input
stringlengths 48
5.67k
| output
stringlengths 0
15.3k
|
|---|---|
redeploy reconcile throw the error NedWorker error for ned_prepare_cli: Day 0 config on the device policy srte_c_3800_ep_61.61.15.10 bandwidth 500 color 3800 end-point ipv4 61.61.15.10 candidate-paths preference 100 dynamic pcep ! metric type igp ! ! ! User created SR policy without bandwidth and pcep. Then tried to redeploy reconcile. Then when deleted the service it went into error.policy srte_c_3800_ep_61.61.15.10 bandwidth 500 <----- this was color 3800 end-point ipv4 61.61.15.10 candidate-paths preference 100 dynamic pcep ! metric type igp ! ! ! <ERROR> 19-Apr-2023::17:05:37.193 NedWorker Ned-Worker-Thread-(cisco-iosxr-cli-7.46:cisco-iosxr-cli-7.46)-23: - NedWorker error for ned_prepare_clicom.tailf.ned.NedCliBaseTemplate$ApplyException: config in diff couldn't be parsed: skipped 1 line in context '/segment-routing/traffic-eng/policy' : (line 5) : ' color 3800'NOTE: this might indicate invalid usage of device model, set config on device and sync-from to verify config syntax/structure at com.tailf.packages.ned.iosxr.IosxrNedCli.doApplyConfig(IosxrNedCli.java:2824) ~ at com.tailf.packages.ned.iosxr.IosxrNedCli.applyConfig(IosxrNedCli.java:2756) ~ at com.tailf.ned.NedCliBaseTemplate.prepare(NedCliBaseTemplate.java:480) ~ at com.tailf.ned.NedWorker.dorun(NedWorker.java:1692) at com.tailf.ned.NedWorker.run(NedWorker.java:329) <INFO> 19-Apr-2023::17:06:35.042 DLMConnection Thread-3397: - DLM has closed the session.<INFO> 19-Apr-2023::17:06:35.264 DlmService Did-11-Worker-6: - ***** Executing the DLM Service | When we do re-deploy reconcile and there is device config related to the service which has been manually added, service would not take ownership of unrelated config. We try to delete the service, It will rollback only the service provisioned config and not the manually created config. If the presence of the manual config on service is not accepted by device, it would throw an error.ThanksSahithiExpected Behavior from the service |
Post config throws error on CFS node: Post config throws error on CFS node ; seeing below error enabling lsa on ncs.confadmin@ncs% load merge SR-multi-vendor-iosxr-netconf.xmladmin@ncs% load merge SR-plan-notification-settings.xmladmin@ncs% load merge SR-status-codes.xmladmin@ncs% commitAborted: 'status-codes core-function-pack L2VPN': Python cb_validate error. Following status codes of core function pack: L2VPN, are not set: 701, 702, 416root@nso-10-194-63-169:/etc/ncs# sudo /etc/init.d/ncs restart-with-package-reloadStopping ncs: Starting ncs: Bad configuration: /etc/ncs/ncs.conf:349: "large-scale" cannot occur here as a child to 'ncs-config'.Daemon died status=21admin@ncs% run show devices device rfs-2 netconf-notifications subscription FAILURENAME STATUS REASON ERROR INFO------------------------------------------------------------------------------------------------------------------rfs-cisco-custom-template-events running - -rfs-dispatch-map-update failed subscribe Failed to create subscription: bad_element: no such streamtsdn-kicker-events running - - | CFS node: 10.194.63.169 (ssh: nso/Vtsisco@123!)RFS node1: 10.194.63.170 (ssh access - same as CFS)RFS node2: 10.194.63.242 (ssh access - same as CFS)In RFS1 and RFS2, I observe that the ncs.conf, the event stream registered as -> <stream> <name>dispatch-map-events</name>The correct stream name is <stream> <name>dispatch-map-update< /name>Please check the LSA installation wiki.Please update the correct stream and restart the rfs nodes.Then re-subscribe.ThanksSahithilarge-scale tag is removed from ncs.conf from M2 release.ThanksCheck out the Wiki on Installation: Thanks Sahithi for the feedback - will fix the ncs.conf for CFS and RFSs |
How to recover from zombie state: Deleting a L3VPN service and failed into zombie state.checked the error handling wiki: still need to understand how to process "load-device-config on zombie ; commit? would you please provide sample NCS commands? | Hi Jason,Please follow clean up action steps from this wiki :Please let me know if you are able to do a clean zombies after following the steps in this wiki.Thanks,Priyanka Hi Jason,Please confirm if you still have any issues.If i don't hear anything by Friday EOD , will consider it closed.ThanksPriyankaHi Jason ,admin@nso-cluster2:~$ ncs_cmd -o -c "delete /zombies/service{/cisco-flat-L3vpn-fp-internal:flat-L3vpn[name='L3NM-L3NM-SRTE-5000-internal'][endpoint-name='CL3-P-TOPLEFT_5000']}"admin@nso-cluster2:~$ ncs_cli -u adminUser admin last logged in 2021-09-22T21:08:40.91434+00:00, to nso-cluster2, from 172.20.200.130 using netconf-sshadmin connected from 10.24.98.183 using ssh on nso-cluster2admin@ncs> show zombies % No entries found.admin@ncs> As we saw zombie was stuck in pending delete so finally we deleted it from ncs_cmd.ThanksPriyanka |
Ethernet-service-oam having Schedule option-- Editing a service to add a Y1731 profile config: NSO: 5.5.2.7TSDN 3.0.0 M6.,While editing a existing L2vpn service.under ethernet-service-oam , while associating a y1731 profile to it, we see option for schedule .May i know why do we need scheudle here ?Schedule is already getting configured in Y1731-profile already.. | Hi Arun,This is at two places one at global and one at service level.You can look at wiki :Global Profile ** l2vpn-ntw y-1731-profile Profile-Delay-1schedule interval 5schedule duration 3Service Definition **l2vpn-ntw vpn-services vpn-service L2VPN_NM-Static-PW-SR-TE-6000vpn-nodes vpn-node VXR-8kv1 VXR-8kv1vpn-network-accesses vpn-network-access 250ethernet-service-oam y-1731 nully-1731-profile Profile-Delay-1schedule interval 50schedule duration forever!ThanksPriyankai see a bug here. why do we need schedule option on the service level?only y1731-profile is good enough. do we have any specific requirement for this.?Hi Arun,As suggested above:Config governance team has decided this :ThanksPriyankadone. |
NSO 5.7 install issue with python: Setup is installed with nso 5.7 with python 3.8.3.After the install, python packages are down in nso cli package status. following error messagesetup details:10.195.165.225optimaqa/Ovatest123@cnctest/Ovatest123@ | Fixed this python issue. However facing another issue with CFS setup, admin@ncs% show cisco-sr-te-cfp:cfp-configurationsNo entries found.CFS : 10.195.165.226 - admin/Ovatest123@RFS : 10.195.165.225 - sameOne more question - Should rfs-node-onboarding.xml, rfs-netconf-notif-sub.xml should be edited with RFS name, ip credentials ? Hello Ramakrishnan,For the python issue, it seems like NSO was able to pick up the correct python version after 1-2 retries. With NSO 5.7, python 3.7+ is a base requirement.For cisco-sr-te-cfp:cfp-configurations, cfp-configurations is used for non-native NED support through dynamic device mapping. For XR CLI NED, no dynamic device mapping is required as it is the native NED. For XR NC NED and XE CLI NED, dynamic device mapping is required. For these non-native NEDs, you can load the dynamic device mapping using included multi-vendor bootstrap payloads (for example SR-TE FP : "SR-multi-vendor-iosxe-cli.xml" and "SR-multi-vendor-iosxr-netconf.xml"). You can find bootstrap instructions for LSA here : .For "rfs-node-onboarding.xml" you may simply load merge the payload as directed in bootstrap steps.Please let me know if you have any other questions or concerns.Thanks,Isaac KimHello Ramakrishnan,As it seems like the requested issue is resolved, I will be closing this ticket.If you have any additional questions or concerns, please feel free to reopen this ticket and leave a reply.Thanks,Isaac Kimit worked. thank you for the information. |
L3VPN service assurance not disabled when device is in unreachable: NSO version : 5.5.2.9CFP build version : M7Device type and version used for testing: 7.3.2.37iAdd clear steps to reproduce an issue 1) device went to unreachable state2) Now I started disabling monitoirng state of the services which are terminating on above mentioned unreachable device3) Disabled service assurance on all L2VPN service is successful, this is fine. But for L3 VPN when I tried to disable, it thrown me error: "Connection timed out "Tortin-Scale-148". Here what we are not understanding is, disable monitoirng on all l2vpn - evpn, p2p, plain and SR was successful, but for L3VPN, having one end as this device, we are not able to move to disable state.NSO logs this Jira ticket is being raised to get clearification if there is such behaviourI have not collected logs at that time, I need to know, while disabling the monitoring of L3VPN service alone, the NSO is trying to check whether device is connected or not ? | has retried these steps to disable the service assurance for L3 vpn service.Issue was un-reproducible.The devices were unreachable yet enable/disabling of service assurance was ok.Hi Sampath, please provide feedback by Friday 10/29/2021 otherwise we will close this ticket. Thank you.Hi, As of now this behaviour is not seen, I Agree. When I raised this ticket, My device was actually in upgrade state, From DLM we have not move the device to detach from CDG, nor manually moved state to down/Unmanaged, So when device actually being upgraded, I found this behaviour. If possible, we can park this ticket as watch item so that if any real maintenance on any device is going to happen in near by feature, we can test this.Hi Sampath, I think this ticket has expired. But if you encounter something similar again in the future, please feel free to open a ticket so we can take another look at it!Thanks,Isaac Kim |
Failed to deploy to L2NM services on XE devices: Tried to deploy L2NM EVPN services to XE devices and it failed with error message below.errors: reason: Python cb_validate error. STATUS_CODE: TSDN-L2VPN-408 REASON: Router NED not supported CATEGORY: user SEVERITY: ERROR Context [name = Router NED not supported: cisco-ios-cli-6.74:cisco-ios-cli-6.74, message = Missing dynamic device mapping state = {'Device': 'AA-ASR-1001x-2', 'Service': 'L2NM-l2nm-evpn-asr1k2-3-internal', 'Device NED ID': 'cisco-ios-cli-6.74:cisco-ios-cli-6.74'}] . paths: /cisco-flat-L2vpn-fp-internal-local-site:flat-L2vpn-internal-local-site-service{L2NM-l2nm-evpn-asr1k2-3-internal AA-ASR-1001x-2}. path: /cisco-flat-L2vpn-fp-internal-local-site:flat-L2vpn-internal-local-site-service{L2NM-l2nm-evpn-asr1k2-3-internal AA-ASR-1001x-2}The testbed environment is:NSO 5.5.2.3 with TSDN 3.0.0-M4XE device: ASR1001x with Cisco IOS XE Software, Version 17.05.01a | Hi Jason,Do you have dynamic mappings loaded? Can you double check if your output matches mine or at the very least there's a mapping for `cisco-ios-cli-6.74`?<div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"><pre>admin@ncs% show cisco-flat-L2vpn-fp:cfp-configurations dynamic-device-mapping cisco-ios-cli-6.74:cisco-ios-cli-6.74 { python-impl-class-name flat_l2vpn_multi_vendors.IosXE;}dynamic-device-mapping cisco-iosxr-nc-7.315:cisco-iosxr-nc-7.315 { python-impl-class-name flat_l2vpn_multi_vendors.NativeXR;}dynamic-device-mapping cisco-iosxr-nc-7.3:cisco-iosxr-nc-7.3 { python-impl-class-name flat_l2vpn_multi_vendors.NativeXR;}dynamic-device-mapping cisco-iosxr-nc-7.4:cisco-iosxr-nc-7.4 { python-impl-class-name flat_l2vpn_multi_vendors.NativeXR;}</pre></div></div>Thanks,RegiHi Regi,I don't see "dynamic-device-mapping cisco-ios-cli-6.74:cisco-ios-cli-6.74" on AA testbed output:admin@ncs% show cisco-flat-L2vpn-fp:cfp-configurationdynamic-device-mapping cisco-iosxr-nc-7.315:cisco-iosxr-nc-7.315 { python-impl-class-name flat_l2vpn_multi_vendors.NativeXR;}dynamic-device-mapping cisco-iosxr-nc-7.32:cisco-iosxr-nc-7.32 { python-impl-class-name flat_l2vpn_multi_vendors.NativeXR;}dynamic-device-mapping cisco-iosxr-nc-7.3:cisco-iosxr-nc-7.3 { python-impl-class-name flat_l2vpn_multi_vendors.NativeXR;}dynamic-device-mapping cisco-iosxr-nc-7.4:cisco-iosxr-nc-7.4 { python-impl-class-name flat_l2vpn_multi_vendors.NativeXR;}Our wiki seems to be the issue, the following section,was missing the following payload for the bootrap configuration for the dynamic mappings ``. I have updated the wiki and informed the submitter.Submitter came back with the following issue:<div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"><pre>message: External error in the NED implementation for device AA-ASR1001x-3: command: label 102 101 : label 102 101 ^ % Invalid input detected at '^' marker. [ interface pseudowire1001 / ]</pre></div></div>After checking on their device I found out the following label was not configurable:<div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"><pre>interface pseudowire1008 source template type pseudowire l2nm-p2p-asr1k2-3 encapsulation mpls neighbor 100.100.100.3 1008 control-word include label 108 109 <-------- not able to configureexit</pre></div></div>After investigating the issue, the pre-requisite MPLS label range was not configuring. After configuring it, the submitter was able to push the payload with no errors.<div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"><pre>mpls label range 1000 2000 static 100 500</pre></div></div>I should mentioned that EVPN for XE is not supported through T-SDN, as such the submitter had updated their payload to P2P as well. |
