports, and registrations. These are not going to be anywhere near as numerous as endpoints, aors, auths, contacts however. Back to allow_unqualified_fetch. If this is set to yes and you have a very large number of objects in the database, the pjsip CLI commands will attempt to retrive ALL of them if not qualified with a LIKE. Worse, if you type "pjsip show endpoint <tab>" guess what's going to happen? :) Having a cache helps but all the objects will have to be retrieved at least once to fill the cache. Setting allow_unqualified_fetch=no prevents the mass retrieve and should be used on endpoints, auths, aors, and contacts. It should NOT be used for identifies, registrations and transports since these MUST be retrieved in bulk. Example sorcery.conf: [res_pjsip] endpoint=config,pjsip.conf,criteria=type=endpoint endpoint=realtime,ps_endpoints,allow_unqualified_fetch=error ASTERISK-25826 #close Reported-by: Ross Beer Tested-by: Ross Beer Change-Id: Id2691e447db90892890036e663aaf907b2dc1c67
2016-03-08 21:55:30 +00:00
enum unqualified_fetch {
UNQUALIFIED_FETCH_NO,
UNQUALIFIED_FETCH_WARN,
UNQUALIFIED_FETCH_YES,
UNQUALIFIED_FETCH_ERROR,
};
struct sorcery_config {
enum unqualified_fetch fetch;
char family[];
};
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
static void *sorcery_realtime_open(const char *data);
static int sorcery_realtime_create(const struct ast_sorcery *sorcery, void *data, void *object);
static void *sorcery_realtime_retrieve_id(const struct ast_sorcery *sorcery, void *data, const char *type, const char *id);
static void *sorcery_realtime_retrieve_fields(const struct ast_sorcery *sorcery, void *data, const char *type, const struct ast_variable *fields);
static void sorcery_realtime_retrieve_multiple(const struct ast_sorcery *sorcery, void *data, const char *type, struct ao2_container *objects,
const struct ast_variable *fields);
static void sorcery_realtime_retrieve_regex(const struct ast_sorcery *sorcery, void *data, const char *type, struct ao2_container *objects, const char *regex);
sorcery: Add ast_sorcery_retrieve_by_prefix() Some consumers of the sorcery API use ast_sorcery_retrieve_by_regex only so that they can anchor the potential match as a prefix and not because they truly need regular expressions. Rather than using regular expressions for simple prefix lookups, add a new operation - ast_sorcery_retrieve_by_prefix - that does them. Change-Id: I56f4e20ba1154bd52281f995c27a429a854f6a79
2017-11-09 14:21:38 +00:00
static void sorcery_realtime_retrieve_prefix(const struct ast_sorcery *sorcery, void *data, const char *type,
struct ao2_container *objects, const char *prefix, const size_t prefix_len);
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
static int sorcery_realtime_update(const struct ast_sorcery *sorcery, void *data, void *object);
static int sorcery_realtime_delete(const struct ast_sorcery *sorcery, void *data, void *object);
static void sorcery_realtime_close(void *data);
static struct ast_sorcery_wizard realtime_object_wizard = {
.name = "realtime",
.open = sorcery_realtime_open,
.create = sorcery_realtime_create,
.retrieve_id = sorcery_realtime_retrieve_id,
.retrieve_fields = sorcery_realtime_retrieve_fields,
.retrieve_multiple = sorcery_realtime_retrieve_multiple,
.retrieve_regex = sorcery_realtime_retrieve_regex,
sorcery: Add ast_sorcery_retrieve_by_prefix() Some consumers of the sorcery API use ast_sorcery_retrieve_by_regex only so that they can anchor the potential match as a prefix and not because they truly need regular expressions. Rather than using regular expressions for simple prefix lookups, add a new operation - ast_sorcery_retrieve_by_prefix - that does them. Change-Id: I56f4e20ba1154bd52281f995c27a429a854f6a79
2017-11-09 14:21:38 +00:00
.retrieve_prefix = sorcery_realtime_retrieve_prefix,
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
.update = sorcery_realtime_update,
.delete = sorcery_realtime_delete,
.close = sorcery_realtime_close,
};
static int sorcery_realtime_create(const struct ast_sorcery *sorcery, void *data, void *object)
{
sorcery/res_pjsip: Refactor for realtime performance There were a number of places in the res_pjsip stack that were getting all endpoints or all aors, and then filtering them locally. A good example is pjsip_options which, on startup, retrieves all endpoints, then the aors for those endpoints, then tests the aors to see if the qualify_frequency is > 0. One issue was that it never did anything with the endpoints other than retrieve the aors so we probably could have skipped a step and just retrieved all aors. But nevermind. This worked reasonably well with local config files but with a realtime backend and thousands of objects, this was a nightmare. The issue really boiled down to the fact that while realtime supports predicates that are passed to the database engine, the non-realtime sorcery backends didn't. They do now. The realtime engines have a scheme for doing simple comparisons. They take in an ast_variable (or list) for matching, and the name of each variable can contain an operator. For instance, a name of "qualify_frequency >" and a value of "0" would create a SQL predicate that looks like "where qualify_frequency > '0'". If there's no operator after the name, the engines add an '=' so a simple name of "qualify_frequency" and a value of "10" would return exact matches. The non-realtime backends decide whether to include an object in a result set by calling ast_sorcery_changeset_create on every object in the internal container. However, ast_sorcery_changeset_create only does exact string matches though so a name of "qualify_frequency >" and a value of "0" returns nothing because the literal "qualify_frequency >" doesn't match any name in the objset set. So, the real task was to create a generic string matcher that can take a left value, operator and a right value and perform the match. To that end, strings.c has a new ast_strings_match(left, operator, right) function. Left and right are the strings to operate on and the operator can be a string containing any of the following: = (or NULL or ""), !=, >, >=, <, <=, like or regex. If the operator is like or regex, the right string should be a %-pattern or a regex expression. If both left and right can be converted to float, then a numeric comparison is performed, otherwise a string comparison is performed. To use this new function on ast_variables, 2 new functions were added to config.c. One that compares 2 ast_variables, and one that compares 2 ast_variable lists. The former is useful when you want to compare 2 ast_variables that happen to be in a list but don't want to traverse the list. The latter will traverse the right list and return true if all the variables in it match the left list. Now, the backends' fields_cmp functions call ast_variable_lists_match instead of ast_sorcery_changeset_create and they can now process the same syntax as the realtime engines. The realtime backend just passes the variable list unaltered to the engine. The only gotcha is that there's no common realtime engine support for regex so that's been noted in the api docs for ast_sorcery_retrieve_by_fields. Only one more change to sorcery was done... A new config flag "allow_unqualified_fetch" was added to reg_sorcery_realtime. "no": ignore fetches if no predicate fields were supplied. "error": same as no but emit an error. (good for testing) "yes": allow (the default); "warn": allow but emit a warning. (good for testing) Now on to res_pjsip... pjsip_options was modified to retrieve aors with qualify_frequency > 0 rather than all endpoints then all aors. Not only was this a big improvement in realtime retrieval but even for config files there's an improvement because we're not going through endpoints anymore. res_pjsip_mwi was modified to retieve only endpoints with something in the mailboxes field instead of all endpoints then testing mailboxes. res_pjsip_registrar_expire was completely refactored. It was retrieving all contacts then setting up scheduler entries to check for expiration. Now, it's a single thread (like keepalive) that periodically retrieves only contacts whose expiration time is < now and deletes them. A new contact_expiration_check_interval was added to global with a default of 30 seconds. Ross Beer reports that with this patch, his Asterisk startup time dropped from around an hour to under 30 seconds. There are still objects that can't be filtered at the database like identifies, transports, and registrations. These are not going to be anywhere near as numerous as endpoints, aors, auths, contacts however. Back to allow_unqualified_fetch. If this is set to yes and you have a very large number of objects in the database, the pjsip CLI commands will attempt to retrive ALL of them if not qualified with a LIKE. Worse, if you type "pjsip show endpoint <tab>" guess what's going to happen? :) Having a cache helps but all the objects will have to be retrieved at least once to fill the cache. Setting allow_unqualified_fetch=no prevents the mass retrieve and should be used on endpoints, auths, aors, and contacts. It should NOT be used for identifies, registrations and transports since these MUST be retrieved in bulk. Example sorcery.conf: [res_pjsip] endpoint=config,pjsip.conf,criteria=type=endpoint endpoint=realtime,ps_endpoints,allow_unqualified_fetch=error ASTERISK-25826 #close Reported-by: Ross Beer Tested-by: Ross Beer Change-Id: Id2691e447db90892890036e663aaf907b2dc1c67
2016-03-08 21:55:30 +00:00
struct sorcery_config *config = data;
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
RAII_VAR(struct ast_variable *, fields, ast_sorcery_objectset_create(sorcery, object), ast_variables_destroy);
struct ast_variable *id = ast_variable_new(UUID_FIELD, ast_sorcery_object_get_id(object), "");
if (!fields || !id) {
ast_variables_destroy(id);
return -1;
}
/* Place the identifier at the front for sanity sake */
id->next = fields;
fields = id;
sorcery/res_pjsip: Refactor for realtime performance There were a number of places in the res_pjsip stack that were getting all endpoints or all aors, and then filtering them locally. A good example is pjsip_options which, on startup, retrieves all endpoints, then the aors for those endpoints, then tests the aors to see if the qualify_frequency is > 0. One issue was that it never did anything with the endpoints other than retrieve the aors so we probably could have skipped a step and just retrieved all aors. But nevermind. This worked reasonably well with local config files but with a realtime backend and thousands of objects, this was a nightmare. The issue really boiled down to the fact that while realtime supports predicates that are passed to the database engine, the non-realtime sorcery backends didn't. They do now. The realtime engines have a scheme for doing simple comparisons. They take in an ast_variable (or list) for matching, and the name of each variable can contain an operator. For instance, a name of "qualify_frequency >" and a value of "0" would create a SQL predicate that looks like "where qualify_frequency > '0'". If there's no operator after the name, the engines add an '=' so a simple name of "qualify_frequency" and a value of "10" would return exact matches. The non-realtime backends decide whether to include an object in a result set by calling ast_sorcery_changeset_create on every object in the internal container. However, ast_sorcery_changeset_create only does exact string matches though so a name of "qualify_frequency >" and a value of "0" returns nothing because the literal "qualify_frequency >" doesn't match any name in the objset set. So, the real task was to create a generic string matcher that can take a left value, operator and a right value and perform the match. To that end, strings.c has a new ast_strings_match(left, operator, right) function. Left and right are the strings to operate on and the operator can be a string containing any of the following: = (or NULL or ""), !=, >, >=, <, <=, like or regex. If the operator is like or regex, the right string should be a %-pattern or a regex expression. If both left and right can be converted to float, then a numeric comparison is performed, otherwise a string comparison is performed. To use this new function on ast_variables, 2 new functions were added to config.c. One that compares 2 ast_variables, and one that compares 2 ast_variable lists. The former is useful when you want to compare 2 ast_variables that happen to be in a list but don't want to traverse the list. The latter will traverse the right list and return true if all the variables in it match the left list. Now, the backends' fields_cmp functions call ast_variable_lists_match instead of ast_sorcery_changeset_create and they can now process the same syntax as the realtime engines. The realtime backend just passes the variable list unaltered to the engine. The only gotcha is that there's no common realtime engine support for regex so that's been noted in the api docs for ast_sorcery_retrieve_by_fields. Only one more change to sorcery was done... A new config flag "allow_unqualified_fetch" was added to reg_sorcery_realtime. "no": ignore fetches if no predicate fields were supplied. "error": same as no but emit an error. (good for testing) "yes": allow (the default); "warn": allow but emit a warning. (good for testing) Now on to res_pjsip... pjsip_options was modified to retrieve aors with qualify_frequency > 0 rather than all endpoints then all aors. Not only was this a big improvement in realtime retrieval but even for config files there's an improvement because we're not going through endpoints anymore. res_pjsip_mwi was modified to retieve only endpoints with something in the mailboxes field instead of all endpoints then testing mailboxes. res_pjsip_registrar_expire was completely refactored. It was retrieving all contacts then setting up scheduler entries to check for expiration. Now, it's a single thread (like keepalive) that periodically retrieves only contacts whose expiration time is < now and deletes them. A new contact_expiration_check_interval was added to global with a default of 30 seconds. Ross Beer reports that with this patch, his Asterisk startup time dropped from around an hour to under 30 seconds. There are still objects that can't be filtered at the database like identifies, transports, and registrations. These are not going to be anywhere near as numerous as endpoints, aors, auths, contacts however. Back to allow_unqualified_fetch. If this is set to yes and you have a very large number of objects in the database, the pjsip CLI commands will attempt to retrive ALL of them if not qualified with a LIKE. Worse, if you type "pjsip show endpoint <tab>" guess what's going to happen? :) Having a cache helps but all the objects will have to be retrieved at least once to fill the cache. Setting allow_unqualified_fetch=no prevents the mass retrieve and should be used on endpoints, auths, aors, and contacts. It should NOT be used for identifies, registrations and transports since these MUST be retrieved in bulk. Example sorcery.conf: [res_pjsip] endpoint=config,pjsip.conf,criteria=type=endpoint endpoint=realtime,ps_endpoints,allow_unqualified_fetch=error ASTERISK-25826 #close Reported-by: Ross Beer Tested-by: Ross Beer Change-Id: Id2691e447db90892890036e663aaf907b2dc1c67
2016-03-08 21:55:30 +00:00
return (ast_store_realtime_fields(config->family, fields) <= 0) ? -1 : 0;
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
}
Remove as much trailing whitespace as possible. Change-Id: I873c1c6d00f447269bd841494459efccdd2c19c0
2017-12-22 14:23:22 +00:00
/*! \brief Internal helper function which returns a filtered objectset.
