Do not attempt to run "systemctl reload" on the open5gs services, unless
they are running. This fixes the logrotate service failing on this
postrotate script, if units are not running or not installed.
Move the config to the sgsn node instead of having a specific route with
specific format "default: route", since anyway internally it's already
applied to the sgsn object.
* [MME] Introduce aging timers
* Creating three new timers
* mirroring work done by gstaa on the AMF
* Implicit detach procedures added
* Fix for detach from unknown UE
* no Purge Timer, no config, expanded code
o Generate the private key as below.
$ openssl genpkey -algorithm X25519 -out /etc/open5gs/hnet/curve25519-1.key
$ openssl ecparam -name prime256v1 -genkey -conv_form compressed -out /etc/open5gs/hnet/secp256r1-2.key
o The private and public keys can be viewed with the command.
The public key is used when creating the SIM.
$ openssl pkey -in /etc/open5gs/hnet/curve25519-1.key -text
$ openssl ec -in /etc/open5gs/hnet/secp256r1-2.key -conv_form compressed -text
In ausf/udm.yaml
hnet:
o Home network public key identifier(PKI) value : 1
Protection scheme identifier : ECIES scheme profile A
- id: 1
scheme: 1
key: /etc/open5gs/hnet/curve25519-1.key
o Home network public key identifier(PKI) value : 2
Protection scheme identifier : ECIES scheme profile B
- id: 2
scheme: 2
key: /etc/open5gs/hnet/secp256r1-2.key
o Home network public key identifier(PKI) value : 3
Protection scheme identifier : ECIES scheme profile A
- id: 3
scheme: 1
key: /etc/open5gs/hnet/curve25519-1.key
o Home network public key identifier(PKI) value : 4
Protection scheme identifier : ECIES scheme profile B
- id: 4
scheme: 2
key: /etc/open5gs/hnet/secp256r1-2.key
Related to #1779
Expose metrics with labels according to ETSI TS 128 552 V16.13.0 in
PCF by using hash.
The metrics are named respecting the rule:
<generation>_<measurement_object_class>_<measurement_family_name>_<metric_name_as_in_TS_128_552>
Since slice itself is not unique, the plmnid label is exposed in
addition to snssai.
AM policy:
fivegs_pcffunction_pa_policyamassoreq and
fivegs_pcffunction_pa_policyamassosucc do not expose snssai label
since it is not available at the time of exposure.
plmnid is defined during AM policy processing, so not to lose the
difference to ...succ, the basic metric
fivegs_pcffunction_pa_policyamassoreq is preserved.
SM policy:
snssai is defined during SM policy processing, so not to lose the
difference to ...succ, the basic metric
fivegs_pcffunction_pa_policysmassoreq is preserved.
Those 2 basic metrics retain their position but are exposed with empty
labels.
Metrics with labels are called later, when the label values are known.
Exposed metrics example:
-standard counters:
fivegs_pcffunction_pa_policyamassoreq{plmnid=""} 3
fivegs_pcffunction_pa_policyamassoreq{plmnid="99970"} 3
fivegs_pcffunction_pa_policyamassosucc{plmnid="99970"} 3
fivegs_pcffunction_pa_policysmassoreq{plmnid="",snssai=""} 3
fivegs_pcffunction_pa_policysmassoreq{plmnid="99970",snssai="1000009"} 3
fivegs_pcffunction_pa_policysmassosucc{plmnid="99970",snssai="1000009"} 3
-nonstandard gauge (added for controlling purposes -
same metric as existing metric on AMF and SMF):
fivegs_pcffunction_pa_sessionnbr{plmnid="99970",snssai="1000009"} 0
Expose metrics with labels according to ETSI TS 128 552 V16.13.0 in
UPF by using hash.
The metrics are named respecting the rule:
<generation>_<measurement_object_class>_<measurement_family_name>_<metric_name_as_in_TS_128_552>
5qi is not available in UPF.
To present 5qi to the user, MN will have to maintain a table qfi->5qi
for each QoS flow (will have to get information from SMF).
So UPF has to expose qfi. qfi itself is not useful. When used, UPF will
have to expose additional label to define the session (e.g. source
interface).
Label dnn is set to value of APN/DNN received in Establishment.
Since SMF does not add APN/DNN to Establishment, the label is empty.
When APN/DNN will be set by SMF, it should be added to sess in UPF
and used in metrics on Modification and Deletion.
Both datavolumeqosleveln3upf are exposed in bytes.
MN is providing the transformation to kbits.
fivegs_upffunction_upf_qosflows should expose the number of QFIs used in
sessions, but exposes number of QER rules, which is currently equal to
QFIs.
The label snsssai is not provided since the slice is not available on UPF.
Exposed metrics example:
Standard counters:
fivegs_ep_n3_gtp_indatapktn3upf 28637
fivegs_ep_n3_gtp_outdatapktn3upf 14729
fivegs_upffunction_sm_n4sessionestabreq 4
fivegs_upffunction_sm_n4sessionestabfail{cause="66"} 1
fivegs_upffunction_sm_n4sessionestabfail{cause="71"} 68
fivegs_upffunction_sm_n4sessionestabfail{cause="68"} 4
fivegs_upffunction_sm_n4sessionestabfail{cause="72"} 15
fivegs_upffunction_sm_n4sessionestabfail{cause="75"} 3
fivegs_upffunction_sm_n4sessionestabfail{cause="65"} 4
fivegs_upffunction_sm_n4sessionreport 0
fivegs_upffunction_sm_n4sessionreportsucc 0
fivegs_ep_n3_gtp_indatavolumeqosleveln3upf{qfi="1"} 39792997
fivegs_ep_n3_gtp_outdatavolumeqosleveln3upf{qfi="1"} 737548
Nonstandard gauge (added for controlling purposes -
same metric as existing metric on AMF and SMF):
fivegs_upffunction_upf_sessionnbr 1
Standard gauge:
fivegs_upffunction_upf_qosflows{dnn=""} 1
Without this change, using metrics with core setup configurations
(configs/vonr.yaml for example) would not be possible. Having one
metrics section for whole config file causes every NF to start metrics
server on same port causing an abort.
* [HSS] Enable Change Streams
* Enable Events and Timers in HSS
* Integrate change streams in dbi
* mongodb should be configured with replica sets enabled to use feature
* Change streams are optional in HSS
* Timer will poll change stream for changes in the database
* As changes are detected, event is created to perform the correct
action
* Changes made as suggested
== Known limitation ==
Placing npcf-smpolicycontrol and pcf-policyauthorization
in different NFs is not supported. Both npcf-smpolicycontrol
and pcf-policyauthorization should be placed in the same NF.
* configs: use proper default IP address for metrics server
Let's use the IP address assigned to each process by default when
configuring the HTTP Prometheus server. Otherwise having several
processes listening on 0.0.0.0 cause collisions.
* configs: mme.yaml: Fix trailing whitespace