L2VPN & L3VPN service types can support on IOSXE devices: Please clarify what types of L2VPN and L3VPN services could be supported on IOSXE devices. | <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"><pre>L2VPN EVPN SR L2VPN EVPN Plain L2VPN P2P SR L2VPN P2P RSVP L2VPN P2P Plain L3VPN SR</pre></div></div>As far as IOSXE is concerned, only L2VPN EVPN is not supported. Please refer to device support matrix: |
Loading TSDN FP along with RON FP .: We have notification generator collision .Which one should we consume ? | Resolved via email. `Consuming RON M3 along with TSDN M6 solved the problems. You can close the tickets.` |
NSO LSA HA setup with L3 mode - Need topology and configs: For 6.0 testing, we need NSO LSA HA setup details with L3 network.Need topology, configs details on router, NSO LSA VMs as well. | Below diagram represents the NSO LSA in HA mode with L3 BGP. Depending on deployment, whether L2, L3 or DNS Redirection, the principle is the same that each NSO function (LSA node) is paired by a primary and a secondary.We have worked with you to bring up a standalone L3 HA setup, please replicate similar config of high-availability and hcc for each CFS, CFS pairs along with LSA config.Thanks Van for providing the topology diagram for NSO LSA L3 mode! We will try this and check.Regards,Gomathi.Hi Gomathi,If nothing else pending, I'd like to close this ticket. Please let me know by EOD Monday. 8/7.Regards,VanHi Van,Sure, I have setup my topology with CFS and RFS nodes. Configuring CSR router configs and NSO LSA configs. Once everything works fine, I will close this ticket.Thanks,Gomathi.Setup completed.Hi Van,I tried to setup NSO LSA with L3 HA mode. Performed below steps:1. Install NSO and configure ncs.conf on all LSA HA nodes.2. Deploy DLM and CFP full bundle from CW UI with HA settings.3. The above step installed packages successfully on all NSO LSA nodes.4. Updated the CFS and RFS nodes in 1st set of cluster nodes (Active CW) as per this wiki - 5. Most of the steps passed and shows output as expected. "run show cluster" shows rfs nodes as down. 6. Seems like CFS is not able to connect to RFS node via 2022 port.7. I re-entered all the steps on cfs to update auth group, credentials and configure rfs on CFS.8. Still CFS cluster shows rfs nodes as down.admin@ncs% run show cluster LAST RECEIVEDNAME NAME STATUS EVENT NOTIFICATIONS--------------------------------------------------nso-cw1-rfs1nso-cw1-rfs2 LOCAL REMOTEREMOTE NODE ADDRESS PORT CHANNELS USER USER STATUS TRACE-----------------------------------------------------------------------------nso-cw1-rfs2 192.168.5.32 2022 - admin admin down disablednso-cw1-rfs1 192.168.5.31 2022 - admin admin down disabledHence, need your help to check this issue. Once this is resolved there are 2 more steps (enable HA and set VIP to VMs) to complete the NSO LSA HA installation. Note: CSR router configs are all configured. |
Validation Error for Service Creation in NSO LSA ( 1 CFS and 2 RFS): Hi,NSO CFS: 10.56.116.121NSO RFS1: 10.56.116.122 ( rfs-1)NSO RFS2: 10.56.116.123(rfs-2)Credentials: nso/Public1234!CFS , RFSs installed by followed the wiki properly.we have devices onboard from CW by "user-defined".60 devices onboarded. ( 31 devices in RFS1, 29 devices in RFS2)but, while trying to create sr policy or services , it throws error . Attached snapshots.Payload attached. | Hi Arun,Looks like the dispatch-map was not in sync . i manually sync the dispatch-map and issue seems to be fixed.Please verify.Please refer the 'service deployment failure troubleshooting' steps<font color="#172b4d"> for more details :Regards,Nishad Subair |
Device Credential when changed on crosswork UI are not changed on the NSO : Problem:----------Crosswork/CNC cannot perform update or delete operations of devices/credentials on NSO.Issue:-------DLM has experienced failures or success with unexpected results when it is trying to do update or delete operations on the NSO devices via DLM interface. This is due to DLM is not exposed to the augmentation ; validation of the NSO yang models from the service models used by different function packs. Due to the NSO limitation, the only operation that DLM does since CNC 2.0 is onboard the missing and read-only operation of devices on NSO. We have a requirement to support update/delete operations of devices/credentials and hence looking for a way for DLM to know upfront if update ; delete operations go through via the other service pack annotations ; validations? | Hi Ravi,Curently NSO has a restconf API call to add/modified device auth-group. Its format like the below.curl --request PATCH 'http://<NSO IP>:<PORT>/restconf/data' --header 'Content-Type: application/yang-data+xml' --header 'Authorization: Basic YWRtaW64YWRtaW4=' --data-raw '<devices xmlns="http://tail-f.com/ns/ncs"> <authgroups> <group> <name>xr-auth</name> <default-map> <remote-name>admin</remote-name> <remote-password>Cisco#123</remote-password> <remote-secondary-password>Cisco#123</remote-secondary-password> </default-map> </group> </authgroups></devices>' It is up to CNC group to implement this API call to change device auth-group credentialThanks,LinhHi Linh,Thanks for your response. DLM uses MAAPI socket API to communicate with NSO. If the device auth-group that DLM want to be updated is having annotation ; validation by other service packs, there is a possibility that the update fails. Is there any way to know upfront if the operation can be successful or not? Thanks,RaviHi Ravi,You may need to consult to this link to get the informatonThanks,LinhClose since the issue is not on TSDN CFPRegards,Linh |
Unable to download CNC Functional Pack TSDN 6.0.0 Drop 5 (DTHO): Unable to download CNC Functional Pack TSDN 6.0.0 Drop 5 (DTHO) | I am able to download the tar with master so please close the case.Customer able to download and requested to close case. |
unable to clear the nso zombies: unable to clear the zombies in nso | Hello, can we please get access to this setup/NSO VM and issue reproduction steps.Thank you.Hello, we have not received further response on this ticket.Please let us know if this issue persists.Note: If no response is received ticket will be closed EOD Aug 2nd.Customer has confirmed that issue is resolved (no fix provided by TSDN). Ticket can be closed. |
Services migration from 5.7.3 m6 to 5.7.4 M7: Hi,We have Uni-d scale setup that was created with cw194/CNC202 build having NSO with 5.7.3 M6 Build.We have 10K services are getting monitored. we would like to know if there is any migration procedure or upgrade to 5.7.4 M7 with 10k services.Thanks,Arun.M.G | This is NSO LSA setup without HA. this is NSO LSA setup . one CFS and Two RFS.Hi Arun,Let's do Webex and go over it.Thanks,PriyankaWe have successfully upgraded both RFS nodes.There is follow-up meeting with Arun tomorrow to upgrade CFS too.ThanksPriyankaHi Priyanka,Thanks a lot for your help on migrating the NSO LSA setup from 5.7.3 M6 to 5.7.4 M7 with all 10k Services.This is really very helpful to get to know steps on migration. Though we are not supporting it officially, this has been a great workaround in scale env that reduces a lot of re-effort and time.We confirmed that the migration is successful and also the services modification/ provisioning worked fine for the devices in both RFSs.Attached file has steps that we followed for migration in both RFS and CFS for Migration.Thanks,Arun.M.G |
sareddyd-esxi-01-vm2: Device is locked in a commit operation by session 1272984: NSO: 5.7.3build: nso-5.7.3-tsdn-4.0.0-M6device: /auto/iox-ucs-077-san2/prod/7.7.1.15I.SIT_IMAGE/xrv9k/wslogs : 10.56.116.151://tmp/data_061022_065731_dirlogin: nso/Public1234!Steps:I was deleting EVPN services. Two services stuck with the "Device is locked in a commit operation by session" Number as mentioned above.Logs are present in the same NSO CFS : 10.56.116.151. Login provided.RFS1: 10.56.116.152RFS2: 10.56.116.153I request you to help in releasing the lock and continue to delete the services. I am not able to operate any action on the devices. Please help.Same log attached with this ticket | /etc/init.d/ncs stop and start solved the issue. But the two evpn services: l2nm-evpn-404,l2nm-evpn-405 are in zombies state. In device they got deleted from NSO before this issue raised. Please check ...Hi Sampath,I would suspect that the stuck zombies are due to the session lock and NSO stop start.I can try to help with unblocking your setup.You mentioned that the config was deleted from the network but still exists in NSO? In this case I would advise to try to execute L2NM cleanup action with no-networking true. Afterwards sync device and your setup should be ready to go.Let me know if this works.Thanks,Isaac KimHello Sampath, I believe we have been able to provide a suitable recovery option for your issue. If there is no reply in few days, I will close this ticket.Thanks,Isaac KimHey, Sorry Isaac, thought of talking to you on this. I have done deleting services in the morning as it has not resolved for a long time I have done stop and start in the evening, hence I am able to delete other service, I have just clean up zombie service. So We need to concentrate only on the issue where I am not able to delete the services because of locking. Let me know if we have investigated on that partHi Sampath, I ran across this ticket when cleaning up my support queue.Apologies, I seemed to have missed the reply for a long time. I will try to analyze the logs and see if there is any unexpected behavior so we can have some closure on this issue.Looking back on the log dump you posted to this ticket,we can see that DLM and/or CNC GUI is having trouble clearing lock on transaction handle 1 as seen below:<div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"><pre class="code-java">2-Jun-2022::04:55:23.983 <0.1337.0> <0.6529.272> ** Generic server <0.6529.272> terminating ** Last message in was {jsonrpc,,[{,1}]}** When Server state == {webui_sess,, <0.6529.272>,https,760286,0,0,[],</pre></div></div>Afterwards, we can see that this th 1 is invalid as it does not exist as seen below:<div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"><pre class="code-java">{, {error_report,<0.1337.0>, {<0.18671.272>,crash_report, [[{initial_call,{cs_trans,init,[]}}, {pid,<0.18671.272>}, {registered_name,[]}, {error_info, {exit, {bad_return_value, {error, {jsonrpc_error, {-32602,, }, [],[], {struct,[{param,}]}, }}},</pre></div></div>This indeed is something we have seen in rare cases in TSDN 4.0 where CNC UI's session transaction lock on NSO is not cleared as described in the ticket. Its strange that the transaction handler id is not found. It seems like due to NSO start/stop during an active commit may have caused CNC UI's state data to be out of sync with NSO. The logical assumption here is that start/stop cleared the transaction, but CNC UI is still attempting to clear what it thinks is an active transaction. In the logs, we can see the above two crash logs repeatedly. This is likely why CNC UI is in unresponsive state.Although changing NSO state during active commits/transactions is not recommended, the workaround would be to log out of CNC GUI (clear CNC state data), restart NSO (force close any locked sessions), and then continue.I hope this helps provide some clarity on this issue.If you face this issue again anytime in the future, please feel free to open a new ticket and reference this JSD for context.Thanks,Isaac KimAnalysis and workaround provided. |
Service Assurance Enable possible for Non-Deployed Services: Hi,NSO: TSDN 4.0.0 M5 with NCS 5.7.4 I have a payload as below where i dont have the services is not deployed at all.when i try to load the payload, it was successfull and created an service entry and Notified to CNC also.but, due to no device config exists, Service Assurance goes to error state.shouldnt we fail the commit the Service assurance part when we dont have the service deployed or when we dont have the respecitve service configs?<config xmlns="http://tail-f.com/ns/config/1.0"> <l2vpn-ntw xmlns="urn:ietf:params:xml:ns:yang:ietf-l2vpn-ntw"> <vpn-services><vpn-service> <vpn-id>test-l2vpn</vpn-id> <service-assurance xmlns="http://cisco.com/ns/nso/fp/examples/ietf-l2vpn-ntw-cisco-augmentations"> <monitoring-state>enable</monitoring-state> <profile-name>Gold_L2VPN_ConfigProfile system</profile-name> <!-<del>profile-name>Silver_L2VPN_ConfigProfile</profile-name</del>-> <rule-name>Rule-L2VPN-NM-P2P system</rule-name> </service-assurance> </vpn-service> </vpn-services> </l2vpn-ntw></config>Thanks,Arun.M.G | Hi ArunCan you check this with the latestTSDN 4.0.0 M8 build? It is failing as expected for us while committing.Thanks,NikhilClosing the ticket as haven't heard anything from the user |