Make res_sorcery_realtime filter unknown retrieved results. When retrieving data from a database or other realtime backend, it's quite possible to retrieve variables that Asterisk does not care about but that are legitimate to exist. Asterisk does not need to throw a hissy fit when these variables are encountered but rather just filter them out. Review: https://reviewboard.asterisk.org/r/3305 ........ Merged revisions 410187 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@410207 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-03-07 21:23:39 +00:00
*
* The following are filtered out of the objectset:
* \li The id field. This is returned to the caller in an out parameter.
* \li Fields that are not registered with sorcery.
*
* \param objectset Objectset to filter.
* \param[out] id The ID of the sorcery object, as found in the objectset.
* \param sorcery The sorcery instance that is requesting an objectset.
* \param type The object type
*
* \return The filtered objectset
*/
static struct ast_variable *sorcery_realtime_filter_objectset(struct ast_variable *objectset, struct ast_variable **id,
const struct ast_sorcery *sorcery, const char *type)
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
{
Make res_sorcery_realtime filter unknown retrieved results. When retrieving data from a database or other realtime backend, it's quite possible to retrieve variables that Asterisk does not care about but that are legitimate to exist. Asterisk does not need to throw a hissy fit when these variables are encountered but rather just filter them out. Review: https://reviewboard.asterisk.org/r/3305 ........ Merged revisions 410187 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@410207 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-03-07 21:23:39 +00:00
struct ast_variable *previous = NULL, *field = objectset;
struct ast_sorcery_object_type *object_type;
object_type = ast_sorcery_get_object_type(sorcery, type);
if (!object_type) {
ast_log(LOG_WARNING, "Unknown sorcery object type %s. Expect errors\n", type);
/* Continue since we still want to filter out the id */
}
while (field) {
int remove_field = 0;
int delete_field = 0;
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
if (!strcmp(field->name, UUID_FIELD)) {
*id = field;
Make res_sorcery_realtime filter unknown retrieved results. When retrieving data from a database or other realtime backend, it's quite possible to retrieve variables that Asterisk does not care about but that are legitimate to exist. Asterisk does not need to throw a hissy fit when these variables are encountered but rather just filter them out. Review: https://reviewboard.asterisk.org/r/3305 ........ Merged revisions 410187 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@410207 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-03-07 21:23:39 +00:00
remove_field = 1;
} else if (object_type &&
!ast_sorcery_is_object_field_registered(object_type, field->name)) {
ast_debug(1, "Filtering out realtime field '%s' from retrieval\n", field->name);
remove_field = 1;
delete_field = 1;
}
if (remove_field) {
struct ast_variable *removed;
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
if (previous) {
previous->next = field->next;
} else {
objectset = field->next;
}
Make res_sorcery_realtime filter unknown retrieved results. When retrieving data from a database or other realtime backend, it's quite possible to retrieve variables that Asterisk does not care about but that are legitimate to exist. Asterisk does not need to throw a hissy fit when these variables are encountered but rather just filter them out. Review: https://reviewboard.asterisk.org/r/3305 ........ Merged revisions 410187 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@410207 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-03-07 21:23:39 +00:00
removed = field;
field = field->next;
removed->next = NULL;
if (delete_field) {
ast_variables_destroy(removed);
}
} else {
previous = field;
field = field->next;
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
}
}
res_sorcery_realtime.c: Fix crash from NULL sorcery object type. If the sorcery object type is not found a NULL is returned. Unfortunately, sorcery_realtime_filter_objectset() will crash after complaining about not finding the object type and saying to expect errors. * Use ao2_cleanup() instead of ao2_ref() to prevent the crash. ASTERISK-25165 Reported by Corey Farrell Change-Id: Ic3b64453ea3058cb68d5c26d97d4fe7b8eea2e97
2015-11-23 20:27:27 +00:00
ao2_cleanup(object_type);
res_sorcery_realtime: Fix leak of sorcery object type. This prevents a leak of a sorcery object type when realtime sorcery objects are retrieved by fields or when multiple objects are retrieved. The extent of this leak is that sorcery object types would be leaked. These are allocated whenever an object type is registered with sorcery, meaning that on module shutdown, these objects would be leaked. This could be problematic if many reloads were performed, but it is not as severe as if every sorcery object retrieved from realtime were being leaked. ASTERISK-25165 #close Reported by Corey Farrell Change-Id: I625c3b50eee4576670b7eeb013c81ad043b4b4f8
2015-06-29 17:45:02 +00:00
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
return objectset;
}
static void *sorcery_realtime_retrieve_fields(const struct ast_sorcery *sorcery, void *data, const char *type, const struct ast_variable *fields)
{
sorcery/res_pjsip: Refactor for realtime performance There were a number of places in the res_pjsip stack that were getting all endpoints or all aors, and then filtering them locally. A good example is pjsip_options which, on startup, retrieves all endpoints, then the aors for those endpoints, then tests the aors to see if the qualify_frequency is > 0. One issue was that it never did anything with the endpoints other than retrieve the aors so we probably could have skipped a step and just retrieved all aors. But nevermind. This worked reasonably well with local config files but with a realtime backend and thousands of objects, this was a nightmare. The issue really boiled down to the fact that while realtime supports predicates that are passed to the database engine, the non-realtime sorcery backends didn't. They do now. The realtime engines have a scheme for doing simple comparisons. They take in an ast_variable (or list) for matching, and the name of each variable can contain an operator. For instance, a name of "qualify_frequency >" and a value of "0" would create a SQL predicate that looks like "where qualify_frequency > '0'". If there's no operator after the name, the engines add an '=' so a simple name of "qualify_frequency" and a value of "10" would return exact matches. The non-realtime backends decide whether to include an object in a result set by calling ast_sorcery_changeset_create on every object in the internal container. However, ast_sorcery_changeset_create only does exact string matches though so a name of "qualify_frequency >" and a value of "0" returns nothing because the literal "qualify_frequency >" doesn't match any name in the objset set. So, the real task was to create a generic string matcher that can take a left value, operator and a right value and perform the match. To that end, strings.c has a new ast_strings_match(left, operator, right) function. Left and right are the strings to operate on and the operator can be a string containing any of the following: = (or NULL or ""), !=, >, >=, <, <=, like or regex. If the operator is like or regex, the right string should be a %-pattern or a regex expression. If both left and right can be converted to float, then a numeric comparison is performed, otherwise a string comparison is performed. To use this new function on ast_variables, 2 new functions were added to config.c. One that compares 2 ast_variables, and one that compares 2 ast_variable lists. The former is useful when you want to compare 2 ast_variables that happen to be in a list but don't want to traverse the list. The latter will traverse the right list and return true if all the variables in it match the left list. Now, the backends' fields_cmp functions call ast_variable_lists_match instead of ast_sorcery_changeset_create and they can now process the same syntax as the realtime engines. The realtime backend just passes the variable list unaltered to the engine. The only gotcha is that there's no common realtime engine support for regex so that's been noted in the api docs for ast_sorcery_retrieve_by_fields. Only one more change to sorcery was done... A new config flag "allow_unqualified_fetch" was added to reg_sorcery_realtime. "no": ignore fetches if no predicate fields were supplied. "error": same as no but emit an error. (good for testing) "yes": allow (the default); "warn": allow but emit a warning. (good for testing) Now on to res_pjsip... pjsip_options was modified to retrieve aors with qualify_frequency > 0 rather than all endpoints then all aors. Not only was this a big improvement in realtime retrieval but even for config files there's an improvement because we're not going through endpoints anymore. res_pjsip_mwi was modified to retieve only endpoints with something in the mailboxes field instead of all endpoints then testing mailboxes. res_pjsip_registrar_expire was completely refactored. It was retrieving all contacts then setting up scheduler entries to check for expiration. Now, it's a single thread (like keepalive) that periodically retrieves only contacts whose expiration time is < now and deletes them. A new contact_expiration_check_interval was added to global with a default of 30 seconds. Ross Beer reports that with this patch, his Asterisk startup time dropped from around an hour to under 30 seconds. There are still objects that can't be filtered at the database like identifies, transports, and registrations. These are not going to be anywhere near as numerous as endpoints, aors, auths, contacts however. Back to allow_unqualified_fetch. If this is set to yes and you have a very large number of objects in the database, the pjsip CLI commands will attempt to retrive ALL of them if not qualified with a LIKE. Worse, if you type "pjsip show endpoint <tab>" guess what's going to happen? :) Having a cache helps but all the objects will have to be retrieved at least once to fill the cache. Setting allow_unqualified_fetch=no prevents the mass retrieve and should be used on endpoints, auths, aors, and contacts. It should NOT be used for identifies, registrations and transports since these MUST be retrieved in bulk. Example sorcery.conf: [res_pjsip] endpoint=config,pjsip.conf,criteria=type=endpoint endpoint=realtime,ps_endpoints,allow_unqualified_fetch=error ASTERISK-25826 #close Reported-by: Ross Beer Tested-by: Ross Beer Change-Id: Id2691e447db90892890036e663aaf907b2dc1c67
2016-03-08 21:55:30 +00:00
struct sorcery_config *config = data;
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
RAII_VAR(struct ast_variable *, objectset, NULL, ast_variables_destroy);
RAII_VAR(struct ast_variable *, id, NULL, ast_variables_destroy);
void *object = NULL;
sorcery/res_pjsip: Refactor for realtime performance There were a number of places in the res_pjsip stack that were getting all endpoints or all aors, and then filtering them locally. A good example is pjsip_options which, on startup, retrieves all endpoints, then the aors for those endpoints, then tests the aors to see if the qualify_frequency is > 0. One issue was that it never did anything with the endpoints other than retrieve the aors so we probably could have skipped a step and just retrieved all aors. But nevermind. This worked reasonably well with local config files but with a realtime backend and thousands of objects, this was a nightmare. The issue really boiled down to the fact that while realtime supports predicates that are passed to the database engine, the non-realtime sorcery backends didn't. They do now. The realtime engines have a scheme for doing simple comparisons. They take in an ast_variable (or list) for matching, and the name of each variable can contain an operator. For instance, a name of "qualify_frequency >" and a value of "0" would create a SQL predicate that looks like "where qualify_frequency > '0'". If there's no operator after the name, the engines add an '=' so a simple name of "qualify_frequency" and a value of "10" would return exact matches. The non-realtime backends decide whether to include an object in a result set by calling ast_sorcery_changeset_create on every object in the internal container. However, ast_sorcery_changeset_create only does exact string matches though so a name of "qualify_frequency >" and a value of "0" returns nothing because the literal "qualify_frequency >" doesn't match any name in the objset set. So, the real task was to create a generic string matcher that can take a left value, operator and a right value and perform the match. To that end, strings.c has a new ast_strings_match(left, operator, right) function. Left and right are the strings to operate on and the operator can be a string containing any of the following: = (or NULL or ""), !