Commit Graph

1289 Commits

Author SHA1 Message Date
Sukchan Lee 433d5f6bf3 Distinguish the type of session creation 2022-04-14 17:34:55 +09:00
Sukchan Lee 80cd9d725f [SBI] fixed HTTP2 header values is 0 (#1488) 2022-04-14 11:28:37 +09:00
Pau Espin e3da7c9934
Rename GTPv2C specifics to gtp2 prefix (#1485)
In the past only GTPv2C was supported, and had the "gtp" generic prefix.
Later on, GTPv1C support was added, and "gtp1" prefix was used.
Let's move GTPv2C specific bits to have "gtp2" prefix too, and leave
"gtp" prefix for generic stuff among different GTP versions.
2022-04-13 07:07:39 +09:00
Sukchan Lee e5ff03b249 [SBI] Change the reference count (#1440)
Change the client's reference count method
to use the same method as nf_instance
2022-04-10 20:09:27 +09:00
Sukchan Lee 49c34605db [SBI] fixed the SMF discover bug (#1440) 2022-04-09 13:24:40 +09:00
Sukchan Lee ff242cd6ea [Gy] Follow-up PRs (#1479) 2022-04-09 10:41:02 +09:00
Pau Espin 2be12903cb
[SMF] Introduce optional Gy interface support (#1479)
The use of the Gy interface (SMF acting as CTF towards an OCS node) is
mandated through configuration file. Default value "enable: auto" will
only make use of it in case a Diameter peer announcing support for the
Credit-Control Application is found.

Upon subscriber session creation, and after auth check over Gx, the SMF
will create a Gy session with the OCS and only after that step the SMF
will accept the session back to the subscriber.
The OCS may then grant some traffic volumes/time and ask to be notified
back with updated measurements.
In order to get the measurements, the SMF relies on PFCP URR configured
to the UPF through Session Repoort Request messages.
When closing the subscriber session, the SMF will also terminate the Gy
session at the OCS.

So far only some specifics parts of the Gy interface as well as the PFCP
side are implemented. Those should be enough to at least have
volume/time thresholds granted by the OCS, which then will be able to
track subsriber resource use.

This patch doesn't implement the OCS side of the Gy interface, that's
left as a future exercise. The interface was tested using an OCS
emulator implemented in TTCN-3 [1]

[1] https://cgit.osmocom.org/osmo-ttcn3-hacks/
2022-04-09 08:26:28 +09:00
Pau Espin 7455424d29
Fix trailing whitespace (#1478) 2022-04-09 08:19:59 +09:00
Sukchan Lee 0b3df3e3db Revert "Introduce Gy interface (#1471)"
This reverts commit 106f2729c3.
2022-04-09 00:45:26 +09:00
Pau Espin 106f2729c3
Introduce Gy interface (#1471)
TODO:

* Use an event for the report, like SMF_EVT_N4_TIMER?

*  Properly set Service identifier in Gy CCR

* SMF: Properly set pkt/octet volumes in Gy CCR
** Update when receiving PFCP Modify Response.

* Figure out best way to require Gy through config file in open5gs-smfd.

* Create a new sess-sm.c which handles smf_sess_t state through
  Gx+Gy+PFCP creation, modification and tear down. This way we can do
stuff in parallel, for instance Gx+Gy. It will alsoavoid duplicating
some code paths due to Gy being optional.
2022-04-09 00:08:58 +09:00
Pau Espin fb8ebcdbea
[UPF] Add initial support for URR Usage Report (#1476) 2022-04-08 23:10:42 +09:00
Pau Espin 52672cff65
[DIAM] Add Gy interface support to lib/diameter (#1474)
This commit adds required blocks in lib/diameter to be able to handle Gy
messages later in open5gs apps.

The Gy interface is mainly decribed in 3GPP TS 32.299  and 3GPP TS
32.251, which in turn refers to Diameter protcols defined in RFC4006.
This interface allows charging managment through an external OCS node.
2022-04-08 21:13:02 +09:00
Pau Espin 77f7bb7af7
[PFCP] Use proper IE to signal triggers in Usage Report (#1472)
The "Report Triggers" is sent SMF->UPF to tell in which situations a
report should be sent.
The "Usage Report Trigger" is sent UPF->SMF to indicate which situation
triggered the report.
2022-04-08 21:12:42 +09:00
Sukchan Lee 3121d183f0 [HSS/PCF] Store IMEISV instead of IMEI (#1464) 2022-04-02 00:46:05 +09:00
EugeneBogush f557fc8eaa
Add/update current imei of UE in subscriber profile (#1464)
* first commit

* fix

Co-authored-by: root <root@lfv.unassigned-domain>
2022-03-31 19:35:30 +09:00
Pau Espin fbfb72c1b7
[PFCP] Support multiple Usage Report IEs in all messages (#1461)
* [PFCP] Properly check down_data_report inclusion building SessionReportRequest