NSO stuck with error - Configuration Namespace is locked by another agent: NSO: 10.195.165.75 - xrd1/xrd1Trying to push a payload but NSO stuck with below error message. message: External error in the NED implementation for device cluster-XRD6: config exclusive: config exclusiveCurrent Configuration Session Line User Date Lock 00000000-00000a46-00000000 LICENSE EVAL-END Thu Jun 9 07:07:04 2022 * Cannot enter exclusive mode. The Configuration Namespace is locked by another agent.Tried NSO restart with package and Ubuntu reload, but it didnt help. Kindly suggest | Hi Ajitha,The above error occurs when device is not cleaned up correctly. Can you login to your device and try to enter the config mode? You will observe similar error. This error is unrelated to NSO.ThanksSahithi Please checkout this ticket for steps on the device: Let me know if I can close the ticket.Hi Sahithi, Thanks for verifying, Checking the issue with XRD team. Seems to be license expired issue. Not an NSO CFP issue. It is a XRD issue. |
We are getting error while provisioning L2 vpn: We are getting following error while provisioning L2VPN in our lab.Validation failed : Unknown error (66): cisco-flat-L2vpn-fp-um-evpn-vpws-template.xml:21 will not apply template to device XRV-A with unsupported ned-id: cisco-iosxr-nc-7.7:cisco-iosxr-nc-7.7STATUS_CODE: TSDN-L2VPN-303REASON: Config push failedCATEGORY: deviceSEVERITY: ERRORContext [name = Configuration Error, message = Could not apply service config on local site state = {'Local Site': 'XRV-A', 'Service': 'L2NM-EVNP_TEST-internal'}] : /cisco-flat-L2vpn-fp-internal-local-site:flat-L2vpn-internal-local-site-service{L2NM-EVNP_TEST-internal XRV-A}Our NSO version : 5.3.2.1We have proper configuration:admin@ncs% show cisco-flat-L2vpn-fp:cfp-configurations dynamic-device-mapping cisco-ios-cli-6.77:cisco-ios-cli-6.77 { python-impl-class-name flat_l2vpn_multi_vendors.IosXE;}dynamic-device-mapping cisco-iosxr-nc-7.315:cisco-iosxr-nc-7.315 { python-impl-class-name flat_l2vpn_multi_vendors.NativeXR;}dynamic-device-mapping cisco-iosxr-nc-7.3:cisco-iosxr-nc-7.3 { python-impl-class-name flat_l2vpn_multi_vendors.NativeXR;}dynamic-device-mapping cisco-iosxr-nc-7.4:cisco-iosxr-nc-7.4 { python-impl-class-name flat_l2vpn_multi_vendors.NativeXR;}dynamic-device-mapping cisco-iosxr-nc-7.5:cisco-iosxr-nc-7.5 { python-impl-class-name flat_l2vpn_multi_vendors.UMXR;}dynamic-device-mapping cisco-iosxr-nc-7.6:cisco-iosxr-nc-7.6 { python-impl-class-name flat_l2vpn_multi_vendors.UMXR;}dynamic-device-mapping cisco-iosxr-nc-7.7:cisco-iosxr-nc-7.7 { python-impl-class-name flat_l2vpn_multi_vendors.UMXR;} | Hi SivaPlease provide NSO setup details and the service payload that has the issue. I need to check the setup to debug it.Thanks,NikhilHi Siva/SagarAs discussed over the call, the nso is not setup properly in your system. Please do a re-install with correct version of NSO - 5.7.3 for M6. NC NED 7.7 is not supported for releases before M6.Let us know if the issues is resolved and we are good to close the ticket.the issue was with nso not setup properly in the system. Suggested to do a re-install with correct version of NSO - 5.7.3 for M6. |
NSO LSA 0 CFS -- Netconf-notificaion subsercription throws error: NSO CFS: 10.56.116.121 NSO RFS1: 10.56.116.122NSO RFS2: 10.56.116.123 Credentials: nso/Public1234!Installed as per the wiki., After installing Tm-TC, it has error for Netconf-notificaion subsercription for rfs-dispatch-map-events.as per the CFS wiki., we have set devices device rfs-1 netconf-notifications subscription rfs-dispatch-map-update stream dispatch-map-update local-user admin store-in-cdb trueas per tm-tc wiki, we have set devices device rfs-1 netconf-notifications subscription rfs-dispatch-map-events stream dispatch-map-events local-user nso store-in-cdb trueafter this, below error.admin@ncs% run show devices device rfs-2 netconf-notifications subscription FAILURENAME STATUS REASON ERROR INFO------------------------------------------------------------------------------------------------------------------rfs-cisco-custom-template-events running - -rfs-dispatch-map-events failed subscribe Failed to create subscription: bad_element: no such streamrfs-dispatch-map-update running - -rfs-kicker-events running - -admin@ncs% run show devices device rfs-1 netconf-notifications subscription FAILURENAME STATUS REASON ERROR INFO------------------------------------------------------------------------------------------------------------------rfs-cisco-custom-template-events running - -rfs-dispatch-map-events failed subscribe Failed to create subscription: bad_element: no such streamrfs-dispatch-map-update running - -rfs-kicker-events running - -admin@ncs%should we have both dispatch-map-update and dispatch-map-events?Thanks,Arun.M.G | Hi Arun,As mentioned in the CFS wiki , we need to configure only 'dispatch-map-update' stream.Thanks,Nishad Subair.thanks Nishad for the Info. |
RFS sync is not happening from CFS - reporting illegal reference device sync for TM-TC Packages: Hi Team, In our whole SH team where everyone are facing issue with the RFS sync from CFS and we all hit illegal reference for TM TC package, due to this we couldn't able to proceed configuring any of the policies or services, it it a blocker issue for the entire team. NCS and TSDN download links are for reference: NCS: CNC FP: NCS CLI output for reference: nso1@ncs> request devices sync-from sync-result { device rfs-1 result false info illegal reference devices device rfs-1 config cisco-tm-tc-fp:tm-tc Crosswork-R10-CE node R10-CE name}nso1@ncs> | Testbed is still in the same state. CFS: 172.20.107.111 nso1/Cnc@1234RFS: 172.20.107.103 nso/Cnc@1234Kindly consider this issue as it is blocking us to provision any of the services. has same issue where we unblock the system by deleting tm-tc instance.RCA: 1. Devices onboarded though rfs1/rfs2 need to sync-from from cfs to update the dispatch map data structure2. If cfs sync-from performed after provisioning services/plan in rfs node, cfs node will reportillegal reference as pre-requisite step missed.Hi Satya,this ticket cannot be moved to Verified until we have right closure on the issue and right behavior on expected. anyways, Thanks for the workaround by Removing the illegal reference from rfs node. Actually, this seems to be an issue in 6.0 where we havent faced this issue in 5.0. Sync-from for devices is successull in RFS , but Sync-from in CFS itself is failing with the same error message.Even after successful sync of dispatch map in CFS, we dont see sync-from is successful in CFS. This seems to be an issue .We face this issue in almost all our 6.0 setups.Note: we dont have any services configured in RFS . It is fresh server. we have alrady opened JIra ticket opened and it was also closed with workaround. Hi Krishna,please reopen this ticket. if not possible, raise another jira and make we are not closing or moving it to V state. Workaround is not accepted as resolution. raise CDETS on tm-tc if required. Thanks,Arun.M.Gcreated which is having the fix related toncs-6.1.2-cisco-tm-tc-fp-cfs-ned-1.0.tar.gz ;ncs-6.1.2-cfs-cisco-tm-tc-fp-6.0.0-35.tar.gz in CFS node |
L3 Slice Creation Issue: Hi,While creating L3 slice, i'm getting the below issue,[{"Message":"/resource-pools/idalloc:id-pool{color-pool}/range/start is not configured"}]Attached the payload.Setup:uid: nsopwd: Vtsisco@123! | Please create the pool ---> color-poolOn NSO, I see as shown in above screenshot.The color-pool is missing hence you are seeing error.Regards,Janaki Duggiralasample bootstrap:<config xmlns="http://tail-f.com/ns/config/1.0"> <resource-pools xmlns="http://tail-f.com/pkg/resource-allocator"> <id-pool xmlns="http://tail-f.com/pkg/id-allocator"> <name>color-pool</name> <range> <start>1</start> <end>4294967295</end> </range> </id-pool> </resource-pools></config> Hello, we have not received further response on this ticket.Please let us know if this issue persists.Note: If no response is received ticket will be closed EOD Aug 2nd.Hi Janaki,Thanks for the responseKindly close the ticket.Thanks,NaveenCustomer has confirmed issue is resolved. Closing ticket. |
CNC NSO notifications issue requires fullResync: NSO notifications processing seems to fail time to time which causes CNC UI to display incorrect/not refreshed information about L3VPN instances and other services.Performing a https://<tt>cnc.vip</tt>:<tt>cnc.port</tt>/crosswork/cat/nso-connector/v1/api/fullResync API call seems to solve these issues.Reproduction the issue is difficult as it is not found to be deterministic.Ask is also to understand if there is a way to detect this issue with the NSO notifications somewhere and if there is an alternate way to perform the full re-sync other than through API.In attachment there is an email thread where it was identified the issue and discussion with CAT visualisation Dev team. | Hi Fabio,I installed TSDN 2.0 and pull l3vpn-route-policy. The restconf call works fine. Please see the attached screen shotIt seems the issue is on CATUI side. Please contact their team for further support.Thanks,LinhHi Linh, thanks, I'll contect the CAT UI team then.Best Regards,Fbio GameiroSince this is not on CFP side, I am going to close this ticket.Regards,Linh |
Service Sync-to and Sync-from actions: Trying to understand when and how to use service Sync-to and Sync-from actionScenario:1. Make headend router OOS (out of sync)2. Provision a policy with async commit queue The provision in provisioning failed state as expected3. Do a device sync-to or sync-from to make device in sync4. Do a service "Sync to"The policy is still in provisioning failed state. Is this expected? service sync-from has the same result.Can you clarify what is the use case for service/policy sync-to and sync-from actions? | Hi Crystal,Please follow the error recovery procedures: The issue that you're facing is expected, since the T-SDN service was not re-deployedHi Reginald,I am a bit confused. I tried the similar steps as the original one except put the device OOS, I made it unreachable. The service Sync-from or Sync-to make the service becomes provisioning successful state.In both scenarios are error-recovery with different sync directions. No re-deploy were done for either one. But with different result. Can you let me know why the results are different for these 2 scenarios? Are both expected results?New Scenario:1. Make headend node reachable2. Provision a policy with async commit queueThe provision in provisioning failed state as expected3. Make headend node reachable4. Do a service "Sync from" or "Sync to"First let me confirm, I maybe missing something important but what do you mean by "service sync-from/sync-to" are you referring to the error-recovery action itself and passing sync-from/sync-to direction? Also for the new scenario, you made "headend node reachable" but you didn't do this for the first scenario? If you outline the reproduction steps with the payloads/actions in question I can probably give you a clearer answerSubmitter was trying to use error-recovery action for persistent failure for the first scenario which is not supported. Using the flow outlined in our we were able to resolve the issue.We did find a bug with the error-recovery action throwing an error, but also returning a "Recovery Complete" which is probably what was confusing. The submitter will file a CDETs for this. |
Failed to deploy L2VPN P2P SR service on XE devices: Tried to deploy the L2VPN P2P with SR policy on XE devices, it failed with following error message, any idea what was wrong? The similar service was able to deploy on XR without issue.message: External error in the NED implementation for device AA-ASR-1001x-2: command: preferred-path segment-routing traffic-eng policy sr-mpls-policy-asr-1-2 : preferred-path segment-routing traffic-eng policy sr-mpls-policy-asr-1-2 ^ % Invalid input detected at '^' marker. [ template type pseudowire l2nm-p2p-sr-asr / ] | ASR1001X is not in our supported device list. It could be this device does not support preferred path.Please refer to device support list : NSO:5.5.2.3CFP: TSDN3.0.0-M4 XE device: ASR1001x with Cisco IOS XE Software, Version 17.05.01aReporter replaced ASR1001x with ASR903, and re-deployed the L2VPN P2P plain service, the service deployment succeeded |