=, >, >=, <, <=, like or regex. If the operator is like or regex, the right string should be a %-pattern or a regex expression. If both left and right can be converted to float, then a numeric comparison is performed, otherwise a string comparison is performed. To use this new function on ast_variables, 2 new functions were added to config.c. One that compares 2 ast_variables, and one that compares 2 ast_variable lists. The former is useful when you want to compare 2 ast_variables that happen to be in a list but don't want to traverse the list. The latter will traverse the right list and return true if all the variables in it match the left list. Now, the backends' fields_cmp functions call ast_variable_lists_match instead of ast_sorcery_changeset_create and they can now process the same syntax as the realtime engines. The realtime backend just passes the variable list unaltered to the engine. The only gotcha is that there's no common realtime engine support for regex so that's been noted in the api docs for ast_sorcery_retrieve_by_fields. Only one more change to sorcery was done... A new config flag "allow_unqualified_fetch" was added to reg_sorcery_realtime. "no": ignore fetches if no predicate fields were supplied. "error": same as no but emit an error. (good for testing) "yes": allow (the default); "warn": allow but emit a warning. (good for testing) Now on to res_pjsip... pjsip_options was modified to retrieve aors with qualify_frequency > 0 rather than all endpoints then all aors. Not only was this a big improvement in realtime retrieval but even for config files there's an improvement because we're not going through endpoints anymore. res_pjsip_mwi was modified to retieve only endpoints with something in the mailboxes field instead of all endpoints then testing mailboxes. res_pjsip_registrar_expire was completely refactored. It was retrieving all contacts then setting up scheduler entries to check for expiration. Now, it's a single thread (like keepalive) that periodically retrieves only contacts whose expiration time is < now and deletes them. A new contact_expiration_check_interval was added to global with a default of 30 seconds. Ross Beer reports that with this patch, his Asterisk startup time dropped from around an hour to under 30 seconds. There are still objects that can't be filtered at the database like identifies, transports, and registrations. These are not going to be anywhere near as numerous as endpoints, aors, auths, contacts however. Back to allow_unqualified_fetch. If this is set to yes and you have a very large number of objects in the database, the pjsip CLI commands will attempt to retrive ALL of them if not qualified with a LIKE. Worse, if you type "pjsip show endpoint <tab>" guess what's going to happen? :) Having a cache helps but all the objects will have to be retrieved at least once to fill the cache. Setting allow_unqualified_fetch=no prevents the mass retrieve and should be used on endpoints, auths, aors, and contacts. It should NOT be used for identifies, registrations and transports since these MUST be retrieved in bulk. Example sorcery.conf: [res_pjsip] endpoint=config,pjsip.conf,criteria=type=endpoint endpoint=realtime,ps_endpoints,allow_unqualified_fetch=error ASTERISK-25826 #close Reported-by: Ross Beer Tested-by: Ross Beer Change-Id: Id2691e447db90892890036e663aaf907b2dc1c67
2016-03-08 21:55:30 +00:00
if (!(objectset = ast_load_realtime_fields(config->family, fields))) {
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
return NULL;
}
Make res_sorcery_realtime filter unknown retrieved results. When retrieving data from a database or other realtime backend, it's quite possible to retrieve variables that Asterisk does not care about but that are legitimate to exist. Asterisk does not need to throw a hissy fit when these variables are encountered but rather just filter them out. Review: https://reviewboard.asterisk.org/r/3305 ........ Merged revisions 410187 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@410207 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-03-07 21:23:39 +00:00
objectset = sorcery_realtime_filter_objectset(objectset, &id, sorcery, type);
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
A couple minor cleanup tweaks. * In res/res_sorcery_realtime.c: Broke long line. * In main/bucket.c: Eliminated unnecessary NULL check as ast_sorcery_unref() is NULL tolerant and set the global object to NULL after unref in the system shutdown bucket_cleanup(). ........ Merged revisions 433420 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@433421 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-03-25 18:37:51 +00:00
if (!id
|| !(object = ast_sorcery_alloc(sorcery, type, id->value))
|| ast_sorcery_objectset_apply(sorcery, object, objectset)) {
res_sorcery_realtime.c: Fix ref leak if object failed to apply. Change-Id: I3c7106ff77009754725cee790eadf5da44154ab6
2018-02-02 21:43:34 +00:00
ao2_cleanup(object);
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
return NULL;
}
return object;
}
static void *sorcery_realtime_retrieve_id(const struct ast_sorcery *sorcery, void *data, const char *type, const char *id)
{
RAII_VAR(struct ast_variable *, fields, ast_variable_new(UUID_FIELD, id, ""), ast_variables_destroy);
return sorcery_realtime_retrieve_fields(sorcery, data, type, fields);
}
static void sorcery_realtime_retrieve_multiple(const struct ast_sorcery *sorcery, void *data, const char *type, struct ao2_container *objects, const struct ast_variable *fields)
{
sorcery/res_pjsip: Refactor for realtime performance There were a number of places in the res_pjsip stack that were getting all endpoints or all aors, and then filtering them locally. A good example is pjsip_options which, on startup, retrieves all endpoints, then the aors for those endpoints, then tests the aors to see if the qualify_frequency is > 0. One issue was that it never did anything with the endpoints other than retrieve the aors so we probably could have skipped a step and just retrieved all aors. But nevermind. This worked reasonably well with local config files but with a realtime backend and thousands of objects, this was a nightmare. The issue really boiled down to the fact that while realtime supports predicates that are passed to the database engine, the non-realtime sorcery backends didn't. They do now. The realtime engines have a scheme for doing simple comparisons. They take in an ast_variable (or list) for matching, and the name of each variable can contain an operator. For instance, a name of "qualify_frequency >" and a value of "0" would create a SQL predicate that looks like "where qualify_frequency > '0'". If there's no operator after the name, the engines add an '=' so a simple name of "qualify_frequency" and a value of "10" would return exact matches. The non-realtime backends decide whether to include an object in a result set by calling ast_sorcery_changeset_create on every object in the internal container. However, ast_sorcery_changeset_create only does exact string matches though so a name of "qualify_frequency >" and a value of "0" returns nothing because the literal "qualify_frequency >" doesn't match any name in the objset set. So, the real task was to create a generic string matcher that can take a left value, operator and a right value and perform the match. To that end, strings.c has a new ast_strings_match(left, operator, right) function. Left and right are the strings to operate on and the operator can be a string containing any of the following: = (or NULL or ""), !=, >, >=, <, <=, like or regex. If the operator is like or regex, the right string should be a %-pattern or a regex expression. If both left and right can be converted to float, then a numeric comparison is performed, otherwise a string comparison is performed. To use this new function on ast_variables, 2 new functions were added to config.c. One that compares 2 ast_variables, and one that compares 2 ast_variable lists. The former is useful when you want to compare 2 ast_variables that happen to be in a list but don't want to traverse the list. The latter will traverse the right list and return true if all the variables in it match the left list. Now, the backends' fields_cmp functions call ast_variable_lists_match instead of ast_sorcery_changeset_create and they can now process the same syntax as the realtime engines. The realtime backend just passes the variable list unaltered to the engine. The only gotcha is that there's no common realtime engine support for regex so that's been noted in the api docs for ast_sorcery_retrieve_by_fields. Only one more change to sorcery was done... A new config flag "allow_unqualified_fetch" was added to reg_sorcery_realtime. "no": ignore fetches if no predicate fields were supplied. "error": same as no but emit an error. (good for testing) "yes": allow (the default); "warn": allow but emit a warning. (good for testing) Now on to res_pjsip... pjsip_options was modified to retrieve aors with qualify_frequency > 0 rather than all endpoints then all aors. Not only was this a big improvement in realtime retrieval but even for config files there's an improvement because we're not going through endpoints anymore. res_pjsip_mwi was modified to retieve only endpoints with something in the mailboxes field instead of all endpoints then testing mailboxes. res_pjsip_registrar_expire was completely refactored. It was retrieving all contacts then setting up scheduler entries to check for expiration. Now, it's a single thread (like keepalive) that periodically retrieves only contacts whose expiration time is < now and deletes them. A new contact_expiration_check_interval was added to global with a default of 30 seconds. Ross Beer reports that with this patch, his Asterisk startup time dropped from around an hour to under 30 seconds. There are still objects that can't be filtered at the database like identifies, transports, and registrations. These are not going to be anywhere near as numerous as endpoints, aors, auths, contacts however. Back to allow_unqualified_fetch. If this is set to yes and you have a very large number of objects in the database, the pjsip CLI commands will attempt to retrive ALL of them if not qualified with a LIKE. Worse, if you type "pjsip show endpoint <tab>" guess what's going to happen? :) Having a cache helps but all the objects will have to be retrieved at least once to fill the cache. Setting allow_unqualified_fetch=no prevents the mass retrieve and should be used on endpoints, auths, aors, and contacts. It should NOT be used for identifies, registrations and transports since these MUST be retrieved in bulk. Example sorcery.conf: [res_pjsip] endpoint=config,pjsip.conf,criteria=type=endpoint endpoint=realtime,ps_endpoints,allow_unqualified_fetch=error ASTERISK-25826 #close Reported-by: Ross Beer Tested-by: Ross Beer Change-Id: Id2691e447db90892890036e663aaf907b2dc1c67
2016-03-08 21:55:30 +00:00
struct sorcery_config *config = data;
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
RAII_VAR(struct ast_config *, rows, NULL, ast_config_destroy);
RAII_VAR(struct ast_variable *, all, NULL, ast_variables_destroy);
manager/config: Support templates and non-unique category names via AMI This patch provides the capability to manipulate templates and categories with non-unique names via AMI. Summary of changes: GetConfig and GetConfigJSON: Added "Filter" parameter: A comma separated list of name_regex=value_regex expressions which will cause only categories whose variables match all expressions to be considered. The special variable name TEMPLATES can be used to control whether templates are included. Passing 'include' as the value will include templates along with normal categories. Passing 'restrict' as the value will restrict the operation to ONLY templates. Not specifying a TEMPLATES expression results in the current default behavior which is to not include templates. UpdateConfig: NewCat now includes options for allowing duplicate category names, indicating if the category should be created as a template, and specifying templates the category should inherit from. The rest of the actions now accept a filter string as defined above. If there are non-unique category names, you can now update specific ones based on variable values. To facilitate the new capabilities in manager, corresponding changes had to be made to config, most notably the addition of filter criteria to many of the APIs. In some cases it was easy to change the references to use the new prototype but others would have required touching too many files for this patch so a wrapper with the original prototype was created. Macros couldn't be used in this case because it would break binary compatibility with modules such as res_digium_phone that are linked to real symbols. Tested-by: George Joseph Review: https://reviewboard.asterisk.org/r/4033/ ........ Merged revisions 425383 from http://svn.asterisk.org/svn/asterisk/branches/12 ........ Merged revisions 425384 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@425385 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-10-13 16:12:17 +00:00
struct ast_category *row = NULL;
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
if (!fields) {
char field[strlen(UUID_FIELD) + 6], value[2];