* [PFCP] Support multiple Usage Report IEs in all messages
2022-03-31 09:43:35 +09:00
Sukchan Lee 038fb680f4 Revert configuration for backward compatibility 2022-03-31 02:15:13 +09:00
Pau Espin 4b8d3a845a
[PFCP] Support associating multiple URRs to a PDR (#1456)
* [PFCP] Fix trailing whitespace in message generation files

* [PFCP] message gen: Support multiple URR ID IEs in Create PDR group

* [PFCP] Support associating multiple URRs to a PDR

According to 3GPP TS 29.244:
"""
A PDR shall contain:
- zero, one or more URRs, which contains instructions related to traffic measurement and reporting.
"""
2022-03-28 21:56:58 +09:00
Sukchan Lee 12c656efd0 Improve Slice/DNN selection (#1438), (#1440) 2022-03-27 17:50:31 +09:00
Sukchan Lee ced37a6201 fix the slice issues (#1438), (#1440) 2022-03-26 23:53:53 +09:00
Pau Espin 38e2dee17f
[PFCP] Handle Measurement Information in Create/Update URR IEs (#1453) 2022-03-26 00:33:32 +09:00
Pau Espin 89636fa8d3
[PFCP] Handle Measurement Information in Create/Update URR IEs (#1452) 2022-03-26 00:07:34 +09:00
Pau Espin cc0d147f5c
Fix trailing whitespace (#1451) 2022-03-25 23:35:27 +09:00
Pau Espin 12bac7a917
[PFCP] Avoid requiring optional fields in Update URR IE (#1449) 2022-03-25 05:58:40 +09:00
Sukchan Lee 2cb06b3085 remove pool related configuration (#1431) 2022-03-23 15:14:18 +09:00
Sukchan Lee 37af21a88d [MEM] fix the pkbuf problem (#1431) 2022-03-22 22:47:45 +09:00
Sukchan Lee bf77318602 [MME] re-factor to check a piggybacked ESM (#1431) 2022-03-16 20:48:48 +09:00
Sukchan Lee a6ec206998 Change EPERM to OGS_EPERM (#1436,#1404) 2022-03-16 20:32:19 +09:00
Pau Espin f060da45d5
[CORE] time: Add APIs to manage NTP 32-bit timestamps (#1439)
These will be further needed in PFCP in the future, as well as in other
Diameter based interfaces (such as Gy).
Let's put all implementation details in APIs so that devs don't need to
care about those details every time.
2022-03-16 11:25:30 +09:00
Pau Espin 80ce991aa6
[CORE] Improve SO_BINDTODEVICE error log line (#1436) 2022-03-16 11:25:21 +09:00
Sukchan Lee bcf53124d5 Improve the socket option configuration (#1404)
o GTP-C Option (Default)
  - so_bindtodevice : NULL

  gtpc:
    addr: 127.0.0.7
    option:
      so_bindtodevice: vrf-blue

o GTP-U Option (Default)
  - so_bindtodevice : NULL

  gtpu:
    addr: 127.0.0.7
    option:
      so_bindtodevice: vrf-blue

o PFCP Option (Default)
  - so_bindtodevice : NULL

  pfcp:
    addr: 127.0.0.7
    option:
      so_bindtodevice: vrf-blue

o SBI Option (Default)
  - tcp_nodelay : true
  - so_linger.l_onoff : false

  sbi:
    addr: 127.0.0.10
    option:
      tcp_nodelay: false
      so_linger:
        l_onoff: true
        l_linger: 10

o NGAP Option (Default)
  - sctp_nodelay : true
  - so_linger.l_onoff : false
ngap:
  addr: 127.0.0.5
  option:
    stcp_nodelay: false
    so_linger:
      l_onoff: true
      l_linger: 10

o NGAP SCTP Option (Default)
  - spp_hbinterval : 5000 (5secs)
  - spp_sackdelay : 200 (200ms)
  - srto_initial : 3000 (3secs)
  - srto_min : 1000 (1sec)
  - srto_max : 5000 (5secs)
  - sinit_num_ostreams : 30
  - sinit_max_instreams : 65535
  - sinit_max_attempts : 4
  - sinit_max_init_timeo : 8000(8secs)
ngap:
  addr: 127.0.0.5
  option:
    sctp:
      spp_hbinterval : 5000
      spp_sackdelay : 200
      srto_initial : 3000
      srto_min : 1000
      srto_max : 5000
      sinit_num_ostreams : 30
      sinit_max_instreams : 65535
      sinit_max_attempts : 4
      sinit_max_init_timeo : 8000
2022-03-15 22:03:50 +09:00
Pau Espin dd5abb79b4
[PFCP] Fix wrong endianess enc of some URR values (#1433)
* [PFCP] Fix trailing whitespace