CNC is receiving continues service update from the NSO and causes the service state to flap continuously: service : L3VPN : The user edited the service and the device was out of sync. The commit failed but the NSO sending service updates continuously to CNC. Please ping me on webex teams to get the testbed access.admin@ncs# *** ALARM service-activation-failure: Network Element Driver: device TSDN-PE-2: out of syncadmin@ncs# *** ALARM service-activation-failure: Network Element Driver: device TSDN-PE-2: out of syncadmin@ncs# *** ALARM service-activation-failure: Network Element Driver: device TSDN-PE-2: out of syncadmin@ncs# *** ALARM service-activation-failure: Network Element Driver: device TSDN-PE-2: out of syncadmin@ncs# *** ALARM service-activation-failure: Network Element Driver: device TSDN-PE-2: out of syncadmin@ncs# | service name TSDN-PE-2-5001 <INFO> 14-Sep-2021::03:21:34.504 cisco-flat-L3vpn-fp ncs-dp-1389714-cisco-flat-L3vpn-fp-internal:main-5-th-2173880: - EXIT_POINT:, servicepoint: flat-L3vpn, service: L3NM-TSDN-PE-2-5001-internal TSDN-PE-2_2, component: ncs:self self, state: ready, nano op: cb_nano_create<INFO> 14-Sep-2021::03:21:34.539 cisco-flat-L3vpn-fp ncs-dp-1389714-cisco-flat-L3vpn-fp-internal:main-5-th-2173880: - ENTRY_POINT: servicepoint: flat-L3vpn, cb: cb_post_modification, kp: /cisco-flat-L3vpn-fp-internal:flat-L3vpn{L3NM-TSDN-PE-2-5001-internal TSDN-PE-2_2}, service op: SERVICE_UPDATE<INFO> 14-Sep-2021::03:21:34.539 cisco-flat-L3vpn-fp ncs-dp-1389714-cisco-flat-L3vpn-fp-internal:main-5-th-2173880: - EXIT_POINT: servicepoint: flat-L3vpn, cb: cb_post_modification, kp: /cisco-flat-L3vpn-fp-internal:flat-L3vpn{L3NM-TSDN-PE-2-5001-internal TSDN-PE-2_2}, service op: SERVICE_UPDATE<INFO> 14-Sep-2021::03:21:34.572 cisco-flat-L3vpn-fp flat_L3vpn_validation-5-th-2173880: - ENTRY_POINT: validate_cb: flat-L3vpn-validation, kp: /cisco-flat-L3vpn-fp:flat-L3vpn{L3NM-TSDN-PE-2-5001-internal}<INFO> 14-Sep-2021::03:21:34.591 cisco-flat-L3vpn-fp flat_L3vpn_validation-5-th-2173880: - EXIT_POINT: validate_cb: flat-L3vpn-validation, kp: /cisco-flat-L3vpn-fp:flat-L3vpn{L3NM-TSDN-PE-2-5001-internal}<INFO> 14-Sep-2021::03:21:34.593 cisco-flat-L3vpn-fp flat_L3vpn_internal_validation-5-th-2173880: - ENTRY_POINT: validate_cb: flat-L3vpn-internal-validation, kp: /cisco-flat-L3vpn-fp-internal:flat-L3vpn{L3NM-TSDN-PE-2-5001-internal TSDN-PE-2_2}<INFO> 14-Sep-2021::03:21:34.608 cisco-flat-L3vpn-fp flat_L3vpn_internal_validation-5-th-2173880: - EXIT_POINT: validate_cb: flat-L3vpn-internal-validation, kp: /cisco-flat-L3vpn-fp-internal:flat-L3vpn{L3NM-TSDN-PE-2-5001-internal TSDN-PE-2_2}<INFO> 14-Sep-2021::03:21:36.754 cisco-flat-L3vpn-fp ncs-dp-1389714-cisco-flat-L3vpn-fp:main-8-usid-18877-l3vpn-internal-plan-change-handler: - Internal plan kicker changed for: flat-l3vpn-internal-plan-kicker /cisco-flat-L3vpn-fp-internal:flat-L3vpn-internal/flat-L3vpn-plan{L3NM-TSDN-PE-2-5001-internal TSDN-PE-2_2} 2173946<INFO> 14-Sep-2021::03:21:36.936 cisco-flat-L3vpn-fp ncs-dp-1389714-cisco-flat-L3vpn-fp:main-8-usid-18877-l3vpn-internal-plan-change-handler: - Attempting to re-deploy zombie: /cisco-flat-L3vpn-fp:flat-L3vpn<INFO> 14-Sep-2021::03:21:36.939 cisco-flat-L3vpn-fp ncs-dp-1389714-cisco-flat-L3vpn-fp:main-8-usid-18877-l3vpn-internal-plan-change-handler: - Attempting to re-deploy service: /cisco-flat-L3vpn-fp:flat-L3vpn{L3NM-TSDN-PE-2-5001-internal}<INFO> 14-Sep-2021::03:21:36.972 cisco-flat-L3vpn-fp ncs-dp-1389714-cisco-flat-L3vpn-fp:main-8-usid-18877-l3vpn-internal-plan-change-handler: - Service re-deploy done for: /cisco-flat-L3vpn-fp:flat-L3vpn{L3NM-TSDN-PE-2-5001-internal}<INFO> 14-Sep-2021::03:21:36.983 cisco-flat-L3vpn-fp ncs-dp-1389714-cisco-flat-L3vpn-fp:main-8-usid-18877-l3vpn-internal-plan-change-handler: - Internal plan change handled for: /cisco-flat-L3vpn-fp-internal:flat-L3vpn-internal/flat-L3vpn-plan{L3NM-TSDN-PE-2-5001-internal TSDN-PE-2_2}<INFO> 14-Sep-2021::03:21:37.546 cisco-flat-L3vpn-fp flat_L3vpn_validation-5-th-2173957: - ENTRY_POINT: validate_cb: flat-L3vpn-validation, kp: /cisco-flat-L3vpn-fp:flat-L3vpn{L3NM-TSDN-PE-2-5001-internal}<INFO> 14-Sep-2021::03:21:37.571 cisco-flat-L3vpn-fp flat_L3vpn_validation-5-th-2173957: - EXIT_POINT: validate_cb: flat-L3Closing this ticket as we are tracking viaCSCvz66023.CSCvz66023 is open for tracking this issue. |
NSO CFP M2 - files changed without revision from git diff: id-allocator-oper.yangipaddress-allocator-oper.yangresource-allocator.yangfrom resource-manager/src/yang folder, files changed from M2 milestone build without any revision. Please see the attached snapshot. | Noted: Created a ticket to resolve this. |
Reconciliation of L2VPN services with CNC/NSO fails with RESTCONF: 1) I create a L2VPN on the routers CLI. These VPN adjusts to the l2vpn-ntw service package included in CNC 2.0.2) I do a sync-from in NSO3) I create the l2vpn service via RESTCONF API, using commit no-networking4) The l2vpn service shows up in CNC as "Success". Everything is normal now.5) If I use NSO's CLI to perform a "l2vpn-ntw ... re-deploy reconcile", everything works fine. But if instead of using NSO's CLI, I perform a RESTCONF API call using "http://<tt>nso_ip</tt>:8080/restconf/data/.../vpn-service=.../re-deploy?reconcile", then the services show up in CNC as "Failed". I attach a screenshot.Besides, NSO's check-sync in the affected devices turns negative, although a "compare-config" shows no differences between CDB and device config. Attached screenshot.These tests have been run on the dCloud demo: cisco_crosswork_network_controller_2.0_with_cdg_2.0_and_nso_5.4.2-1The same behavior is not observed when repeating the same process for l3vpn-ntw services.Thank you. | Hi Oscar,Is there any NSO logs that you can provide from when you encountered this failure? This may help with debugging.Alternatively, if the NSO is still running with the failed service intact, I can check few things on your NSO if you can provide credentials (will not make any state changes).Thanks,Isaac KimHello Isaac, unfortunately I don't have the logs as this was executed in a lab and it has been reconfigured for other purposes. Once it's available for me again, I'll repeat the exercise and let you know. Thanks.Hello Oscar,It has been a month since this ticket has been awaiting feedback. In this case, if you encounter reproduction and obtain logs.If there is no response in a week, I will assume that this case is closed.Thanks,Isaac KimHello, you can proceed to close the ticket, please.Regards,Oscar. |
Y1731 configuarion query for the existing l2vpn service: Hi,1] I already have a l2vpn Service configured and it is working fine. Payload attached L2VPN_NM-P2P-RSVPTE-202.xml2] Now, i am modifying the service with adding y1731 config to it. attached payload L2VPN_NM-P2P-RSVPTE-202-y1731-m6.xmlAfer step 2, i am seeing that it is deleting the existing config and adding new config to the devices with y1731.why do we need to delete the existing service conifg and readd?"show configuration commit changes last 2"atatched output from device cli. | Hi Arun,Looks like this bug :ThanksPriyanka |
Unable to create L2 service type in NSO M8: Hi,Unable to create L2 service type in NSO M8.Getting below error:errors: reason: is not configured. paths: /ralloc:resource-pools/idalloc:id-pool{None}/range/start. path: /ralloc:resource-pools/idalloc:id-pool{None}/range/startSetup:172.27.226.127Credentials: nso/Vtsisco@123!Regards,Magesh | Hi Magesh,Could you please provide the payload used, so we can reproduce this error.Hi Babu,Attaching the payload. Hi Magesh,The NSS Functionality was removed from TSDN in M8. So that payload will not be compatible. Also it looks like the NSO build present on your machine is outdated for M8. You can find the supported build here in the release notes.Hi Babu,But i was able to create L3 type slices in the same setup.It should have thrown error if NSS was not supported.Hi Babu,I think this is custom package for Transport Slicing on top of M8.So adding Regi to the ticket.Thanks,NaveenHi Naveen,Seems like you're missing L2 resource pools. You can configure them via the following:<div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"><pre><config xmlns="http://tail-f.com/ns/config/1.0"> <l2vpn-ntw xmlns="urn:ietf:params:xml:ns:yang:ietf-l2vpn-ntw"> <id-pools xmlns="http://cisco.com/ns/nso/fp/examples/cisco-l2vpn-ntw"> <evi-id-pool-name>evi-id-pool</evi-id-pool-name> <evi-source-target-pool-name>evi-source-target-pool</evi-source-target-pool-name> <rt-pool-name>rt-id-pool</rt-pool-name> </id-pools> </l2vpn-ntw></config></pre></div></div>There is already a ticket being tracked for this issue to add appropriate error messages.Thanks,RegiHi Regi,Thanks , I am not getting ID-pool error after adding the id-pools.But I am getting below error while trying to push the configs.interface GigabitEthernet0/0/0/2.311rn!!% Invalid argument: VPLS Bridge domains not supported on this platformrnBut looks like there is specific SMU has to be installed on top of XR7.8.1.Let me give a try.Thanks,NaveenYes, that's correct XRv9k does not support L2 EVPN MP because it doesn't support bridge domains.You'll need to use ASR9K or some other platform which does support it.Thanks,Regi |
L2VPN services stuck in-progress state after provisioning from NSO UI: L2VPN services stuck in-progress state after provisioning from NSO UISteps:1. Load all types of services one by one from NSO UI - either copy paste the payload content or attach the file.2. We can see few services will be provisioned successfully and couple of them will be stuck in-progress state during provisioning.Attached nso logs for reference. | One more step part of the below issue:*2. Enable or edit already provisioned services one by one. And then provision services from NSO UI one by one.Hi Gomathi,As discussed, there was a clean-up done for the services directly(without any delete) and then the same services were created again which resulted in some race condition and hence the services were stuck in in-progress state.We don't encourage the user to do clean-up of the services directly.It should be invokedwhen certain elements of the service fail to delete during normal service deletion. This can happen in scenarios where device not reachable during deletion.As a workaround for the services stuck in in-progress state, please execute the below steps on cfs node:Do a rfs-node sync-from on the cfs node ----> admin@cfs-node% request devices device rfs-node-1 sync-fromre-deploy the service on cfs node ----> admin@cfs-node% request l2vpn-ntw vpn-services vpn-service EVPN-MP-ELAN re-deployThanks,NikhilAdvised the user to delete the services instead of directly doing a clean-up. Also, gave a workaround for the stuck services |
devices from RFS3 not available in CFS: We have HA scale setup with two CFS and 3 RFS. There are 25K devices spread between 3 RFS. However, devices from RFS-3 are not visible in CFS.HA VIP 172.20.63.132CFS1 172.20.63.133CFS2 172.20.63.134RFS1 172.20.63.148RFS2 172.20.63.149RFS3 172.20.63.150username: cnctest/Ovatest123@ | Hi Ramakrishnan,I am able to see the devices from RFS-3 in CFS.Please suggest .Thanks,Nishad SubairHi Nishad,Updating the case as Ramakrishnan is offline.Actually when tried to create a policy in CFS using script, it's erroring out when Head-End/End-Point Devices are belonging to RFS3 as shown in below output below.Similarly from CNC Provisioning UI, when tried to provision any policy/service - the drop-down for HE/EP or VPN nodes does'nt show devices belonging to RFS3 though it's added and reachable to CNC. Please see the below snaps - Devices from RFS3 are missing in dispatch map. Nodes - Node_100_4, Node_100_10, Node_100_11 are from RFS3.I see that device from RFS3 are now available. Can you share the changes that you did in this setup.Hi Ramakrishnan,No changes were done in the set up . i was able to see the mentioned devices in the dispatch map.Please suggest if i can close the ticket.Thanks,Nishad SubairGo ahead and close the case. Sorry for the delay |