sorcery/res_pjsip: Refactor for realtime performance There were a number of places in the res_pjsip stack that were getting all endpoints or all aors, and then filtering them locally. A good example is pjsip_options which, on startup, retrieves all endpoints, then the aors for those endpoints, then tests the aors to see if the qualify_frequency is > 0. One issue was that it never did anything with the endpoints other than retrieve the aors so we probably could have skipped a step and just retrieved all aors. But nevermind. This worked reasonably well with local config files but with a realtime backend and thousands of objects, this was a nightmare. The issue really boiled down to the fact that while realtime supports predicates that are passed to the database engine, the non-realtime sorcery backends didn't. They do now. The realtime engines have a scheme for doing simple comparisons. They take in an ast_variable (or list) for matching, and the name of each variable can contain an operator. For instance, a name of "qualify_frequency >" and a value of "0" would create a SQL predicate that looks like "where qualify_frequency > '0'". If there's no operator after the name, the engines add an '=' so a simple name of "qualify_frequency" and a value of "10" would return exact matches. The non-realtime backends decide whether to include an object in a result set by calling ast_sorcery_changeset_create on every object in the internal container. However, ast_sorcery_changeset_create only does exact string matches though so a name of "qualify_frequency >" and a value of "0" returns nothing because the literal "qualify_frequency >" doesn't match any name in the objset set. So, the real task was to create a generic string matcher that can take a left value, operator and a right value and perform the match. To that end, strings.c has a new ast_strings_match(left, operator, right) function. Left and right are the strings to operate on and the operator can be a string containing any of the following: = (or NULL or ""), !=, >, >=, <, <=, like or regex. If the operator is like or regex, the right string should be a %-pattern or a regex expression. If both left and right can be converted to float, then a numeric comparison is performed, otherwise a string comparison is performed. To use this new function on ast_variables, 2 new functions were added to config.c. One that compares 2 ast_variables, and one that compares 2 ast_variable lists. The former is useful when you want to compare 2 ast_variables that happen to be in a list but don't want to traverse the list. The latter will traverse the right list and return true if all the variables in it match the left list. Now, the backends' fields_cmp functions call ast_variable_lists_match instead of ast_sorcery_changeset_create and they can now process the same syntax as the realtime engines. The realtime backend just passes the variable list unaltered to the engine. The only gotcha is that there's no common realtime engine support for regex so that's been noted in the api docs for ast_sorcery_retrieve_by_fields. Only one more change to sorcery was done... A new config flag "allow_unqualified_fetch" was added to reg_sorcery_realtime. "no": ignore fetches if no predicate fields were supplied. "error": same as no but emit an error. (good for testing) "yes": allow (the default); "warn": allow but emit a warning. (good for testing) Now on to res_pjsip... pjsip_options was modified to retrieve aors with qualify_frequency > 0 rather than all endpoints then all aors. Not only was this a big improvement in realtime retrieval but even for config files there's an improvement because we're not going through endpoints anymore. res_pjsip_mwi was modified to retieve only endpoints with something in the mailboxes field instead of all endpoints then testing mailboxes. res_pjsip_registrar_expire was completely refactored. It was retrieving all contacts then setting up scheduler entries to check for expiration. Now, it's a single thread (like keepalive) that periodically retrieves only contacts whose expiration time is < now and deletes them. A new contact_expiration_check_interval was added to global with a default of 30 seconds. Ross Beer reports that with this patch, his Asterisk startup time dropped from around an hour to under 30 seconds. There are still objects that can't be filtered at the database like identifies, transports, and registrations. These are not going to be anywhere near as numerous as endpoints, aors, auths, contacts however. Back to allow_unqualified_fetch. If this is set to yes and you have a very large number of objects in the database, the pjsip CLI commands will attempt to retrive ALL of them if not qualified with a LIKE. Worse, if you type "pjsip show endpoint <tab>" guess what's going to happen? :) Having a cache helps but all the objects will have to be retrieved at least once to fill the cache. Setting allow_unqualified_fetch=no prevents the mass retrieve and should be used on endpoints, auths, aors, and contacts. It should NOT be used for identifies, registrations and transports since these MUST be retrieved in bulk. Example sorcery.conf: [res_pjsip] endpoint=config,pjsip.conf,criteria=type=endpoint endpoint=realtime,ps_endpoints,allow_unqualified_fetch=error ASTERISK-25826 #close Reported-by: Ross Beer Tested-by: Ross Beer Change-Id: Id2691e447db90892890036e663aaf907b2dc1c67
2016-03-08 21:55:30 +00:00
if (config->fetch == UNQUALIFIED_FETCH_NO) {
return;
}
if (config->fetch == UNQUALIFIED_FETCH_ERROR) {
ast_log(LOG_ERROR, "Unqualified fetch prevented on %s\n", config->family);
return;
}
if (config->fetch == UNQUALIFIED_FETCH_WARN) {
res_sorcery_realtime.c: Fix unqualified fetch warning. The allow_unqualified_fetch option for the sorcery realtime backend blocked actually fetching all rows when the option is set to warn. * Made issue a warning and actually do the request when allow_unqualified_fetch=warn is set. Change-Id: I74456c80a03a62dce66fc3dc3cb0cf2351ac4312
2018-08-09 23:46:19 +00:00
ast_log(LOG_WARNING, "Unqualified fetch requested on %s\n", config->family);
sorcery/res_pjsip: Refactor for realtime performance There were a number of places in the res_pjsip stack that were getting all endpoints or all aors, and then filtering them locally. A good example is pjsip_options which, on startup, retrieves all endpoints, then the aors for those endpoints, then tests the aors to see if the qualify_frequency is > 0. One issue was that it never did anything with the endpoints other than retrieve the aors so we probably could have skipped a step and just retrieved all aors. But nevermind. This worked reasonably well with local config files but with a realtime backend and thousands of objects, this was a nightmare. The issue really boiled down to the fact that while realtime supports predicates that are passed to the database engine, the non-realtime sorcery backends didn't. They do now. The realtime engines have a scheme for doing simple comparisons. They take in an ast_variable (or list) for matching, and the name of each variable can contain an operator. For instance, a name of "qualify_frequency >" and a value of "0" would create a SQL predicate that looks like "where qualify_frequency > '0'". If there's no operator after the name, the engines add an '=' so a simple name of "qualify_frequency" and a value of "10" would return exact matches. The non-realtime backends decide whether to include an object in a result set by calling ast_sorcery_changeset_create on every object in the internal container. However, ast_sorcery_changeset_create only does exact string matches though so a name of "qualify_frequency >" and a value of "0" returns nothing because the literal "qualify_frequency >" doesn't match any name in the objset set. So, the real task was to create a generic string matcher that can take a left value, operator and a right value and perform the match. To that end, strings.c has a new ast_strings_match(left, operator, right) function. Left and right are the strings to operate on and the operator can be a string containing any of the following: = (or NULL or ""), !=, >, >=, <, <=, like or regex. If the operator is like or regex, the right string should be a %-pattern or a regex expression. If both left and right can be converted to float, then a numeric comparison is performed, otherwise a string comparison is performed. To use this new function on ast_variables, 2 new functions were added to config.c. One that compares 2 ast_variables, and one that compares 2 ast_variable lists. The former is useful when you want to compare 2 ast_variables that happen to be in a list but don't want to traverse the list. The latter will traverse the right list and return true if all the variables in it match the left list. Now, the backends' fields_cmp functions call ast_variable_lists_match instead of ast_sorcery_changeset_create and they can now process the same syntax as the realtime engines. The realtime backend just passes the variable list unaltered to the engine. The only gotcha is that there's no common realtime engine support for regex so that's been noted in the api docs for ast_sorcery_retrieve_by_fields. Only one more change to sorcery was done... A new config flag "allow_unqualified_fetch" was added to reg_sorcery_realtime. "no": ignore fetches if no predicate fields were supplied. "error": same as no but emit an error. (good for testing) "yes": allow (the default); "warn": allow but emit a warning. (good for testing) Now on to res_pjsip... pjsip_options was modified to retrieve aors with qualify_frequency > 0 rather than all endpoints then all aors. Not only was this a big improvement in realtime retrieval but even for config files there's an improvement because we're not going through endpoints anymore. res_pjsip_mwi was modified to retieve only endpoints with something in the mailboxes field instead of all endpoints then testing mailboxes. res_pjsip_registrar_expire was completely refactored. It was retrieving all contacts then setting up scheduler entries to check for expiration. Now, it's a single thread (like keepalive) that periodically retrieves only contacts whose expiration time is < now and deletes them. A new contact_expiration_check_interval was added to global with a default of 30 seconds. Ross Beer reports that with this patch, his Asterisk startup time dropped from around an hour to under 30 seconds. There are still objects that can't be filtered at the database like identifies, transports, and registrations. These are not going to be anywhere near as numerous as endpoints, aors, auths, contacts however. Back to allow_unqualified_fetch. If this is set to yes and you have a very large number of objects in the database, the pjsip CLI commands will attempt to retrive ALL of them if not qualified with a LIKE. Worse, if you type "pjsip show endpoint <tab>" guess what's going to happen? :) Having a cache helps but all the objects will have to be retrieved at least once to fill the cache. Setting allow_unqualified_fetch=no prevents the mass retrieve and should be used on endpoints, auths, aors, and contacts. It should NOT be used for identifies, registrations and transports since these MUST be retrieved in bulk. Example sorcery.conf: [res_pjsip] endpoint=config,pjsip.conf,criteria=type=endpoint endpoint=realtime,ps_endpoints,allow_unqualified_fetch=error ASTERISK-25826 #close Reported-by: Ross Beer Tested-by: Ross Beer Change-Id: Id2691e447db90892890036e663aaf907b2dc1c67
2016-03-08 21:55:30 +00:00
}
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
/* If no fields have been specified we want all rows, so trick realtime into doing it */
snprintf(field, sizeof(field), "%s LIKE", UUID_FIELD);
snprintf(value, sizeof(value), "%%");
if (!(all = ast_variable_new(field, value, ""))) {
return;
}
fields = all;
}
sorcery/res_pjsip: Refactor for realtime performance There were a number of places in the res_pjsip stack that were getting all endpoints or all aors, and then filtering them locally. A good example is pjsip_options which, on startup, retrieves all endpoints, then the aors for those endpoints, then tests the aors to see if the qualify_frequency is > 0. One issue was that it never did anything with the endpoints other than retrieve the aors so we probably could have skipped a step and just retrieved all aors. But nevermind. This worked reasonably well with local config files but with a realtime backend and thousands of objects, this was a nightmare. The issue really boiled down to the fact that while realtime supports predicates that are passed to the database engine, the non-realtime sorcery backends didn't. They do now. The realtime engines have a scheme for doing simple comparisons. They take in an ast_variable (or list) for matching, and the name of each variable can contain an operator. For instance, a name of "qualify_frequency >" and a value of "0" would create a SQL predicate that looks like "where qualify_frequency > '0'". If there's no operator after the name, the engines add an '=' so a simple name of "qualify_frequency" and a value of "10" would return exact matches. The non-realtime backends decide whether to include an object in a result set by calling ast_sorcery_changeset_create on every object in the internal container. However, ast_sorcery_changeset_create only does exact string matches though so a name of "qualify_frequency >" and a value of "0" returns nothing because the literal "qualify_frequency >" doesn't match any name in the objset set. So, the real task was to create a generic string matcher that can take a left value, operator and a right value and perform the match. To that end, strings.c has a new ast_strings_match(left, operator, right) function. Left and right are the strings to operate on and the operator can be a string containing any of the following: = (or NULL or ""), !=, >, >=, <, <=, like or regex. If the operator is like or regex, the right string should be a %-pattern or a regex expression. If both left and right can be converted to float, then a numeric comparison is performed, otherwise a string comparison is performed. To use this new function on ast_variables, 2 new functions were added to config.c. One that compares 2 ast_variables, and one that compares 2 ast_variable lists. The former is useful when you want to compare 2 ast_variables that happen to be in a list but don't want to traverse the list. The latter will traverse the right list and return true if all the variables in it match the left list. Now, the backends' fields_cmp functions call ast_variable_lists_match instead of ast_sorcery_changeset_create and they can now process the same syntax as the realtime engines. The realtime backend just passes the variable list unaltered to the engine. The only gotcha is that there's no common realtime engine support for regex so that's been noted in the api docs for ast_sorcery_retrieve_by_fields. Only one more change to sorcery was done... A new config flag "allow_unqualified_fetch" was added to reg_sorcery_realtime. "no": ignore fetches if no predicate fields were supplied. "error": same as no but emit an error. (good for testing) "yes": allow (the default); "warn": allow but emit a warning. (good for testing) Now on to res_pjsip... pjsip_options was modified to retrieve aors with qualify_frequency > 0 rather than all endpoints then all aors. Not only was this a big improvement in realtime retrieval but even for config files there's an improvement because we're not going through endpoints anymore. res_pjsip_mwi was modified to retieve only endpoints with something in the mailboxes field instead of all endpoints then testing mailboxes. res_pjsip_registrar_expire was completely refactored. It was retrieving all contacts then setting up scheduler entries to check for expiration. Now, it's a single thread (like keepalive) that periodically retrieves only contacts whose expiration time is < now and deletes them. A new contact_expiration_check_interval was added to global with a default of 30 seconds. Ross Beer reports that with this patch, his Asterisk startup time dropped from around an hour to under 30 seconds. There are still objects that can't be filtered at the database like identifies, transports, and registrations. These are not going to be anywhere near as numerous as endpoints, aors, auths, contacts however. Back to allow_unqualified_fetch. If this is set to yes and you have a very large number of objects in the database, the pjsip CLI commands will attempt to retrive ALL of them if not qualified with a LIKE. Worse, if you type "pjsip show endpoint <tab>" guess what's going to happen? :) Having a cache helps but all the objects will have to be retrieved at least once to fill the cache. Setting allow_unqualified_fetch=no prevents the mass retrieve and should be used on endpoints, auths, aors, and contacts. It should NOT be used for identifies, registrations and transports since these MUST be retrieved in bulk. Example sorcery.conf: [res_pjsip] endpoint=config,pjsip.conf,criteria=type=endpoint endpoint=realtime,ps_endpoints,allow_unqualified_fetch=error ASTERISK-25826 #close Reported-by: Ross Beer Tested-by: Ross Beer Change-Id: Id2691e447db90892890036e663aaf907b2dc1c67
