Reset skip error counter on a fabric link if it was down. #3247
Conversation
|
Related to sonic-net/sonic-buildimage#19288 |
|
@saksarav-nokia to review too |
|
Tested the fix and seems to be working. However i see lot of these log messages and not sure if we need have them at NOTICE level 2024 Aug 5 20:50:16.520538 ixre-cpm-chassis12 NOTICE swss0#orchagent: :- updateStateDbTable: PORT0 updates POLL_WITH_ERRORS to 0 0 |
|
I can change the message level to INFO |
Changed the log level to INFO, so it will not show up by default now |
|
can anyone help review again and see if we can merge up this ? thank you |
|
test_HashSwitchGlobalConfiguration[inner-frame-ecmp-hash] failed (1 runs remaining out of 2). |
|
/azpw run Azure.sonic-swss |
|
/AzurePipelines run Azure.sonic-swss |
|
Azure Pipelines successfully started running 1 pipeline(s). |
|
I don't quite get the coverage results, I added test to cover the change and get coverage tests passed before retarting the pipeline. How it can fail the coverage test after the rerun ... start again /azpw run Azure.sonic-swss |
|
/azpw run Azure.sonic-swss |
|
/AzurePipelines run Azure.sonic-swss |
|
Azure Pipelines successfully started running 1 pipeline(s). |
|
/azpw run coverage.Azure.sonic-swss.vstest |
|
@prsunny, can you please help merge this PR |
…3247) What I did reset the skip error counters for fabric link monitoring if a link was down and up again. Why I did it The fabric link monitoring feature does not take care the peer card restart cases. Peer card reload could lead to links go in to isolation state, and that's a case where we can ignore the error due to the init churn. So in this change , the skip error on init counters get reset if a link was down and up again.
|
Cherry-pick PR to 202405: #3281 |
What I did reset the skip error counters for fabric link monitoring if a link was down and up again. Why I did it The fabric link monitoring feature does not take care the peer card restart cases. Peer card reload could lead to links go in to isolation state, and that's a case where we can ignore the error due to the init churn. So in this change , the skip error on init counters get reset if a link was down and up again.
mssonicbld
added
Included in 202405 Branch
and removed
Created PR to 202405 Branch
labels
What I did
reset the skip error counters for fabric link monitoring if a link was down and up again.
Why I did it
The fabric link monitoring feature does not take care the peer card restart cases. Peer card reload could lead to links go in to isolation state, and that's a case where we can ignore the error due to the init churn. So in this change , the skip error on init counters get reset if a link was down and up again.
How I verified it
Details if related