Not able to access yang variables for core-fp-common from NSO cli: I have downloaded the packages of TSDN from software.cisco.com and did packages reload in our environment. We are using NSO version 5.7.1.1. We have done below activities but not able to access core-fp-common from NSO cli.1. We have done the package reload and saw the load-dir, I can find the .fxs files but core yang is not able to access. Below is the screenshot for the same.2. I have tried to build the package core-fp-common but getting below result. It is strange for me. This activity I have done on my local machine which is having NSO version 5.7.1 and have local install but above activities I performed on system install. | Hi Faisal,Can we do webex ?I need to go over these steps over webex with you ?I want to understand whether you want to use local install or system install. Let's sync up and i can help you with resolution.ThanksPriyankaHi Priyanka,Yes, sure. Can we do it today. Please let me know when can we do that?Regards,FaisalHi Faisal ,I work in PST , please let me suggest a time that can work for both us.you time 9:30 PM IST works for you ?ThanksPriyankaHi Priyanka,I saw your message late night by 11 PM. Can we do it today 9:30 PM IST?Please setup a call by 9:30 PM IST. I will be available today.Regards,FaisalHi Faisal ,I accepted your invite for 9:30 PM IST , but I waited in the meeting for quite some time and left.As you were not available today, let's do the same time tomorrow if that works.ThanksPriyankaHi Faisal ,I am closing as per your confirmation.Please let me know if you need any further help.ThanksPriyanka |
provisioning L3VPN slow in scale setup: Our NSO scale setup has 25K devices, 200K VPN services.Compared to earlier NSO 5.4.x, current NSO 6.0 is slow in provisioning L3VPN services.In this release, we are trying L3VPN with 5 end points.L3VPN 5 EP - 1 service - 240 to 250 sec (without actually sending configs to routers -no-networking - true)L3VPN (point-to-point) - between 2 devices - 1 service - 104 to 110 sec. Is there any bug on L3VPN? Can we speed this up in setup?Question 2: The setup has only 3 months license. This setup needs to be tested over 5 months. Can we extend the license? Can you share the steps.Setup: 172.20.63.133Creds: cnctest/Ovatest | Hi ,Cannot login in your NSO .Could you please share correct working credentials ?Also can you please share usecase and comparison for NSO 5.4 vs NSO 6.0 test you executed .Please update steps of testing used came in both ?Types of devices and ned types same ?I see in one comparison no networking true - In this case you see different numbers in NSO 5.4 vs NSO 6.0 .Please share details and set up .I will look into .ThanksPriyankaWe use REST API through shell script and create L3VPN service through payloads.Device types are XR device ( simulated ) Major difference is for L3VPN, we were only using 2 end points in NSO 5.4.x. For NSO 6.0, we create L3VPN with 5 end points and also L3VPN with 2 end point. In both NSO cases, we were only executing no-networking is true.Other difference is updated payload between new NSO 6.0 and old NSO 5.4.x.In terms of memory, CPU, Swap, everything is same. It is the same machine.Setup details 172.20.63.133UI, CLI - cnctest/Ovatest123@ Hi Rama,First question comparing 2 EP (NSO 5.4.x ) vs 5 EP (NSO 6.0 ) is not apple to apple comparison.Please share a comparison for 2 EP with the same payload for both use cases.As there is no point in checking where orange is compared to apple.Usecase makes lots of difference in results.I hope you got my point.ThanksPriyankaAgree for 5 EP and 2 EP difference. Even comparing 2 end point L3VPN between NSO 6.0 and NSO 5.4.x also yields the same result, Time is higher in NSO 6.0. Please clarify on the license part also.Hi Rama ,Can you be specific what is the exact difference you see between NSO 5.4.x and NSO 6.0 2 EP in terms of commit times and also share exact outputs along with plan for both usecase.I am really not sure what licensing you are asking us ?We as CFP team not responsible for any license.ThanksPriyankaProvided setup details for both NSO and the scripts that are used along with payload. Kindly check why NSO 6.0 is taking long time for L3VPN. Hi Rama,I compared both your labs .Here is my observation :1. All RFS nodes on NSO 5.7 x -> has 26 packages 2.All RFS nodes on NSO 6.0 > has 31 packages .Here the packages which are causing degradation of performance are NETCONF NED packages.Please remove all netconf ned packages from both nodes .Then only we can say these set ups are comparable .right now NSO 6.0 has more number of netconf neds.Here is the platform ticket with reference :Here is my recommendation :for faster performance :1.Remove all NETCONF neds from RFS ndoes .2.Make sure to keep same number of packages on both set up if you are comparing them .We should keep identical labs if we are comparing them .Also please close CDET which you have raised :As without confirmation no point in keeping duplicate tracking of issues.ThanksPriyankaDevices in both setup use only cli ned. How can the unused netconf packages can cause performance issues?NSO 5.7.5.1devices use 'device-type cli ned-id cisco-iosxr-cli-7.39'NSO 6.0 device use 'device-type cli ned-id cisco-iosxr-cli-7.43'We used only the default packages available in TSDN FP for both NSO version. Removed netconf related packages in NSO. But there are no changes in time taken for provisioning.Hi Ramakrishnan,Priyanka mentioned the RT(Ticket on NSO platform) in the previous reply on degradation of performance due to netconf NED presence. Regarding "how", We believe NSO Platform folks are more well versed with the technical implementation here. You should be able to file RT with help of Crystal and find out the how.You mentioned that you removed netconf packages and tested. Please provide the following details to proceed further. Can you please let us know the exact time degradation?What is time to provision in 5.7.x vs time to provision in 6.0 ? Please give us the times. Along with "show packages package oper-status up" on the system.Also are you using 5.4 or 5.7 ? In the description I see 5.4 while you mention in the reply to be 5.7.We do not see any kind of degradation from our testing. So, we would need as much info specific as possible. If it would be easier to webex, let us know that as well.Hi Rama,I tried to create service using your payload in your labs and it ended up in ned error in NSO 5.7.5.1 lab and rpc error in NSO 6.0 lab.Can we do webex and get closure as no point of back and forth .ThanksPriyankaHi Team,I just tried now and able to provision L3VPN services in both NSO's. Will attach the snaps to show exact times for both including "show packages package oper-status up" o/p. In NSO 6.0, we see a huge delay of 10 to 20 times than earlier releases. We have one more standalone NSO 6.1, there also same issue seen. This delay is seen ONLY for L3VPN and not for L2VPN. Every release from the time* NSO is introduced into COE/CNC, am doing the scale testing using these scripts for scaling up 200K ; 300K VPN services. And never hit this huge delay for L3VPN.Thanks for sharing Platform tickets but these are not transparent to every one like CDETS. As we don't have access to platform tickets, neither description nor the content are visible to us.Please refer the attached notepad "repro-CSCwe21126" for exact times and show package o/p along with NSO logins.Will be available online for 1 more hour before i sign-off for the day. Please ping me, if required to repro again. "ramasrin" will be available tomorrow morning IST to provide more info. if needed for you to help us triage this at earliest.Thanks for your support !!-Sreedhar. Hi Sreedhar ,Please join :ThanksPriyankaHi Sreedhar,As per our webex discussion :Please try 50 service creation using multi-thread on NSO 6.x - Please share a comparison of 50 service (single thread ) vs multi thread (using 5 thread .each thread 10 services without device conflicts between threads.)Let's first focus on standalone for this.'ThanksPriyankaHi Priyanka, Our script has the payload for 10 services each with same device. We can modify it so that all devices used are different. When you mean by multi threading, should we run 5 instance of that script simultaneously from 5 terminals? By single thread, do you mean single terminal executing 10 service one by one? Or when i refer the following link, it talks about changes in package-meta-data.xml. Should we do anything of that sort ? I will try to sync up with you. with regards,Srinivasan RHi Rama,As per our webex discussion - Try the way i suggested distributing devices across services on different threads.Please try that and let me know how it goes.ThanksPriyankal3vpn.xml is the payload that caused lot of delay. scale-l3nm_fast is the one that NSO team gave, that is faster. We observed that vpn-instance-profile created for each service is not referenced in the vpn-node in scale-l3nm_fast.xml. If we refer the vpn-instance-profile under vpn-node, it takes around 25 seconds for each service. If we dont refer it, it takes just 1 second to create the service.Hi Team, With the modified payload (removed the reference of vpn-instance-profile under vpn-node), provisioning of L3VPN is very fast. Able to provision 10 L3VPN services in less than 20 seconds currently. Unblocked with the issue with this workaround. Thanks,Sreedhar.changing payload helped team to unblocked. |
Information needed on Re-activate Re-deploy and Re-Deploy Re-Concile: Hi Team,I am looking for more details on Re-activate Re-deploy and Re-Deploy Re-Concile to perform testing. Can some help with steps how can I execute these?Thanks,Kanchan | Hi Kanchan,As re-deploy is a native NSO action, you can find more information from NSO docs.For re-deploy, you can check nso_user_guide-6.1_XXXX.pdf page 203 for all details. You can find this pdf in your NSO install directory under <NSO_ROOT>/doc/pdf/nso_user_guide-6.1_XXXX.pdfThanks,Isaac Kim Isaac to provide testing scenario for redeploy reconcile via REST on Monday~Hi Kanchan, I will try to provide the testing procedure today.Thanks,Isaac KimHi Kanchan, I have finished the reproduction + REST procedure for redeploy reconcile.I have attached a postman collection to this reply with the requests in order.The procedure follows the requests in order:<ol>CREATE SERVICE<ol>Basic ODN serviceReplace head-end with your own device in request body</ol>QUERY DEVICE CONFIG BACKPOINTER EXISTS<ol>Query NSO device config with service meta-data and verify refcounter=1 and backpointer existsIn request URI, replace device selector with your own device</ol>DELETE DEVICE CONFIG COLOR<ol>Deletes NSO device config color with no-networking (only deleted from NSO CDB)In request URI, replace device selector with your own device</ol>SYNC FROM DEVICE TO PULL COLOR<ol>As title states, this will pull the color config from the actual device</ol>QUERY DEVICE CONFIG BACKPOINTER DOESNT EXIST<ol>Query NSO device config with service meta-data and verify refcounter and backpointer are NOT presentIn request URI, replace device selector with your own device</ol>REDEPLOY RECONCILE<ol>Redeploy reconcile service to retake ownership of NSO device config color</ol>QUERY DEVICE CONFIG BACKPOINTER EXISTS<ol>Query NSO device config with service meta-data and verify refcounter=1 and backpointer are reestablishedIn request URI, replace device selector with your own device</ol></ol>If you have any questions, please let me know.Thanks,Isaac KimThere was no issue. Sample re-deploy reconcile RESTCONF procedure provided. |
Incorrect Payloads in M2 Release Wiki: L2VPN payloads in the M2 release wiki are incorrect. They still point to the M1 payload format. | Payloads have been fixed for TSDN 5.0 M2 detailed wiki for L2NM. Please let me know if you face any issues with the new sample payloads! |
Sync from and Sync to from CNC for services failing with error application communication failure: Create L3VPN service from CNC creation is successful - performed sync from and sync to action from cnc getting below error"[{"Message":"application communication failure","nsoResponse":"application communication failure"}]" | Hi, a identical CDETS has been filed for this (). As such, we will continue any discussion on the CDETS and close this JSD.Thanks,Isaac Kim |