2016-03-08 21:55:30 +00:00
if (!(rows = ast_load_realtime_multientry_fields(config->family, fields))) {
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
return;
}
manager/config: Support templates and non-unique category names via AMI This patch provides the capability to manipulate templates and categories with non-unique names via AMI. Summary of changes: GetConfig and GetConfigJSON: Added "Filter" parameter: A comma separated list of name_regex=value_regex expressions which will cause only categories whose variables match all expressions to be considered. The special variable name TEMPLATES can be used to control whether templates are included. Passing 'include' as the value will include templates along with normal categories. Passing 'restrict' as the value will restrict the operation to ONLY templates. Not specifying a TEMPLATES expression results in the current default behavior which is to not include templates. UpdateConfig: NewCat now includes options for allowing duplicate category names, indicating if the category should be created as a template, and specifying templates the category should inherit from. The rest of the actions now accept a filter string as defined above. If there are non-unique category names, you can now update specific ones based on variable values. To facilitate the new capabilities in manager, corresponding changes had to be made to config, most notably the addition of filter criteria to many of the APIs. In some cases it was easy to change the references to use the new prototype but others would have required touching too many files for this patch so a wrapper with the original prototype was created. Macros couldn't be used in this case because it would break binary compatibility with modules such as res_digium_phone that are linked to real symbols. Tested-by: George Joseph Review: https://reviewboard.asterisk.org/r/4033/ ........ Merged revisions 425383 from http://svn.asterisk.org/svn/asterisk/branches/12 ........ Merged revisions 425384 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@425385 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-10-13 16:12:17 +00:00
while ((row = ast_category_browse_filtered(rows, NULL, row, NULL))) {
struct ast_variable *objectset = ast_category_detach_variables(row);
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
RAII_VAR(struct ast_variable *, id, NULL, ast_variables_destroy);
RAII_VAR(void *, object, NULL, ao2_cleanup);
Make res_sorcery_realtime filter unknown retrieved results. When retrieving data from a database or other realtime backend, it's quite possible to retrieve variables that Asterisk does not care about but that are legitimate to exist. Asterisk does not need to throw a hissy fit when these variables are encountered but rather just filter them out. Review: https://reviewboard.asterisk.org/r/3305 ........ Merged revisions 410187 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@410207 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-03-07 21:23:39 +00:00
objectset = sorcery_realtime_filter_objectset(objectset, &id, sorcery, type);
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
res_sorcery_realtime.c: Fix ref leak if object failed to apply. Change-Id: I3c7106ff77009754725cee790eadf5da44154ab6
2018-02-02 21:43:34 +00:00
if (id
&& (object = ast_sorcery_alloc(sorcery, type, id->value))
&& !ast_sorcery_objectset_apply(sorcery, object, objectset)) {
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
ao2_link(objects, object);
}
Make res_sorcery_realtime filter unknown retrieved results. When retrieving data from a database or other realtime backend, it's quite possible to retrieve variables that Asterisk does not care about but that are legitimate to exist. Asterisk does not need to throw a hissy fit when these variables are encountered but rather just filter them out. Review: https://reviewboard.asterisk.org/r/3305 ........ Merged revisions 410187 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@410207 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-03-07 21:23:39 +00:00
ast_variables_destroy(objectset);
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
}
}
static void sorcery_realtime_retrieve_regex(const struct ast_sorcery *sorcery, void *data, const char *type, struct ao2_container *objects, const char *regex)
{
res_sorcery_realtime: Remove leading ^ requirement. res_sorcery_realtime's search-by-regex callback performed a check to ensure that the passed-in regex began with a caret (^). If it did not, then no results would be returned. This callback only started to become used when "like" support was added to PJSIP CLI commands. The CLI command for listing objects would pass an empty regex ("") to the sorcery backend if no "like" statement was present. For most sorcery backends, this resulted in returning all objects. However, for realtime, this resulted in returning no objects. This commit seeks to fix the regression by removing the requirement from res_sorcery_realtime for the passed-in-regex to begin with a caret. ASTERISK-25689 #close Reported by Marcelo Terres Change-Id: I22b4dc5d7f3f11bb29ac2e42ef94682e9bab3b20
2016-01-12 16:36:15 +00:00
char field[strlen(UUID_FIELD) + 6], value[strlen(regex) + 3];
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
RAII_VAR(struct ast_variable *, fields, NULL, ast_variables_destroy);
sorcery/res_pjsip: Refactor for realtime performance There were a number of places in the res_pjsip stack that were getting all endpoints or all aors, and then filtering them locally. A good example is pjsip_options which, on startup, retrieves all endpoints, then the aors for those endpoints, then tests the aors to see if the qualify_frequency is > 0. One issue was that it never did anything with the endpoints other than retrieve the aors so we probably could have skipped a step and just retrieved all aors. But nevermind. This worked reasonably well with local config files but with a realtime backend and thousands of objects, this was a nightmare. The issue really boiled down to the fact that while realtime supports predicates that are passed to the database engine, the non-realtime sorcery backends didn't. They do now. The realtime engines have a scheme for doing simple comparisons. They take in an ast_variable (or list) for matching, and the name of each variable can contain an operator. For instance, a name of "qualify_frequency >" and a value of "0" would create a SQL predicate that looks like "where qualify_frequency > '0'". If there's no operator after the name, the engines add an '=' so a simple name of "qualify_frequency" and a value of "10" would return exact matches. The non-realtime backends decide whether to include an object in a result set by calling ast_sorcery_changeset_create on every object in the internal container. However, ast_sorcery_changeset_create only does exact string matches though so a name of "qualify_frequency >" and a value of "0" returns nothing because the literal "qualify_frequency >" doesn't match any name in the objset set. So, the real task was to create a generic string matcher that can take a left value, operator and a right value and perform the match. To that end, strings.c has a new ast_strings_match(left, operator, right) function. Left and right are the strings to operate on and the operator can be a string containing any of the following: = (or NULL or ""), !=, >, >=, <, <=, like or regex. If the operator is like or regex, the right string should be a %-pattern or a regex expression. If both left and right can be converted to float, then a numeric comparison is performed, otherwise a string comparison is performed. To use this new function on ast_variables, 2 new functions were added to config.c. One that compares 2 ast_variables, and one that compares 2 ast_variable lists. The former is useful when you want to compare 2 ast_variables that happen to be in a list but don't want to traverse the list. The latter will traverse the right list and return true if all the variables in it match the left list. Now, the backends' fields_cmp functions call ast_variable_lists_match instead of ast_sorcery_changeset_create and they can now process the same syntax as the realtime engines. The realtime backend just passes the variable list unaltered to the engine. The only gotcha is that there's no common realtime engine support for regex so that's been noted in the api docs for ast_sorcery_retrieve_by_fields. Only one more change to sorcery was done... A new config flag "allow_unqualified_fetch" was added to reg_sorcery_realtime. "no": ignore fetches if no predicate fields were supplied. "error": same as no but emit an error. (good for testing) "yes": allow (the default); "warn": allow but emit a warning. (good for testing) Now on to res_pjsip... pjsip_options was modified to retrieve aors with qualify_frequency > 0 rather than all endpoints then all aors. Not only was this a big improvement in realtime retrieval but even for config files there's an improvement because we're not going through endpoints anymore. res_pjsip_mwi was modified to retieve only endpoints with something in the mailboxes field instead of all endpoints then testing mailboxes. res_pjsip_registrar_expire was completely refactored. It was retrieving all contacts then setting up scheduler entries to check for expiration. Now, it's a single thread (like keepalive) that periodically retrieves only contacts whose expiration time is < now and deletes them. A new contact_expiration_check_interval was added to global with a default of 30 seconds. Ross Beer reports that with this patch, his Asterisk startup time dropped from around an hour to under 30 seconds. There are still objects that can't be filtered at the database like identifies, transports, and registrations. These are not going to be anywhere near as numerous as endpoints, aors, auths, contacts however. Back to allow_unqualified_fetch. If this is set to yes and you have a very large number of objects in the database, the pjsip CLI commands will attempt to retrive ALL of them if not qualified with a LIKE. Worse, if you type "pjsip show endpoint <tab>" guess what's going to happen? :) Having a cache helps but all the objects will have to be retrieved at least once to fill the cache. Setting allow_unqualified_fetch=no prevents the mass retrieve and should be used on endpoints, auths, aors, and contacts. It should NOT be used for identifies, registrations and transports since these MUST be retrieved in bulk. Example sorcery.conf: [res_pjsip] endpoint=config,pjsip.conf,criteria=type=endpoint endpoint=realtime,ps_endpoints,allow_unqualified_fetch=error ASTERISK-25826 #close Reported-by: Ross Beer Tested-by: Ross Beer Change-Id: Id2691e447db90892890036e663aaf907b2dc1c67
2016-03-08 21:55:30 +00:00
if (!ast_strlen_zero(regex)) {
/* The realtime API provides no direct ability to do regex so for now we support a limited subset using pattern matching */
snprintf(field, sizeof(field), "%s LIKE", UUID_FIELD);
if (regex[0] == '^') {
snprintf(value, sizeof(value), "%s%%", regex + 1);
} else {
snprintf(value, sizeof(value), "%%%s%%", regex);
}
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