* [PFCP] Fix wrong endianess enc of some URR values

u32 tlvs are already converted to big endian automatically. Manually
doing so ends up in double conversion and hence in wrong endianness
being sent over the wire.
Similar issue was also fixed recently in the PFCP decoding path.
Related: https://github.com/open5gs/open5gs/issues/1349
2022-03-15 13:34:22 +09:00
Pau Espin 5b81802be9
[Gx] Fix trailing whitespace (#1420) 2022-03-08 21:50:20 +09:00
Pau Espin 4388f9bf3a
[Gx] Use OGS_DIAM_GX_APPLICATION_ID define instead of hardcoded val (#1419) 2022-03-08 21:50:05 +09:00
Sukchan Lee c1fc25958c Change default to bindtodevice:false 2022-03-08 19:49:16 +09:00
Sukchan Lee 253e2ad98a [GTP/PFCP] Support VRF (#1404) 2022-03-07 22:43:18 +09:00
Pau Espin 7bddc92322 [GTP] Support binding socket to device
This is useful, among other possible applications, to make use of VRFs [1],
in this case for GTP-C and GTP-U traffic in the PGW.

The bind_dev field is added to the ogs_socknode_t so that it's easy to
extend its use into lots of other sockets being set up based on config
file information.

[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/networking/vrf.rst
2022-03-02 20:07:51 +01:00
Sukchan Lee 3217e93124 [LINUX] initialize struct epoll_event (#1399) 2022-03-01 21:03:27 +09:00
Pau Espin d06acc7f37
Fix trailing whitespace (#1402) 2022-03-01 17:13:40 +09:00
Sukchan Lee efe37c1e84 [ASN1C] asn.1 issues in NGAP (#1394 from @nkout)
mouse07410/asn1c#89
Found when tried to encode NGAP_CauseRadioNetwork_release_due_to_pre_emption

mouse07410/asn1c#90
Found when tried to decode messages encoded with newer schema
2022-02-26 16:09:38 +09:00
Supreeth Herle 52bc4be547
Bugfix packet filter for pre rel11 LTE devices (#1393)
* Fix conversion from IPFilterRule to packet filter

As per 3GPP TS 24.008, following Packet filter component type identifier
are not supported on the LTE pre release-11 UEs:

IPv4 local address type
IPv6 remote address/prefix length type
IPv6 local address/prefix length type

And,
IPv6 remote address/prefix length type and
IPv6 local address/prefix length type shall be used when both MS and
Network support Local Address in TFTs.

This commit add logic to omit adding local address in packet filters
for compatibility with pre-release LTE 11 devices. The following parameter
could be used to toggle omit/no to omit behavior.

parameter:
	no_ipv4v6_local_addr_in_packet_filter: <true/false>

* Remove logic of supporting pre-release LTE 11 devices in PCRF
2022-02-25 22:11:51 +09:00
Pau Espin 3b6da3ddb2
[GTPv1C] Fix sequm not added to EchoResponse and VersionNotSupported (#1390)
According to TS 29.060 they should be added.

section 7.6:
"if it is a request for which a response has been defined, shall be sent
with a Sequence Number"

section 8.2:
"""
Sequence number flag (S) shall be set to "1"
...
For GTP-C messages not having a defined response message for a request
message, i.e. for messages Version Not Supported, RAN Information Relay
and Supported Extension Headers Notification, the Sequence Number shall
be ignored by the receiver.
"
2022-02-24 23:38:52 +09:00
Pau Espin dbd9e8bd78
[GTPv1C] Set EUA spare field to '1111' as per spec (#1388) 2022-02-24 06:03:54 +09:00
Sukchan Lee d61127afcc [SMF] fix the seperate DNS in PCO IPCP (#1358) 2022-02-23 17:38:07 +09:00
Pau Espin b0fc06ab6c
Fix trailing whitespace (#1381) 2022-02-22 07:00:06 +09:00
Sukchan Lee e3fa731c4a [SBI] fix the timezone string converter (#1376) 2022-02-20 20:18:32 +09:00
Sukchan Lee 420c1d5ad3 [SMF] fix the PCO bug (#1357, #1358)
- SMF can handle PCO ID PAP (0xc023)
- SMF adds DNS servers to PCO IPCP only if UE requested
2022-02-19 16:05:07 +09:00
Sukchan Lee b4f382d360 [MEM] fix pkbuf problem in multi-threads (#1353)
We should use talloc in multi-threads instead of pkbuf.
Now, pkbuf library only uses single-thread environment.
2022-02-19 09:47:44 +09:00
Pau Espin 137ed99dd5
[GTPv1C] Send conditional IE Reordering Required in CreatePDPCtxResp (#1374)
It seems to be actually mandatory in GGSN->SGN directon, though it is
ignored in Release99 upwards. Let's hardcode it to 0 for now, should be
fine in the majority of cases.
2022-02-19 06:36:54 +09:00
Pau Espin f7999a2cad
[GTPv1C] Fix outer message length field containing wrong value (#1373)
The mandatory part of the header must be excluded from the header, that
is flags(1)+type(1)+length(2)+teid(4) = 8 bytes.
2022-02-19 06:36:40 +09:00