L2VPN Create Failure: Payload Used:{ "ietf-l2vpn-ntw:vpn-service": { "vpn-id": "l2nm-vpn-o-sr-mpls-2401", "ietf-l2vpn-ntw-cisco:control-word": "yes", "vpn-type": "ietf-vpn-common:vpws", "vpn-nodes": { "vpn-node": [ { "vpn-node-id": "TB3-PE-A", "ietf-l2vpn-ntw-cisco:te-service-mapping": { "te-mapping": { "sr-policy": { "policy": "sr-mpls-policy-a-2401" } } }, "signaling-option": { "ldp-or-l2tp": { "pw-peer-list": [ { "vc-id": 2401, "peer-addr": "100.100.100.6", "ietf-l2vpn-ntw-cisco:mpls-label": 241 } ] } }, "vpn-network-accesses": { "vpn-network-access": [ { "id": "l2vpn-p2p-ac1", "interface-id": "GigabitEthernet0/0/0/1", "service": { "mtu": 65 }, "connection": { "encapsulation": { "encap-type": "ietf-vpn-common:dot1q", "dot1q": { "cvlan-id": 601, "tag-operations": { "push": [ null ], "tag-1": 123, "ietf-l2vpn-ntw-cisco:mode": "symmetric" } } } } } ] } }, { "vpn-node-id": "TB3-PE-B", "signaling-option": { "ldp-or-l2tp": { "pw-peer-list": [ { "vc-id": 2401, "peer-addr": "100.100.100.5", "ietf-l2vpn-ntw-cisco:mpls-label": 242 } ] } }, "vpn-network-accesses": { "vpn-network-access": [ { "id": "l2vpn-p2p-ac1", "interface-id": "GigabitEthernet0/0/0/1", "service": { "mtu": 64 }, "connection": { "encapsulation": { "encap-type": "ietf-vpn-common:dot1q", "dot1q": { "cvlan-id": 601, "tag-operations": { "push": [ null ], "tag-1": 234, "ietf-l2vpn-ntw-cisco:mode": "symmetric" } } } } } ] } } ] } }}Error:{ "ietf-restconf:errors": { "error": [ { "error-type": "application", "error-tag": "malformed-message", "error-path": "/ietf-l2vpn-ntw:l2vpn-ntw/vpn-services/vpn-service", "error-message": "/flat-L2vpn-internal-local-site-service{L2NM-l2nm-vpn-o-sr-mpls-2401-internal TB3-PE-A}: Python cb_validate error. exceptions must derive from BaseException" } ] }}ncs.log:<INFO> 22-Sep-2022::22:05:12.672 nso-10-194-63-180 ncs: ncs progress usid=1315 tid=2476 datastore=running context=rest subsystem=flat_L2vpn_validation validate<INFO> 22-Sep-2022::22:05:12.681 nso-10-194-63-180 ncs: ncs progress usid=1315 tid=2476 datastore=running context=rest subsystem=flat_L2vpn_internal_local_site_validation validate<ERR> 22-Sep-2022::22:05:12.686 nso-10-194-63-180 ncs: devel-c validate error {application, "Python cb_validate error. exceptions must derive from BaseException"} for path /cisco-flat-L2vpn-fp-internal-local-site:flat-L2vpn-internal-local-site-service{L2NM-l2nm-vpn-o-sr-mpls-2401-internal TB3-PE-A}<INFO> 22-Sep-2022::22:05:12.686 nso-10-194-63-180 ncs: ncs progress usid=1315 tid=2476 datastore=running context=rest run validation over the changeset: error (0.023 s)<INFO> 22-Sep-2022::22:05:12.698 nso-10-194-63-180 ncs: ncs delete "/var/opt/ncs/rollbacks/rollback10104.confirm" resulted in enoent<INFO> 22-Sep-2022::22:05:12.698 nso-10-194-63-180 ncs: ncs delete "/var/opt/ncs/rollbacks/rollback10104.confirm.bup" resulted in enoent<INFO> 22-Sep-2022::22:05:12.699 nso-10-194-63-180 ncs: ncs progress usid=1315 tid=2476 datastore=running context=rest releasing transaction lock | Hi SachinPlease provide us the setup details so that we can look into this error. The payload seems to be working fine for us.Thanks,NikhilHad a call with Sachin and the payload seems to be going through fine now. He'll be trying out a few more times and will update in case of any failures.The payload seems to be going through fine and user was able to provision the service. |
Services are getting stuck in zombie: Single Node NSO VM (NSO 5.7.5.1 with M8 - re-install of NSO from a previous version)After re-install able to create a new service but when deleting the same service it's getting stuck in a zombie. Not sure what is going on as have the right NACM and other settings in place, checked M8 install wiki too. | Had webex with Arun ,we tried different things like nacm settings ,sudoers ,even cleaned up cdb ,no luck .For now Arun will bring up new VM and then we will try fresh install to see .ThanksPriyankaIssue was Ha settings in ncs.conf .Arun was able to resolve it .Closing the case as per his confirmation .ThanksPriyanka |
Options for CS-SR: When a CS policy is configured with auto-color enabled, what is the color that is sent by NSO to CNC? What is the range of valid colors for these policies? Is 0 considered a valid color? | Hello Sachin,CS SR-TE auto color utilizes Resource Manager to allocate color values based on a resource pool. The color resource pool should first be defined under ralloc:resource-pools -> idalloc:id-pool which should then be referenced from cs-sr-te-resource-pools -> color-pool-name.When defining your resource pool parameters under ralloc:resource-pools, user can specify a range from which allocation will be done. This resource pool allocation range must match the range of the corresponding config node so that a valid value can be allocated. This means for color, it will be 1-4294967295 as you can see in the model:<div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"><pre class="code-java">leaf color { tailf:info ; description ; type uint32 { range ; }}</pre></div></div>As you can see, 0 is not a valid value as it is not within the color range.Please let me know if you have any other questions.Thanks,Isaac KimHi Sachin,To see what the allocated color value is, you can check (OPER DATA) cs-sr-te-policy -> color-oper-data -> color.Thanks,Isaac KimAs discussed in the meeting we will refactor the allocated oper-data from service to plan. You can expect the oper-data to be stored in the following paths::cs-sr-te-plan -> color-oper-data -> color :cs-sr-te-plan ->bidirectional-association-id-oper-data ->working-path-bidirectional-association-idcs-sr-te-plan ->bidirectional-association-id-oper-data ->protect-path-bidirectional-association-idcs-sr-te-plan ->bidirectional-association-id-oper-data ->restore-path-bidirectional-association-idcs-sr-te-plan ->disjoint-group-id-oper-data ->forward-group-idcs-sr-te-plan ->disjoint-group-id-oper-data ->reverse-group-idFixed in M6 Patch Build |
LSA - M8 - NetConf Notifications are not working: LSA - M8 - NetConf Notifications are not working. CFS is not receiving any notification from any of the RFS nodes. Tried multiple times - the reboot of the CFS and RFS, try running the disconnect and re-connect netconfg subscription command no luck. Finally - had to remove the config as per wiki 'Onboard RFS device onto CFS device tree' and re-do and after that notification started to work. | Had to re-create subscription for the subscription to work. |
NCA FP: When upgrading from NSO6.1 M7 to M8 NSO failed with callback-node is not configured: root@nso-cnc-4-0-ha-1:~#root@nso-cnc-4-0-ha-1:~# date;sudo NCS_RELOAD_PACKAGES=force /etc/init.d/ncs restart --ignore-initial-validation;date;Thu 06 Apr 2023 10:59:12 AM PDTStopping ncs: connection refused (stop)Starting ncs: CDB: Upgrade failed: Upgrade transaction failed to validate: /ncs:devices/authgroups/group{sr-device-1}/umap{cwuser}/callback-node is not configuredDaemon died status=13.Thu 06 Apr 2023 11:22:26 AM PDT Backup files : /auto/nfv_orch/AUTO_DEV/jobjames/logs/5.0/CSCwe95313and config used -> set devices authgroups group juno-device-1 default-mapset devices authgroups group juno-device-1 default-map remote-name adminset devices authgroups group juno-device-1 default-map remote-password $9$20vECdX1uquBxN3UQ45LuPLqYaMJTQ8M9ZWFQCgGeLU=set devices authgroups group sr-device-0 default-mapset devices authgroups group sr-device-0 default-map remote-name adminset devices authgroups group sr-device-0 default-map remote-password $9$oWiTHykfTdqFfiZCDtpe7awxtPo7JEd36jFxsZnURY4=set devices authgroups group sr-device-1 default-mapset devices authgroups group sr-device-1 default-map remote-name adminset devices authgroups group sr-device-1 default-map remote-password $9$NNWOVfK4wI2VoyRCu1VVrUpyZCEjMISqPkItiOAm/3Y=set devices authgroups group sr-device-1 umap cwuser callback-node /cw-creds-getset devices authgroups group sr-device-1 umap cwuser action-name getset devices authgroups group xe-device-1 default-mapset devices authgroups group xe-device-1 default-map remote-name adminset devices authgroups group xe-device-1 default-map remote-password $9$mjuwSEdaOxLpGZWrJynhYGgvKBaeUFU+ClcYXp/C95Y=set devices authgroups group xe-device-2 default-mapset devices authgroups group xe-device-2 default-map remote-name adminset devices authgroups group xe-device-2 default-map remote-password $9$TpMGV/EY/H6GN91nQwVn7I7boWFxmJZl8jPaUKi9CZw=set devices authgroups group xe-device-2 umap cwuser callback-node /cw-creds-getset devices authgroups group xe-device-2 umap cwuser action-name get | Hi Mohamed,Fortunately we have seen this issue before. Please take a look at the definition of callback-node:<div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"><pre class="code-java">4108 leaf callback-node {4109 description4110 "Invoke a standalone action to retrieve login credentials 4111 managed devices on the instance.4112 4113 The action is invoked on the callback node that4114 is specified by an instance identifer.";4115 mandatory ;4116 type instance-identifier;4117 }</pre></div></div>Here we can see callback-node has a type of instance-identifier. Here is the definition of instance identifier:<div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"><pre class="code-java">9.13. The instance-identifier Built-In Type The instance-identifier built-in type is used to uniquely identify a particular instance node in the data tree.</pre></div></div>This means callback-node must reference a valid existing node in NSO CDB. Here are the two callback-node commands you mentioned:<div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"><pre class="code-java">set devices authgroups group sr-device-1 umap cwuser callback-node /cw-creds-getset devices authgroups group xe-device-2 umap cwuser callback-node /cw-creds-get</pre></div></div>Here we are configuring /cw-creds-get which is not native NSO CDB config. This means we are referencing an external model that was loaded from a package. Based on the naming of the node, we can assume this node is part of a CW (Crosswork) package model. Based on the error message "/callback-node is not configured", we can assume that the package which contains /cw-creds-get is not loaded into NSO as the reference does not exist.To fix this, please make sure all dependent CW packages are defined in your NSO packages directory before reloading.Thanks,Isaac Kim |
Need help in Decoupling Netconf template from Functional pack : There is a requirement to decouple the Netconf template we package along with FP. If we load this template manually it throws error. Need help in fixing this error. I have attached the template. Following is the error. nso@ncs# config Entering configuration mode terminalnso@ncs(config)# load merge cisco-tm-tc-native-um-template.xmlLoading.Error: on line 7: no mount id exists for: /ncs:devices/ncs:device[ncs:name='{/name}']/ncs:confignso@ncs(config)# nso@ncs(config)# | Based on the template, it seems like this is a service template which is intended to be applied with the context of your TM TC service.Why are we trying to load this service template from NSO CLI? The service template should be automatically loaded when the TM TC package is loaded into NSO.Thanks,Isaac KimYes, it is a service template but if we have this loaded as part of TMTC package, the installation will fail if there is no netconf ned package installed. That is because this template has reference of netconf ned package. The requirement is to load TM-TC package even if there is no netconf ned package installed.What you could try is using an if-ned-id condition inside the template to protect the XR NC namespace. This should get rid of the NC NED dependency afaik.For example:<div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"><pre class="code-java"><config xmlns=></span> <devices xmlns=></span> <device> <name>{/name}</name> <?set DEVICE = {/name} ?> <config> <?-ned-id cisco-iosxr-nc-7.3:cisco-iosxr-nc-7.3 cisco-iosxr-nc-7.4:cisco-iosxr-nc-7.4 <OTHER SUPPORTED NEDS FOR THIS NAMESPACE>?> <telemetry xmlns=></span> ... <?end?></pre></div></div>But what this will mean is that if this template is applied and there are no matching NEDs for this if-ned-id condition, nothing inside the if-ned-id block will be pushed to device. If the service code applies different templates based on NED type, this should be okay.Please try this out and let me know if it works. Again this fix is for loading the package into NSO when there is no NC NED present, not for NSO CLI load merge.Thanks,Isaac Kimthe package installation is failing with following error admin@ncs# *** ALARM package-load-failure: cisco-tm-tc-native-um-template.xml:9 unknown ned-id: cisco-iosxr-nc-7.3:cisco-iosxr-nc-7.3, cisco-iosxr-nc-7.4:cisco-iosxr-nc-7.4Hi Moksha,Upon further evaluation, it seems like TM-TC is already setup for multivendor support.cisco-tm-tc-fp -> Native NED Support + Service Logictm-tc-multivendors -> Multivendor NED SupportIf NC NED is not the native NED for cisco-tm-tc-fp and we want to decouple the dependency from cisco-tm-tc-fp, we should move the related files to tm-tc-multivendors.Thanks,Isaac KimA solution was provided to submitter and absorbed. |