sorcery/res_pjsip: Refactor for realtime performance There were a number of places in the res_pjsip stack that were getting all endpoints or all aors, and then filtering them locally. A good example is pjsip_options which, on startup, retrieves all endpoints, then the aors for those endpoints, then tests the aors to see if the qualify_frequency is > 0. One issue was that it never did anything with the endpoints other than retrieve the aors so we probably could have skipped a step and just retrieved all aors. But nevermind. This worked reasonably well with local config files but with a realtime backend and thousands of objects, this was a nightmare. The issue really boiled down to the fact that while realtime supports predicates that are passed to the database engine, the non-realtime sorcery backends didn't. They do now. The realtime engines have a scheme for doing simple comparisons. They take in an ast_variable (or list) for matching, and the name of each variable can contain an operator. For instance, a name of "qualify_frequency >" and a value of "0" would create a SQL predicate that looks like "where qualify_frequency > '0'". If there's no operator after the name, the engines add an '=' so a simple name of "qualify_frequency" and a value of "10" would return exact matches. The non-realtime backends decide whether to include an object in a result set by calling ast_sorcery_changeset_create on every object in the internal container. However, ast_sorcery_changeset_create only does exact string matches though so a name of "qualify_frequency >" and a value of "0" returns nothing because the literal "qualify_frequency >" doesn't match any name in the objset set. So, the real task was to create a generic string matcher that can take a left value, operator and a right value and perform the match. To that end, strings.c has a new ast_strings_match(left, operator, right) function. Left and right are the strings to operate on and the operator can be a string containing any of the following: = (or NULL or ""), !=, >, >=, <, <=, like or regex. If the operator is like or regex, the right string should be a %-pattern or a regex expression. If both left and right can be converted to float, then a numeric comparison is performed, otherwise a string comparison is performed. To use this new function on ast_variables, 2 new functions were added to config.c. One that compares 2 ast_variables, and one that compares 2 ast_variable lists. The former is useful when you want to compare 2 ast_variables that happen to be in a list but don't want to traverse the list. The latter will traverse the right list and return true if all the variables in it match the left list. Now, the backends' fields_cmp functions call ast_variable_lists_match instead of ast_sorcery_changeset_create and they can now process the same syntax as the realtime engines. The realtime backend just passes the variable list unaltered to the engine. The only gotcha is that there's no common realtime engine support for regex so that's been noted in the api docs for ast_sorcery_retrieve_by_fields. Only one more change to sorcery was done... A new config flag "allow_unqualified_fetch" was added to reg_sorcery_realtime. "no": ignore fetches if no predicate fields were supplied. "error": same as no but emit an error. (good for testing) "yes": allow (the default); "warn": allow but emit a warning. (good for testing) Now on to res_pjsip... pjsip_options was modified to retrieve aors with qualify_frequency > 0 rather than all endpoints then all aors. Not only was this a big improvement in realtime retrieval but even for config files there's an improvement because we're not going through endpoints anymore. res_pjsip_mwi was modified to retieve only endpoints with something in the mailboxes field instead of all endpoints then testing mailboxes. res_pjsip_registrar_expire was completely refactored. It was retrieving all contacts then setting up scheduler entries to check for expiration. Now, it's a single thread (like keepalive) that periodically retrieves only contacts whose expiration time is < now and deletes them. A new contact_expiration_check_interval was added to global with a default of 30 seconds. Ross Beer reports that with this patch, his Asterisk startup time dropped from around an hour to under 30 seconds. There are still objects that can't be filtered at the database like identifies, transports, and registrations. These are not going to be anywhere near as numerous as endpoints, aors, auths, contacts however. Back to allow_unqualified_fetch. If this is set to yes and you have a very large number of objects in the database, the pjsip CLI commands will attempt to retrive ALL of them if not qualified with a LIKE. Worse, if you type "pjsip show endpoint <tab>" guess what's going to happen? :) Having a cache helps but all the objects will have to be retrieved at least once to fill the cache. Setting allow_unqualified_fetch=no prevents the mass retrieve and should be used on endpoints, auths, aors, and contacts. It should NOT be used for identifies, registrations and transports since these MUST be retrieved in bulk. Example sorcery.conf: [res_pjsip] endpoint=config,pjsip.conf,criteria=type=endpoint endpoint=realtime,ps_endpoints,allow_unqualified_fetch=error ASTERISK-25826 #close Reported-by: Ross Beer Tested-by: Ross Beer Change-Id: Id2691e447db90892890036e663aaf907b2dc1c67
2016-03-08 21:55:30 +00:00
if (!(fields = ast_variable_new(field, value, ""))) {
return;
}
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
}
sorcery_realtime_retrieve_multiple(sorcery, data, type, objects, fields);
}
sorcery: Add ast_sorcery_retrieve_by_prefix() Some consumers of the sorcery API use ast_sorcery_retrieve_by_regex only so that they can anchor the potential match as a prefix and not because they truly need regular expressions. Rather than using regular expressions for simple prefix lookups, add a new operation - ast_sorcery_retrieve_by_prefix - that does them. Change-Id: I56f4e20ba1154bd52281f995c27a429a854f6a79
2017-11-09 14:21:38 +00:00
static void sorcery_realtime_retrieve_prefix(const struct ast_sorcery *sorcery, void *data, const char *type,
struct ao2_container *objects, const char *prefix, const size_t prefix_len)
{
char field[strlen(UUID_FIELD) + 6], value[prefix_len + 2];
RAII_VAR(struct ast_variable *, fields, NULL, ast_variables_destroy);
if (prefix_len) {
snprintf(field, sizeof(field), "%s LIKE", UUID_FIELD);
snprintf(value, sizeof(value), "%.*s%%", (int) prefix_len, prefix);
if (!(fields = ast_variable_new(field, value, ""))) {
return;
}
}
sorcery_realtime_retrieve_multiple(sorcery, data, type, objects, fields);
}
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
static int sorcery_realtime_update(const struct ast_sorcery *sorcery, void *data, void *object)
{
sorcery/res_pjsip: Refactor for realtime performance There were a number of places in the res_pjsip stack that were getting all endpoints or all aors, and then filtering them locally. A good example is pjsip_options which, on startup, retrieves all endpoints, then the aors for those endpoints, then tests the aors to see if the qualify_frequency is > 0. One issue was that it never did anything with the endpoints other than retrieve the aors so we probably could have skipped a step and just retrieved all aors. But nevermind. This worked reasonably well with local config files but with a realtime backend and thousands of objects, this was a nightmare. The issue really boiled down to the fact that while realtime supports predicates that are passed to the database engine, the non-realtime sorcery backends didn't. They do now. The realtime engines have a scheme for doing simple comparisons. They take in an ast_variable (or list) for matching, and the name of each variable can contain an operator. For instance, a name of "qualify_frequency >" and a value of "0" would create a SQL predicate that looks like "where qualify_frequency > '0'". If there's no operator after the name, the engines add an '=' so a simple name of "qualify_frequency" and a value of "10" would return exact matches. The non-realtime backends decide whether to include an object in a result set by calling ast_sorcery_changeset_create on every object in the internal container. However, ast_sorcery_changeset_create only does exact string matches though so a name of "qualify_frequency >" and a value of "0" returns nothing because the literal "qualify_frequency >" doesn't match any name in the objset set. So, the real task was to create a generic string matcher that can take a left value, operator and a right value and perform the match. To that end, strings.c has a new ast_strings_match(left, operator, right) function. Left and right are the strings to operate on and the operator can be a string containing any of the following: = (or NULL or ""), !=, >, >=, <, <=, like or regex. If the operator is like or regex, the right string should be a %-pattern or a regex expression. If both left and right can be converted to float, then a numeric comparison is performed, otherwise a string comparison is performed. To use this new function on ast_variables, 2 new functions were added to config.c. One that compares 2 ast_variables, and one that compares 2 ast_variable lists. The former is useful when you want to compare 2 ast_variables that happen to be in a list but don't want to traverse the list. The latter will traverse the right list and return true if all the variables in it match the left list. Now, the backends' fields_cmp functions call ast_variable_lists_match instead of ast_sorcery_changeset_create and they can now process the same syntax as the realtime engines. The realtime backend just passes the variable list unaltered to the engine. The only gotcha is that there's no common realtime engine support for regex so that's been noted in the api docs for ast_sorcery_retrieve_by_fields. Only one more change to sorcery was done... A new config flag "allow_unqualified_fetch" was added to reg_sorcery_realtime. "no": ignore fetches if no predicate fields were supplied. "error": same as no but emit an error. (good for testing) "yes": allow (the default); "warn": allow but emit a warning. (good for testing) Now on to res_pjsip... pjsip_options was modified to retrieve aors with qualify_frequency > 0 rather than all endpoints then all aors. Not only was this a big improvement in realtime retrieval but even for config files there's an improvement because we're not going through endpoints anymore. res_pjsip_mwi was modified to retieve only endpoints with something in the mailboxes field instead of all endpoints then testing mailboxes. res_pjsip_registrar_expire was completely refactored. It was retrieving all contacts then setting up scheduler entries to check for expiration. Now, it's a single thread (like keepalive) that periodically retrieves only contacts whose expiration time is < now and deletes them. A new contact_expiration_check_interval was added to global with a default of 30 seconds. Ross Beer reports that with this patch, his Asterisk startup time dropped from around an hour to under 30 seconds. There are still objects that can't be filtered at the database like identifies, transports, and registrations. These are not going to be anywhere near as numerous as endpoints, aors, auths, contacts however. Back to allow_unqualified_fetch. If this is set to yes and you have a very large number of objects in the database, the pjsip CLI commands will attempt to retrive ALL of them if not qualified with a LIKE. Worse, if you type "pjsip show endpoint <tab>" guess what's going to happen? :) Having a cache helps but all the objects will have to be retrieved at least once to fill the cache. Setting allow_unqualified_fetch=no prevents the mass retrieve and should be used on endpoints, auths, aors, and contacts. It should NOT be used for identifies, registrations and transports since these MUST be retrieved in bulk. Example sorcery.conf: [res_pjsip] endpoint=config,pjsip.conf,criteria=type=endpoint endpoint=realtime,ps_endpoints,allow_unqualified_fetch=error ASTERISK-25826 #close Reported-by: Ross Beer Tested-by: Ross Beer Change-Id: Id2691e447db90892890036e663aaf907b2dc1c67
2016-03-08 21:55:30 +00:00
struct sorcery_config *config = data;
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
RAII_VAR(struct ast_variable *, fields, ast_sorcery_objectset_create(sorcery, object), ast_variables_destroy);
if (!fields) {
return -1;
}
res_sorcery_realtime: fix bug when successful UPDATE is treated as failed If the SQL UPDATE statement changes nothing then SQLRowCount returns 0. This value should be treated as success. But the function sorcery_realtime_update treats it as failed. This bug was found using stress tests on PJSIP. If there are 2 consecutive SIP REGISTER requests with the same contact data during 1 second then res_pjsip_registrar adds contact location on 1st request and tries to update contact location on 2nd. The update fails and res_pjsip_registrar even removes correct contact location. The test "object_update_uncreated" was removed from test_sorcery_realtime.c because it's now a valid situation. This patch also adds missing debug of extra SQL parameter. ASTERISK-26172 #close Change-Id: I05a7f3051455336c9dda29efc229decf86071303
2016-07-04 21:38:57 +00:00
return (ast_update_realtime_fields(config->family, UUID_FIELD, ast_sorcery_object_get_id(object), fields) < 0) ? -1 : 0;
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
}
static int sorcery_realtime_delete(const struct ast_sorcery *sorcery, void *data, void *object)
{