get-modifications reverse on CFS does not include device config delta: Procedure and Issue:have NSO LSA setupdeployed the L2NM service on CFS.the 'get-modifications reverse' returned only '-flat-L2vpn internal' payload, without device config delta.Note:L2NM service payload is attached.device CL5-PE-A is configured on rfs-1, device CL5-PE-C is configured on rfs-2.the return of 'get-modifications reverse':admin@ncs% request l2vpn-ntw vpn-services vpn-service L2VPN_NM-P2P-PE-A-PE-C get-modifications reverse cli { local-node { data -flat-L2vpn L2NM-L2VPN_NM-P2P-PE-A-PE-C-internal {service-type p2p;flat-L2vpn-p2p {pw-id 301;local-site {pe CL5-PE-A;if-type GigabitEthernet;if-id 0/0/0/1;if-encap dot1q;vlan-id 301;sub-if-id 301;mtu 65;xconnect-group-name L2VPN_NM-P2P-PE-A-PE-C;p2p-name L2VPN_NM-P2P-PE-A-PE-C;sr-te {preferred-path { - policy SR-Policy-1; - }}control-word no;pw-class L2VPN_NM-P2P-PE-A-PE-C;xconnect-local-ip 200.200.200.5;xconnect-remote-ip 200.200.200.7;mpls-local-label 114;mpls-remote-label 115;}remote-site {pe CL5-PE-C;if-type GigabitEthernet;if-id 0/0/0/1;if-encap dot1q;vlan-id 301;sub-if-id 301;mtu 65;xconnect-group-name L2VPN_NM-P2P-PE-A-PE-C;p2p-name L2VPN_NM-P2P-PE-A-PE-C;sr-te {preferred-path { - policy SR-Policy-2; - }}control-word no;pw-class L2VPN_NM-P2P-PE-A-PE-C;}} -}</ul> }}admin@ncs% | Had verified, the config delta was pushed correctly on on the both devices.Hi JasonWe filed an RT for getting support to get-modifications from top level service.Workaround:Currently as work-around to get the device config, we need to do get-modifications reverse from second level service.For Example:When you create l2NM service, get-modifications reverse on flat-l2 service will work as intended.Tested this on your system and was able to verify:```admin@ncs% request flat-L2vpn L2NM-L2VPN_NM-P2P-PE-A-PE-C-internal get-modifications reversecli { local-node { data } lsa-service { service-id /devices/device/config/cisco-flat-L2vpn-fp-internal-local-site:flat-L2vpn-internal-local-site-service data devices { device CL5-PE-A { config { interface {<ul class="alternate" type="square">GigabitEthernet 0/0/0/1 {description "to CL5-P-BOTTOMLEFT";ipv4 {address { - ip 20.20.15.5; - mask 255.255.255.252; - }}} GigabitEthernet-subinterface {GigabitEthernet 0/0/0/1.301 {mode l2transport;description "T-SDN Interface";mtu 65;encapsulation {dot1q {```Please let me know if this workaround works until we make progress with RT.Please also let me know if this functionality was working/not-working for LSA on M1 Release too?I was able to capture the "device config" from the output of "get-modifications reverse" on one of CFP QA's testbed during M1 timeframe (earlier this year), so I believe it was working in M1.Hi Sahithi,Thanks for providing the workaround. Although it could retrieve the device config data in this way, but it is not practically applicable to SH implementation. I'd rather be waiting for the issue getting resolved. Here is the reason:<ul class="alternate" type="square">with this workaround, SH needs to query NSO twice to retrieve the data, because SH should not assume the flat-vpn-internal service yang path and name.the output of workaround is in different data structure (hierarchy).Both of above have to be reworked once the issue is resolved.Can you please re-check if it was LSA setup you captured the "device config" on M1?PLMK if you have the working get-modifications reverse config for LSA on M1, it would be helpful too.Just trying understand if this is a regression issue.ThanksSahithiHi Sahithi,Sorry, I do not have M1 setup. The one I accessed was CFP QA's.Hi Jason,Confirmed with CFP QA that this issue exists in LSA setup on M1 too. I understand this is not a regression issue and a new issue reported. As informed, we filed a RT on NSO platform and working with them on this.ThanksSahithiFixed with M3 release |
y1731-profile -- Needed y1731 profile profile having Probe type with packet: Hi,I am trying to configure y1731 with profile having Probe type option as Packet. Can you please give me the respective xml payload with all possible attibutes for y1731 profile having probe type with packet? | Hi Arun,You can try out creating payloads from CNC or NSO UI .but here is sample :first create profile.set l2vpn-ntw y-1731-profile qa probe burst message-count 60set l2vpn-ntw y-1731-profile qa probe burst message-period 1000set l2vpn-ntw y-1731-profile qa delay-params statistic delay-two-wayThen add profile in L2nm service.ThanksPriyankaHi Priyanka,Thanks for the update.I would need the XML payload having probe type as Packet. this helps for automation purposes.i am able to create with NSO UI.but payload helpsThanks,Arun.M.GHi Arun,Please use NSO CLI and create XML from there.We have provided enough samples on detailed documentation wikis and now test team needs to try their own payload is what we recommend .Thanks,Priyankaplease provide commands to generate the XML from NCS CLI . Hi Arun,As discussed on chat :admin@ncs% show l2vpn-ntw | display xml<config xmlns="http://tail-f.com/ns/config/1.0"> <l2vpn-ntw xmlns="urn:ietf:params:xml:ns:yang:ietf-l2vpn-ntw"> <vpn-services> <vpn-service> <vpn-id>ARUN_XR_L2NM_EVPN</vpn-id> <evi-id>1000</evi-id> <evi-source>200</evi-source> <evi-target>3000</evi-target> <vpn-svc-type>vpn-common:evpn-bgp</vpn-svc-type> <vpn-nodes> <vpn-node> <vpn-node-id>CNC-653</vpn-node-id> <ne-id>CNC-653</ne-id> <te-service-mapping> <te-mapping>.......................--------Please refer to Ncs documentation for NCS hands on:{{admin@nso-192-24:/opt/ncs/current/doc/pdf$ lsdeployment.pdf nso_cdm_migration_guide-5.5.2.9.pdf nso-esc.pdf nso_lsa-5.5.2.9.pdf nso_northbound-5.5.2.9.pdfnso-5-ned-migr.pdf nso-cdm.pdf nso_getting_started-5.5.2.9.pdf nso_man-5.5.2.9.pdf nso_user_guide-5.5.2.9.pdfnso_admin_guide-5.5.2.9.pdf nso_development-5.5.2.9.pdf nso_installation_guide-5.5.2.9.pdf nso_ned-5.5.2.9.pdf nso_webui-5.5.2.9.pdfPlease read ncs user guide and admin guide for more hands on NSO .ThanksPriyankadone. |
Unable to sync device if controller interface has configuration speed 400g-In NCS540: Unable to sync device if controller interface has configuration speed 400g-In NCS540admin@ncs% request devices device ron-ncs540-2 sync-from result falseinfo ron-ncs540-2: invalid value for: speed in /ncs:devices/ncs:device/ncs:config/Cisco-IOS-XR-ifmgr-cfg:interface-configurations/Cisco-IOS-XR-ifmgr-cfg:interface-configuration/Cisco-IOS-XR-optics-speed-cfg:speed: "400g" is an invalid value.admin@ncs% request devices device ron-ncs540-2 check-sync result out-of-syncinfo got: 1000000149 expected: 1000000123admin@ncs% *** ALARM out-of-sync: got: 1000000149 expected: 1000000123Setup: 10.194.57.127 - nso38/Ovatest123@Device - 172.29.11.40 | Hi Ajitha,:<ol>NSO versionCFP build versionDevice type and version used for testingAdd clear steps to reproduce an issue this includes details if some steps were done through NSO UI/CLI or through CNC UI, details on commit options used, order in which payloads were pushed, plan status in between, etc.All relevant payloads.NSO logs including trace if there was something pushed to the deviceIn some cases, wed like to understand the use-case where we see users performing certain actions especially when some steps are done through NSO UI/CLI.</ol>Hi Ajitha,The issue was known and fixed but didn't make CNC 3.0 M6 release due to hardware availability. Please standby for the next release (RON 1.0 M4) where this issue is addressed and fixed.Linmao and I spoke to Ajitha and explained that the additional support for ncs-540 is done but was just not able to make RON 1.0 M3 (CNC 3.0 M6) release. She agreed to take a private build with the speed setting support for ncs-540 and will test it out.Followed up with Ajitha today for updates. She said she will get back to me after consuming M4 early buildFollowed up with Ajitha today regarding additional Speed Setting support on NCS-540 included in M4 build provided and was confirmed and verified. Moving this ticket to resolved and closing the case.09/30 - Followed up with Ajitha regarding additional speed setting support for NCS-540 that was included in RON 1.0 M4 build was tested and verified. Marking this ticket as resolved and closing. |
Please remove the CS SR-TE from TSDN 4.0.0 - M2 - Model and Payload Changes: From - the CS SR-TE section found with Yang file and Sample Payload. Please remove this section as the CS is no longer part of 4.0 | Hi Quy,I will take care of this.Thanks,Isaac KimDone |
RON ML service creation throws error message with correct payload: Setup: Crosswork Platform -163 , COE -134 , CNC -131 , ONC CAPP - w-na-onc-1.1.0-673-test-210726-EXTERNAL-CB2.3.tar.gz , NSO - 5.5.2.3 M4Crosswork/ONC CAPP: 10.195.165.76 - UI: admin/Admin*12345 - CLI: cw-admin/Ovatest123@NSO - 10.195.165.211 - nso210/nso210Creating RON ML service with attached payload which worked well earlier and has valid service name. But NSO throws below error message,RPC error towards onc-real-new: operation_failed: Exception in callback: Missing valid value for name-type SERVICE_NAME for Connectivity service with UUID:. Valid value shall contain alpha numeric with only underscore as the special character with length up to 64 characters.Interlayer link:P-BOTTOMLEFT-VXR- 0/0/0/3- ROADM_SITE3- 1/2003/2/30,29P-BOTTOMRIGHT-VXR- 0/0/0/3- ROADM_SITE10- 1/2018/1/28,27Looks like a defect. Kindly help troubleshoot further | Reached out to Ajitha for troubleshoot session but she was not available.Asked Ajitha to delete the deployed service and try again, but again its the same error. Tried different sips still same error. Will troubleshoot further. Do we know what is the payload sent to the ONC ? do you have device ned log ? Hey Sri, Payload is attached to the ticket. I think logs will be available. She has provided the setup details in descriptionHi Ajitha,Lets connect again once you are available, so that we can troubleshoot this together. It seems like some issue with the sips you are using. ThanksHi Ajitha,As discussed we were able to troubleshoot the issue and find out the root cause for the above issue. It seems like creating the service with same name BL3toBR3 throws that same error everytime. It is removed from NSO but somehow still exist in ONC. Kindly see attachments. We were able to deploy the same payload with different name and it works fine.Kindly work with ONC team how this can be removed.I will be closing this ticket as you have already raised a CDET for this and will redirect it to ONC team.Thanks,Mohit B$ curl --location --request GET 'http://10.195.165.76:30603/crosswork/onc-tapi/restconf/data/tapi-common:context/tapi-connectivity:connectivity-context/connectivity-service/?fields=name(value-name;value);uuid;administrative-state;operational-state;lifecycle-state;connection' --header 'Accept: application/yang-data+json' --header 'Authorization: Basic YWRtaW46QWRtaW4qMTIzNDU=' -k{ "tapi-connectivity:connectivity-service": [ { "uuid": "049e66d5-e1fb-315f-b246-7d12781fc200", "connection": [ { "connection-uuid": "4d0eb749-940a-34ec-8868-8d0ea2a95e49" }, { "connection-uuid": "7a75608d-a1e2-3577-8d45-b34c5bf596d3" } ], "name": [ { "value-name": "SERVICE_NAME", "value": "2" } ], "administrative-state": "UNLOCKED", "operational-state": "DISABLED", "lifecycle-state": "INSTALLED" }, { "uuid": "3e43dfed-da22-37ff-8a70-1fa1c92977ea", "connection": [ { "connection-uuid": "0c504cae-7c62-3828-bb6a-b2deea1a6be5" }, { "connection-uuid": "7a13c323-26f6-3a94-a72b-af86135af0b2" } ], "name": [ { "value-name": "SERVICE_NAME", "value": "Neel_ronml_sr" } ], "administrative-state": "UNLOCKED", "operational-state": "DISABLED", "lifecycle-state": "INSTALLED" }, { "uuid": "a132d136-b0ff-365e-953f-65c1b675f552", "connection": [ { "connection-uuid": "56829965-be33-3d3a-a0f4-64cae3c67068" }, { "connection-uuid": "eb484e7d-9181-326b-ab89-312e720b8d47" } ], "name": [ { "value-name": "SERVICE_NAME", "value": "BL3toBR3" } ], "administrative-state": "UNLOCKED", "operational-state": "DISABLED", "lifecycle-state": "INSTALLED" }, { "uuid": "efae3ee7-8328-3408-83c5-acc249c653dc", "connection": [ { "connection-uuid": "0c93d428-9ebc-3335-b874-1c4081b5f1b1" }, { "connection-uuid": "f2df4efa-c28c-38cd-8f9e-d2ef31efbc5d" } ], "name": [ { "value-name": "SERVICE_NAME", "value": "bl2tobr2" } ], "administrative-state": "UNLOCKED", "operational-state": "DISABLED", "lifecycle-state": "INSTALLED" } ]}Nso lookup says service deleted------------admin@ncs% run show ron ron-service-lookup RON ML OPTICALSERVICE SERVICE CONTROLLER-------------------------------------------------------------------3e43dfed-da22-37ff-8a70-1fa1c92977ea Neel_ronml_sr onc-real-newefae3ee7-8328-3408-83c5-acc249c653dc bl2tobr2 onc-real-new |
Chicken and the egg problem in appending / prepending route policies : I want to use the policy append/prepend statements in route policydefinition <l3vpn-route-policyxmlns=><name>L3VPN_NM-SRTE-ODN-PE-A-44</name><color><id>44</id><ipv4><prefix>40.40.44.0/30</prefix></ipv4></color><extra-policy><name>L3VPN_NM-SRTE-ODN-PE-A-45</name><operation>append</operation><address>both</address> </extra-policy> </l3vpn-route-policy>The thing is that policy I append/prepend has to be on the device,because otherwise I get this error <errorsxmlns="urn:ietf:params:xml:ns:yang:ietf-restconf"> <error><error-type>application</error-type><error-tag>malformed-message</error-tag><error-path></error-path> <error-message>RPCerror towards PE-C: operation_failed: for ns2:vrfs/ns2:vrf[vrf-name ='L3VPN_NM-SRTE-ODN-40']/ns2:afs/ns2:af[af-name = 'ipv4' and saf-name ='unicast' and topology-name = 'default']/ns1:bgp/ns1:export-route-policy:Verify failed for policy: Policy usespolicy . must be definedbefore can be attached.</error-message> </error></errors>On the other hand, I dont have a manner to create the route policy onthe device without creating VPN service .This leads to a conclusion that I can not create policies I want toappend/prepend through NSO, but only using CLI. As far as I see it, this is a problem ! | Are you able to use custom-template to resolve this issue? You can refer to flow here: Hi Yevgeny, did the custom-templates resolve your issue?Closing ticket due to no response from submitter |