sorcery/res_pjsip: Refactor for realtime performance There were a number of places in the res_pjsip stack that were getting all endpoints or all aors, and then filtering them locally. A good example is pjsip_options which, on startup, retrieves all endpoints, then the aors for those endpoints, then tests the aors to see if the qualify_frequency is > 0. One issue was that it never did anything with the endpoints other than retrieve the aors so we probably could have skipped a step and just retrieved all aors. But nevermind. This worked reasonably well with local config files but with a realtime backend and thousands of objects, this was a nightmare. The issue really boiled down to the fact that while realtime supports predicates that are passed to the database engine, the non-realtime sorcery backends didn't. They do now. The realtime engines have a scheme for doing simple comparisons. They take in an ast_variable (or list) for matching, and the name of each variable can contain an operator. For instance, a name of "qualify_frequency >" and a value of "0" would create a SQL predicate that looks like "where qualify_frequency > '0'". If there's no operator after the name, the engines add an '=' so a simple name of "qualify_frequency" and a value of "10" would return exact matches. The non-realtime backends decide whether to include an object in a result set by calling ast_sorcery_changeset_create on every object in the internal container. However, ast_sorcery_changeset_create only does exact string matches though so a name of "qualify_frequency >" and a value of "0" returns nothing because the literal "qualify_frequency >" doesn't match any name in the objset set. So, the real task was to create a generic string matcher that can take a left value, operator and a right value and perform the match. To that end, strings.c has a new ast_strings_match(left, operator, right) function. Left and right are the strings to operate on and the operator can be a string containing any of the following: = (or NULL or ""), !=, >, >=, <, <=, like or regex. If the operator is like or regex, the right string should be a %-pattern or a regex expression. If both left and right can be converted to float, then a numeric comparison is performed, otherwise a string comparison is performed. To use this new function on ast_variables, 2 new functions were added to config.c. One that compares 2 ast_variables, and one that compares 2 ast_variable lists. The former is useful when you want to compare 2 ast_variables that happen to be in a list but don't want to traverse the list. The latter will traverse the right list and return true if all the variables in it match the left list. Now, the backends' fields_cmp functions call ast_variable_lists_match instead of ast_sorcery_changeset_create and they can now process the same syntax as the realtime engines. The realtime backend just passes the variable list unaltered to the engine. The only gotcha is that there's no common realtime engine support for regex so that's been noted in the api docs for ast_sorcery_retrieve_by_fields. Only one more change to sorcery was done... A new config flag "allow_unqualified_fetch" was added to reg_sorcery_realtime. "no": ignore fetches if no predicate fields were supplied. "error": same as no but emit an error. (good for testing) "yes": allow (the default); "warn": allow but emit a warning. (good for testing) Now on to res_pjsip... pjsip_options was modified to retrieve aors with qualify_frequency > 0 rather than all endpoints then all aors. Not only was this a big improvement in realtime retrieval but even for config files there's an improvement because we're not going through endpoints anymore. res_pjsip_mwi was modified to retieve only endpoints with something in the mailboxes field instead of all endpoints then testing mailboxes. res_pjsip_registrar_expire was completely refactored. It was retrieving all contacts then setting up scheduler entries to check for expiration. Now, it's a single thread (like keepalive) that periodically retrieves only contacts whose expiration time is < now and deletes them. A new contact_expiration_check_interval was added to global with a default of 30 seconds. Ross Beer reports that with this patch, his Asterisk startup time dropped from around an hour to under 30 seconds. There are still objects that can't be filtered at the database like identifies, transports, and registrations. These are not going to be anywhere near as numerous as endpoints, aors, auths, contacts however. Back to allow_unqualified_fetch. If this is set to yes and you have a very large number of objects in the database, the pjsip CLI commands will attempt to retrive ALL of them if not qualified with a LIKE. Worse, if you type "pjsip show endpoint <tab>" guess what's going to happen? :) Having a cache helps but all the objects will have to be retrieved at least once to fill the cache. Setting allow_unqualified_fetch=no prevents the mass retrieve and should be used on endpoints, auths, aors, and contacts. It should NOT be used for identifies, registrations and transports since these MUST be retrieved in bulk. Example sorcery.conf: [res_pjsip] endpoint=config,pjsip.conf,criteria=type=endpoint endpoint=realtime,ps_endpoints,allow_unqualified_fetch=error ASTERISK-25826 #close Reported-by: Ross Beer Tested-by: Ross Beer Change-Id: Id2691e447db90892890036e663aaf907b2dc1c67
2016-03-08 21:55:30 +00:00
struct sorcery_config *config = data;
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
sorcery/res_pjsip: Refactor for realtime performance There were a number of places in the res_pjsip stack that were getting all endpoints or all aors, and then filtering them locally. A good example is pjsip_options which, on startup, retrieves all endpoints, then the aors for those endpoints, then tests the aors to see if the qualify_frequency is > 0. One issue was that it never did anything with the endpoints other than retrieve the aors so we probably could have skipped a step and just retrieved all aors. But nevermind. This worked reasonably well with local config files but with a realtime backend and thousands of objects, this was a nightmare. The issue really boiled down to the fact that while realtime supports predicates that are passed to the database engine, the non-realtime sorcery backends didn't. They do now. The realtime engines have a scheme for doing simple comparisons. They take in an ast_variable (or list) for matching, and the name of each variable can contain an operator. For instance, a name of "qualify_frequency >" and a value of "0" would create a SQL predicate that looks like "where qualify_frequency > '0'". If there's no operator after the name, the engines add an '=' so a simple name of "qualify_frequency" and a value of "10" would return exact matches. The non-realtime backends decide whether to include an object in a result set by calling ast_sorcery_changeset_create on every object in the internal container. However, ast_sorcery_changeset_create only does exact string matches though so a name of "qualify_frequency >" and a value of "0" returns nothing because the literal "qualify_frequency >" doesn't match any name in the objset set. So, the real task was to create a generic string matcher that can take a left value, operator and a right value and perform the match. To that end, strings.c has a new ast_strings_match(left, operator, right) function. Left and right are the strings to operate on and the operator can be a string containing any of the following: = (or NULL or ""), !=, >, >=, <, <=, like or regex. If the operator is like or regex, the right string should be a %-pattern or a regex expression. If both left and right can be converted to float, then a numeric comparison is performed, otherwise a string comparison is performed. To use this new function on ast_variables, 2 new functions were added to config.c. One that compares 2 ast_variables, and one that compares 2 ast_variable lists. The former is useful when you want to compare 2 ast_variables that happen to be in a list but don't want to traverse the list. The latter will traverse the right list and return true if all the variables in it match the left list. Now, the backends' fields_cmp functions call ast_variable_lists_match instead of ast_sorcery_changeset_create and they can now process the same syntax as the realtime engines. The realtime backend just passes the variable list unaltered to the engine. The only gotcha is that there's no common realtime engine support for regex so that's been noted in the api docs for ast_sorcery_retrieve_by_fields. Only one more change to sorcery was done... A new config flag "allow_unqualified_fetch" was added to reg_sorcery_realtime. "no": ignore fetches if no predicate fields were supplied. "error": same as no but emit an error. (good for testing) "yes": allow (the default); "warn": allow but emit a warning. (good for testing) Now on to res_pjsip... pjsip_options was modified to retrieve aors with qualify_frequency > 0 rather than all endpoints then all aors. Not only was this a big improvement in realtime retrieval but even for config files there's an improvement because we're not going through endpoints anymore. res_pjsip_mwi was modified to retieve only endpoints with something in the mailboxes field instead of all endpoints then testing mailboxes. res_pjsip_registrar_expire was completely refactored. It was retrieving all contacts then setting up scheduler entries to check for expiration. Now, it's a single thread (like keepalive) that periodically retrieves only contacts whose expiration time is < now and deletes them. A new contact_expiration_check_interval was added to global with a default of 30 seconds. Ross Beer reports that with this patch, his Asterisk startup time dropped from around an hour to under 30 seconds. There are still objects that can't be filtered at the database like identifies, transports, and registrations. These are not going to be anywhere near as numerous as endpoints, aors, auths, contacts however. Back to allow_unqualified_fetch. If this is set to yes and you have a very large number of objects in the database, the pjsip CLI commands will attempt to retrive ALL of them if not qualified with a LIKE. Worse, if you type "pjsip show endpoint <tab>" guess what's going to happen? :) Having a cache helps but all the objects will have to be retrieved at least once to fill the cache. Setting allow_unqualified_fetch=no prevents the mass retrieve and should be used on endpoints, auths, aors, and contacts. It should NOT be used for identifies, registrations and transports since these MUST be retrieved in bulk. Example sorcery.conf: [res_pjsip] endpoint=config,pjsip.conf,criteria=type=endpoint endpoint=realtime,ps_endpoints,allow_unqualified_fetch=error ASTERISK-25826 #close Reported-by: Ross Beer Tested-by: Ross Beer Change-Id: Id2691e447db90892890036e663aaf907b2dc1c67
2016-03-08 21:55:30 +00:00
return (ast_destroy_realtime_fields(config->family, UUID_FIELD, ast_sorcery_object_get_id(object), NULL) <= 0) ? -1 : 0;
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
}
static void *sorcery_realtime_open(const char *data)
{
sorcery/res_pjsip: Refactor for realtime performance There were a number of places in the res_pjsip stack that were getting all endpoints or all aors, and then filtering them locally. A good example is pjsip_options which, on startup, retrieves all endpoints, then the aors for those endpoints, then tests the aors to see if the qualify_frequency is > 0. One issue was that it never did anything with the endpoints other than retrieve the aors so we probably could have skipped a step and just retrieved all aors. But nevermind. This worked reasonably well with local config files but with a realtime backend and thousands of objects, this was a nightmare. The issue really boiled down to the fact that while realtime supports predicates that are passed to the database engine, the non-realtime sorcery backends didn't. They do now. The realtime engines have a scheme for doing simple comparisons. They take in an ast_variable (or list) for matching, and the name of each variable can contain an operator. For instance, a name of "qualify_frequency >" and a value of "0" would create a SQL predicate that looks like "where qualify_frequency > '0'". If there's no operator after the name, the engines add an '=' so a simple name of "qualify_frequency" and a value of "10" would return exact matches. The non-realtime backends decide whether to include an object in a result set by calling ast_sorcery_changeset_create on every object in the internal container. However, ast_sorcery_changeset_create only does exact string matches though so a name of "qualify_frequency >" and a value of "0" returns nothing because the literal "qualify_frequency >" doesn't match any name in the objset set. So, the real task was to create a generic string matcher that can take a left value, operator and a right value and perform the match. To that end, strings.c has a new ast_strings_match(left, operator, right) function. Left and right are the strings to operate on and the operator can be a string containing any of the following: = (or NULL or ""), !=, >, >=, <, <=, like or regex. If the operator is like or regex, the right string should be a %-pattern or a regex expression. If both left and right can be converted to float, then a numeric comparison is performed, otherwise a string comparison is performed. To use this new function on ast_variables, 2 new functions were added to config.c. One that compares 2 ast_variables, and one that compares 2 ast_variable lists. The former is useful when you want to compare 2 ast_variables that happen to be in a list but don't want to traverse the list. The latter will traverse the right list and return true if all the variables in it match the left list. Now, the backends' fields_cmp functions call ast_variable_lists_match instead of ast_sorcery_changeset_create and they can now process the same syntax as the realtime engines. The realtime backend just passes the variable list unaltered to the engine. The only gotcha is that there's no common realtime engine support for regex so that's been noted in the api docs for ast_sorcery_retrieve_by_fields. Only one more change to sorcery was done... A new config flag "allow_unqualified_fetch" was added to reg_sorcery_realtime. "no": ignore fetches if no predicate fields were supplied. "error": same as no but emit an error. (good for testing) "yes": allow (the default); "warn": allow but emit a warning. (good for testing) Now on to res_pjsip... pjsip_options was modified to retrieve aors with qualify_frequency > 0 rather than all endpoints then all aors. Not only was this a big improvement in realtime retrieval but even for config files there's an improvement because we're not going through endpoints anymore. res_pjsip_mwi was modified to retieve only endpoints with something in the mailboxes field instead of all endpoints then testing mailboxes. res_pjsip_registrar_expire was completely refactored. It was retrieving all contacts then setting up scheduler entries to check for expiration. Now, it's a single thread (like keepalive) that periodically retrieves only contacts whose expiration time is < now and deletes them. A new contact_expiration_check_interval was added to global with a default of 30 seconds. Ross Beer reports that with this patch, his Asterisk startup time dropped from around an hour to under 30 seconds. There are still objects that can't be filtered at the database like identifies, transports, and registrations. These are not going to be anywhere near as numerous as endpoints, aors, auths, contacts however. Back to allow_unqualified_fetch. If this is set to yes and you have a very large number of objects in the database, the pjsip CLI commands will attempt to retrive ALL of them if not qualified with a LIKE. Worse, if you type "pjsip show endpoint <tab>" guess what's going to happen? :) Having a cache helps but all the objects will have to be retrieved at least once to fill the cache. Setting allow_unqualified_fetch=no prevents the mass retrieve and should be used on endpoints, auths, aors, and contacts. It should NOT be used for identifies, registrations and transports since these MUST be retrieved in bulk. Example sorcery.conf: [res_pjsip] endpoint=config,pjsip.conf,criteria=type=endpoint endpoint=realtime,ps_endpoints,allow_unqualified_fetch=error ASTERISK-25826 #close Reported-by: Ross Beer Tested-by: Ross Beer Change-Id: Id2691e447db90892890036e663aaf907b2dc1c67