Need help on configure NSO HA (L2) on GEO-HA setup : Please help us to configure NSO HA on GEO-HA lab setup where we only use DNS to point to the primary instead of VIPThanks and regards. | Hi Quy, We had helped configuring an L2 setup in your lab. If no other request pending, I'd like to close this ticket. You can always open a new ticket as needed.Regards,VanThank you. Please close the ticket.Regards.Setup completed. |
While deleting the devices from NSO, getting this error "illegal reference 'core-fp-delete-shutdown-service R1 GigabitEthernet 0/0/0/4 device'": Though there are no zombies in NSO and when we try to delete the devices from NSO , get this error " "illegal reference 'core-fp-delete-shutdown-service R1 GigabitEthernet 0/0/0/4 device'"NSO logs - /auto/ncahi/Users/kbhateja/cw/data_042822_123350.tgz | Hi, KritikaI looked the logs but still don't see any root cause to this issue. Wondering there is service that referred to this device and then you delete the devices?Can you provide the steps that cause this issue?Thanks,LinhHi Linh,The steps that we are doing is - 1. Creating CNC policy (SR, L3VPN)2. Updating the policy created in step-1.3. Deleting the policy.Now sometimes in step-3 , we get the zombies created on NSO. The created zombies are cleaned by re-deploy, resurrect or cleanup command.Later on when we try to delete the devices from NSO , the error "illegal reference 'core-fp-delete-shutdown-service R1 Gigabit Ethernet 0/0/0/4 device'" is seen. Thanks,KritikaAlso attaching the configs that are pushed to the device while creation of policy. Hi Kritika,You created two services in step 1 (SR-Policy and L3VPN) and step 3 you ONLY delete one service which is SR-Policy. L3VPN service is still exist and then you delete devices. NSO CDB still has dangling reference to interface GigabitEthernet 0/0/0/4 which defined in L3VPN service. You should delete L3VPN service before deleting devices. The best practice is when deleting devices, you should do a commit dry-run to see if there is still reference to devices from some services. If there is then you go back and delete then service first then delete the devicesSo it seems the issue occurs when improper action is performed.Thanks,LinhHi Linh,Actually we are creating a policy and deleting it before creating another policy.Also we are making sure to delete all policies and clean the zombies before deleting the devices from NSO but still hitting the issue.Thanks,KritikaHi Kritika,Please try do reproduce this issue again with proper steps. If it still happens then we could have webex to debug it.Thanks,Linh |
Provisioning service on XRD throws "External error in the NED implementation": Hi-Provisioning service on XRD throws below error message.Fri May 13 17:16:38 IST 2022:DEBUG: << " "error-message": "External error in the NED implementation for device xrd-6: config exclusive: config exclusiveFri May 13 11:46:30.782 UTC""admin@ncs> show devices listNAME ADDRESS DESCRIPTION NED ID ADMIN STATE--------------------------------------------------------------------------------TB3-P-BOTTOMLEFT 172.27.226.53 - cisco-iosxr-cli-7.39 unlockedTB3-P-BOTTOMRIGHT 172.27.226.54 - cisco-iosxr-cli-7.39 unlockedTB3-P-TOPLEFT 172.27.226.51 - cisco-iosxr-cli-7.39 unlockedTB3-P-TOPRIGHT 172.27.226.52 - cisco-iosxr-cli-7.39 unlockedTB3-PE-A 172.27.226.55 - cisco-iosxr-cli-7.39 unlockedTB3-PE-B 172.27.226.56 - cisco-iosxr-cli-7.39 unlockedTB3-PE-C 172.27.226.57 - cisco-iosxr-cli-7.39 unlockedTB3-PE-D 172.27.226.93 - cisco-iosxr-cli-7.39 unlockedTB3-PE-E 172.27.226.94 - cisco-iosxr-cli-7.39 unlockedxrd-1 192.168.5.17 - cisco-iosxr-cli-7.39 unlockedxrd-2 192.168.5.18 - cisco-iosxr-cli-7.39 unlockedxrd-3 192.168.5.19 - cisco-iosxr-cli-7.39 unlockedxrd-4 192.168.5.20 - cisco-iosxr-cli-7.39 unlockedxrd-5 192.168.5.21 - cisco-iosxr-cli-7.39 unlockedxrd-6 192.168.5.22 - cisco-iosxr-cli-7.39 unlockedxrd-7 192.168.5.23 - cisco-iosxr-cli-7.39 unlockedxrd-8 192.168.5.24 - cisco-iosxr-cli-7.39 unlocked | hi NSO team, please hold off on this issue as this is only a private build for M6. But if this is a known issue please let us know. Thanks and best regards.Closing this as its a duplicate of .ThanksSahithi |
RONCFPSD-532 - CFS - reporting illegal reference device sync for TM-TC Packages: Following with the below JIRA ticket: As we don't have an option to re-open the ticket, Opening the new one to track the same issue as removing TM-TC package from the system to provision the services is not an acceptable work around. Until we get the closure for this issue, Please keep the ticket open and let us know the right component to track the issue in the CDETS. | created which is having the fix related toncs-6.1.2-cisco-tm-tc-fp-cfs-ned-1.0.tar.gz ;ncs-6.1.2-cfs-cisco-tm-tc-fp-6.0.0-35.tar.gz in CFS node Hello Krishnaraj,We have scanned the logs for any obvious symptoms, but there are no obvious errors. We did observe some kicker and read/write conflict errors, but there is no details in the trace so we cannot come to any conclusion as of yet.Having a live setup with the reproduced error would be the ideal scenario and RCA can be done much faster due to the context we can observe on your setup. I would suggest to continue testing setup with TMTC package and contact a TMTC team member with the live setup details once reproduced.In the meantime, we will continue to parse the logs to see if we can find anything.Thanks,Isaac Kim |
delete device crashes NSO in scale setup: I have a CNC 3.0 TSDN setup with NSO 5.5.2.9. It has 25K devices. I want to delete these devices, since migration is taking time. I tried deleting 25K devices. It took 1 hour 5 mins for the response. When i committed the change, NSO is restarted. I tried this twice. Can you help me to delete these devices starting with Node_*setup : 172.20.63.189 - cnctest/Ovatest123@ | Hello, just to help isolate the crash through logs, approximately what day and time did you encounter the crash? Doesnt have to be exact, but can help me filter through other error logs. Also just want to confirm you are committing through CNC WebUI when you encounter the crash.Thanks,Isaac KimHi Isaac, Ramakrishna has signed-off for the day. He has tried this today IST day time. Not sure how the commit was done. FYI.Thanks,Sreedhar.Thank you, I will try to go off of this.I will post update if I find anything.Thanks,Isaac KimHi as requested to help with device deletion, can I try to delete Node_101_1 device to see behavior? There seems to be connection timeout with Node_X devices.Thanks,Isaac KimYes. Please go ahead and delete all nodes with Node_X. These are sheer simulator nodes and don't have network connectivity directly to NSO. They are reachable to DLM only.Seems like deleting device Node_101_1 was successful, but seems like dependency validation took much longer than expected (253.609s).Hi,When deleting devices in batches of Node_10X and Node_11X, there is some delay but it goes through fine. When deleting devices in bulk (all), I have observed the crash. Whats strange is that there is no indication of a NSO crash as there is no NSO crash dump logs.In any case, NSO crash is not acceptable. But not sure if this crash is due to resource overload or due to some NSO internal mechanism. This is definitely not a CFP issue, so I will have to raise an NSO platform RT for further investigation.In the meantime, to move forward with your device cleanup, I would suggest to delete the devices in batches. I would try 1K devices first and if that crashes NSO then 500. Unfortunately, with no crash logs, there is nothing to indicate what is source of crash.I will post an update here when the RT is created.Thanks,Isaac KimI tried deleting the devices only through ncs_cli. "no device Node_*" , and "commit"Hello Rama, because there is no crash log from NSO upon crash, we suspect that the process was immediately killed for some reason. Maybe due to resource overload or some other system related cause. To verify NSO's capability in this scenario,our QA team has cloned your CDB and executed a bulk device delete (all devices) and NSO did not crash despite taking 1 hour 15 minutes for the delete to complete. For consistency, they have run through this scenario two times and encountered no crash.QA has recommended to try to switch to different vm and esxi. At this point we believe NSO crash is related to system, not NSO itself.Please let me know if this recommendation works. If not, we can discuss further!Thanks,Isaac KimWhen i delete device (bulk) as such, NSO return the prompt after 1 hour, plus few minutes. Only when commit is issued, i see the crash. Did they try with commit as well?Hi Rama, I apologize for the late update.I have been able to reproduce your crash upon commit. We have tried to look for the source of crash, but there is no evidence in any system logs or NSO crash dump logs.In this case, I will be creating a platform ticket so that the platform team can take a look. Once the ticket is raised, I will update this comment section.Thanks,Isaac KimHi Rama, I have filed the ticket with platform team with reproduction setup. I will update you if we hear back from them.Thanks,Isaac KimHi Rama, apologies for the late reply.The NSO platform team have found two relevant issues in regards to the bulk delete NSO crash issue. Both have been fixed starting NSO 5.5.3 and NSO 5.7.4.Because this issue stems from resource overload (combination of host system and NSO causing crash), the workaround for TSDN 3.0 is to delete devices in batches (commit each batch).Thanks,Isaac Kim |
RFS nodes gets only 12520+16 nodes after attaching to 25K node CW scale setup: Added new NSO LSA to CW which have 25K nodes. After more than half a day, RFS1 learnt only 12520 nodes and RFS2 have 16 nodes.RFS1 - 172.20.63.144 - 12520 DevicesRFS2 - 172.20.63.145 - 16 DevicesCFS - 172.20.63.143All NSO Logins : cnctest/Ovatest123@CW/CNC : 172.20.63.202 (cw-admin/admin/Ovatest123@)Kindly check and help resolve the issue. Both RFS nodes should combinely should get 25K nodes. (in earlier versions using NSO standard mode, it use to always work in few hours) | Hi Sreedhar,Could you please share what exact steps you performed?ThanksPriyankaHi Sreedhar,I do not see any issue with NSO at this point .CFS is showing the correct number of nodes it fetched from RFS nodes.Please check with the CW/CNC team about this not reflecting on NSO.ThanksPriyankaHi Priyanka,After raising case, have realized that I did'nt edit the NSO sync policy in CW under DLM before adding NSO's to CW. I have removed the NSO's from CW now and will edit the NSO sync policy and re-add NSO's under provider page. If still see same issue, will update the case.Steps done were - NSO LSA setup earlier has only xrv9k's devices (around 32) which was learnt from DLM when added to previous CW setup. Then using script we have increased VPN services to 300K, while NSO is not added to any CW. Now in the current CW setup which also have 25K sheer nodes, added NSO's under provider. After half a day, when noticed RFS nodes - we only have 12520 (RFS1) + 16 (RFS2) nodes learnt. It was in same state from more than a day. From DLM, could see all 25K nodes have RFS entries split across 50% each.Thanks,Sreedhar.Hi ,Repeated this test after setting LSA mode in CW. Still same issue. Looks like RFS2 there is no sync of devices happened. So is this issue with DLM sync ? Any logs from NSO showing sync from CW is not happening to this RFS, will be helpful to reach right team. (either CW/CNC)Thanks,Sreedhar.Yes please follow up with DLM team.ThanksPriyanka |
L2VPN dry-run failed with PASS_ALL is not allowed as parent route policy name: Steps:1. Provision L2VPN with ODN attached. 2. Set parent-rr-route-policy = PASS_ALL3. Do Dry-run It failed with PASS_ALL is not allowed error. See screenshot for complete message.Is this expected?Note: Tried parent-rr-route-policy with different values such as PASS, PASS1. Both worked fine.payload and screenshot attached. | Screenshot from wiki Hi Crystal ,Please check the Release wiki for M7 .PASS_ALL" is a reserved policy name for CNC, therefore it is not allowed as L2 parent-rr-route-policy.*ThanksPriyanka |