2016-03-08 21:55:30 +00:00
struct sorcery_config *config;
char *tmp;
char *family;
char *option;
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
/* We require a prefix for family string generation, or else stuff could mix together */
sorcery/res_pjsip: Refactor for realtime performance There were a number of places in the res_pjsip stack that were getting all endpoints or all aors, and then filtering them locally. A good example is pjsip_options which, on startup, retrieves all endpoints, then the aors for those endpoints, then tests the aors to see if the qualify_frequency is > 0. One issue was that it never did anything with the endpoints other than retrieve the aors so we probably could have skipped a step and just retrieved all aors. But nevermind. This worked reasonably well with local config files but with a realtime backend and thousands of objects, this was a nightmare. The issue really boiled down to the fact that while realtime supports predicates that are passed to the database engine, the non-realtime sorcery backends didn't. They do now. The realtime engines have a scheme for doing simple comparisons. They take in an ast_variable (or list) for matching, and the name of each variable can contain an operator. For instance, a name of "qualify_frequency >" and a value of "0" would create a SQL predicate that looks like "where qualify_frequency > '0'". If there's no operator after the name, the engines add an '=' so a simple name of "qualify_frequency" and a value of "10" would return exact matches. The non-realtime backends decide whether to include an object in a result set by calling ast_sorcery_changeset_create on every object in the internal container. However, ast_sorcery_changeset_create only does exact string matches though so a name of "qualify_frequency >" and a value of "0" returns nothing because the literal "qualify_frequency >" doesn't match any name in the objset set. So, the real task was to create a generic string matcher that can take a left value, operator and a right value and perform the match. To that end, strings.c has a new ast_strings_match(left, operator, right) function. Left and right are the strings to operate on and the operator can be a string containing any of the following: = (or NULL or ""), !=, >, >=, <, <=, like or regex. If the operator is like or regex, the right string should be a %-pattern or a regex expression. If both left and right can be converted to float, then a numeric comparison is performed, otherwise a string comparison is performed. To use this new function on ast_variables, 2 new functions were added to config.c. One that compares 2 ast_variables, and one that compares 2 ast_variable lists. The former is useful when you want to compare 2 ast_variables that happen to be in a list but don't want to traverse the list. The latter will traverse the right list and return true if all the variables in it match the left list. Now, the backends' fields_cmp functions call ast_variable_lists_match instead of ast_sorcery_changeset_create and they can now process the same syntax as the realtime engines. The realtime backend just passes the variable list unaltered to the engine. The only gotcha is that there's no common realtime engine support for regex so that's been noted in the api docs for ast_sorcery_retrieve_by_fields. Only one more change to sorcery was done... A new config flag "allow_unqualified_fetch" was added to reg_sorcery_realtime. "no": ignore fetches if no predicate fields were supplied. "error": same as no but emit an error. (good for testing) "yes": allow (the default); "warn": allow but emit a warning. (good for testing) Now on to res_pjsip... pjsip_options was modified to retrieve aors with qualify_frequency > 0 rather than all endpoints then all aors. Not only was this a big improvement in realtime retrieval but even for config files there's an improvement because we're not going through endpoints anymore. res_pjsip_mwi was modified to retieve only endpoints with something in the mailboxes field instead of all endpoints then testing mailboxes. res_pjsip_registrar_expire was completely refactored. It was retrieving all contacts then setting up scheduler entries to check for expiration. Now, it's a single thread (like keepalive) that periodically retrieves only contacts whose expiration time is < now and deletes them. A new contact_expiration_check_interval was added to global with a default of 30 seconds. Ross Beer reports that with this patch, his Asterisk startup time dropped from around an hour to under 30 seconds. There are still objects that can't be filtered at the database like identifies, transports, and registrations. These are not going to be anywhere near as numerous as endpoints, aors, auths, contacts however. Back to allow_unqualified_fetch. If this is set to yes and you have a very large number of objects in the database, the pjsip CLI commands will attempt to retrive ALL of them if not qualified with a LIKE. Worse, if you type "pjsip show endpoint <tab>" guess what's going to happen? :) Having a cache helps but all the objects will have to be retrieved at least once to fill the cache. Setting allow_unqualified_fetch=no prevents the mass retrieve and should be used on endpoints, auths, aors, and contacts. It should NOT be used for identifies, registrations and transports since these MUST be retrieved in bulk. Example sorcery.conf: [res_pjsip] endpoint=config,pjsip.conf,criteria=type=endpoint endpoint=realtime,ps_endpoints,allow_unqualified_fetch=error ASTERISK-25826 #close Reported-by: Ross Beer Tested-by: Ross Beer Change-Id: Id2691e447db90892890036e663aaf907b2dc1c67
2016-03-08 21:55:30 +00:00
if (ast_strlen_zero(data)) {
return NULL;
}
tmp = ast_strdupa(data);
family = strsep(&tmp, ",");
if (!ast_realtime_is_mapping_defined(family)) {
return NULL;
}
config = ast_calloc(1, sizeof(*config) + strlen(family) + 1);
if (!config) {
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
return NULL;
}
sorcery/res_pjsip: Refactor for realtime performance There were a number of places in the res_pjsip stack that were getting all endpoints or all aors, and then filtering them locally. A good example is pjsip_options which, on startup, retrieves all endpoints, then the aors for those endpoints, then tests the aors to see if the qualify_frequency is > 0. One issue was that it never did anything with the endpoints other than retrieve the aors so we probably could have skipped a step and just retrieved all aors. But nevermind. This worked reasonably well with local config files but with a realtime backend and thousands of objects, this was a nightmare. The issue really boiled down to the fact that while realtime supports predicates that are passed to the database engine, the non-realtime sorcery backends didn't. They do now. The realtime engines have a scheme for doing simple comparisons. They take in an ast_variable (or list) for matching, and the name of each variable can contain an operator. For instance, a name of "qualify_frequency >" and a value of "0" would create a SQL predicate that looks like "where qualify_frequency > '0'". If there's no operator after the name, the engines add an '=' so a simple name of "qualify_frequency" and a value of "10" would return exact matches. The non-realtime backends decide whether to include an object in a result set by calling ast_sorcery_changeset_create on every object in the internal container. However, ast_sorcery_changeset_create only does exact string matches though so a name of "qualify_frequency >" and a value of "0" returns nothing because the literal "qualify_frequency >" doesn't match any name in the objset set. So, the real task was to create a generic string matcher that can take a left value, operator and a right value and perform the match. To that end, strings.c has a new ast_strings_match(left, operator, right) function. Left and right are the strings to operate on and the operator can be a string containing any of the following: = (or NULL or ""), !=, >, >=, <, <=, like or regex. If the operator is like or regex, the right string should be a %-pattern or a regex expression. If both left and right can be converted to float, then a numeric comparison is performed, otherwise a string comparison is performed. To use this new function on ast_variables, 2 new functions were added to config.c. One that compares 2 ast_variables, and one that compares 2 ast_variable lists. The former is useful when you want to compare 2 ast_variables that happen to be in a list but don't want to traverse the list. The latter will traverse the right list and return true if all the variables in it match the left list. Now, the backends' fields_cmp functions call ast_variable_lists_match instead of ast_sorcery_changeset_create and they can now process the same syntax as the realtime engines. The realtime backend just passes the variable list unaltered to the engine. The only gotcha is that there's no common realtime engine support for regex so that's been noted in the api docs for ast_sorcery_retrieve_by_fields. Only one more change to sorcery was done... A new config flag "allow_unqualified_fetch" was added to reg_sorcery_realtime. "no": ignore fetches if no predicate fields were supplied. "error": same as no but emit an error. (good for testing) "yes": allow (the default); "warn": allow but emit a warning. (good for testing) Now on to res_pjsip... pjsip_options was modified to retrieve aors with qualify_frequency > 0 rather than all endpoints then all aors. Not only was this a big improvement in realtime retrieval but even for config files there's an improvement because we're not going through endpoints anymore. res_pjsip_mwi was modified to retieve only endpoints with something in the mailboxes field instead of all endpoints then testing mailboxes. res_pjsip_registrar_expire was completely refactored. It was retrieving all contacts then setting up scheduler entries to check for expiration. Now, it's a single thread (like keepalive) that periodically retrieves only contacts whose expiration time is < now and deletes them. A new contact_expiration_check_interval was added to global with a default of 30 seconds. Ross Beer reports that with this patch, his Asterisk startup time dropped from around an hour to under 30 seconds. There are still objects that can't be filtered at the database like identifies, transports, and registrations. These are not going to be anywhere near as numerous as endpoints, aors, auths, contacts however. Back to allow_unqualified_fetch. If this is set to yes and you have a very large number of objects in the database, the pjsip CLI commands will attempt to retrive ALL of them if not qualified with a LIKE. Worse, if you type "pjsip show endpoint <tab>" guess what's going to happen? :) Having a cache helps but all the objects will have to be retrieved at least once to fill the cache. Setting allow_unqualified_fetch=no prevents the mass retrieve and should be used on endpoints, auths, aors, and contacts. It should NOT be used for identifies, registrations and transports since these MUST be retrieved in bulk. Example sorcery.conf: [res_pjsip] endpoint=config,pjsip.conf,criteria=type=endpoint endpoint=realtime,ps_endpoints,allow_unqualified_fetch=error ASTERISK-25826 #close Reported-by: Ross Beer Tested-by: Ross Beer Change-Id: Id2691e447db90892890036e663aaf907b2dc1c67
2016-03-08 21:55:30 +00:00
strcpy(config->family, family); /* Safe */
config->fetch = UNQUALIFIED_FETCH_YES;
while ((option = strsep(&tmp, ","))) {
char *name = strsep(&option, "=");
char *value = option;
if (!strcasecmp(name, "allow_unqualified_fetch")) {
if (ast_strlen_zero(value) || !strcasecmp(value, "yes")) {
config->fetch = UNQUALIFIED_FETCH_YES;
} else if (!strcasecmp(value, "no")) {
config->fetch = UNQUALIFIED_FETCH_NO;
} else if (!strcasecmp(value, "warn")) {
config->fetch = UNQUALIFIED_FETCH_WARN;
} else if (!strcasecmp(value, "error")) {
config->fetch = UNQUALIFIED_FETCH_ERROR;
} else {
ast_log(LOG_ERROR, "Unrecognized value in %s:%s: '%s'\n", family, name, value);
return NULL;
}
} else {
ast_log(LOG_ERROR, "Unrecognized option in %s: '%s'\n", family, name);
return NULL;
}
}
return config;
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
}
static void sorcery_realtime_close(void *data)
{
ast_free(data);
}
static int load_module(void)
{
if (ast_sorcery_wizard_register(&realtime_object_wizard)) {
return AST_MODULE_LOAD_DECLINE;
}
return AST_MODULE_LOAD_SUCCESS;
}
static int unload_module(void)
{
ast_sorcery_wizard_unregister(&realtime_object_wizard);
return 0;
}
AST_MODULE_INFO(ASTERISK_GPL_KEY, AST_MODFLAG_GLOBAL_SYMBOLS | AST_MODFLAG_LOAD_ORDER, "Sorcery Realtime Object Wizard",
Add module support level to ast_module_info structure. Print it in CLI "module show" . ASTERISK-23919 #close Reported by Malcolm Davenport Review: https://reviewboard.asterisk.org/r/3802 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@419592 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-07-25 16:47:17 +00:00
.support_level = AST_MODULE_SUPPORT_CORE,
Add support for a realtime sorcery module. This change does the following: 1. Adds the sorcery realtime module 2. Adds unit tests for the sorcery realtime module 3. Changes the realtime core to use an ast_variable list instead of variadic arguments 4. Changes all realtime drivers to accept an ast_variable list Review: https://reviewboard.asterisk.org/r/2424/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@386731 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2013-04-27 12:01:29 +00:00
.load = load_module,
.unload = unload_module,
.load_pri = AST_MODPRI_REALTIME_DRIVER,
);