[202311] [Mellanox] run module initialization when any SFP related API is called (#18930) #18937
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…ed (sonic-net#18930) Why I did it Currently, there are a few issues related to module host management: 1. module initialization flow is triggered when chassis.get_change_event is called for the first time. It is too late which delays fast-reboot/warm-reboot convergence time 2. module initialization flow always send change event for all SFP objects to xcvrd even if there is no real module change event. It causes ports flapping during warm-reboot 3. legacy mode and module host management mode are mixed into the same code logic which makes it hard to maintain and extend To address above issues, the PR introduces following changes: 1. module initialization flow is triggered when chassis SFP related API is called for the first time. It means that any of the following API will trigger module init: chassis.get_sfp, chassis.get_all_sfps, chassis.get_change_event. The init flow shall only execute once under xcvrd context. 2. module initialization flow does not send change event anymore. Change event shall be sent only if a real cable event is detected by chassis.get_change_event (plug-in/plug-out/error) 3. legacy mode and module host management mode are decoupled to avoid affecting each other How to verify it 1. Unit test covered most of the new change 2. Full sonic-mgmt regression test make sure there is no degradation (test branch 202311) 3. Manual test
Junchao-Mellanox
requested review from
keboliu and
lguohan
and removed request for
lguohan
11 tasks
keboliu
approved these changes
May 10, 2024
|
@Junchao-Mellanox can you confirm that this change has been tested on 202311 branch? |
|
Hi @yxieca , yes, it was tested based on 202311 for all nvidia platforms with a few rounds of regression. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Backport PR #18930
Why I did it
Currently, there are a few issues related to module host management:
chassis.get_change_eventis called for the first time. It is too late which delays fast-reboot/warm-reboot convergence timeWork item tracking
How I did it
To address above issues, the PR introduces following changes:
chassis.get_sfp,chassis.get_all_sfps,chassis.get_change_event. The init flow shall only execute once under xcvrd context.chassis.get_change_event(plug-in/plug-out/error)How to verify it
Which release branch to backport (provide reason below if selected)
Tested branch (Please provide the tested image version)
Description for the changelog
Link to config_db schema for YANG module changes
A picture of a cute animal (not mandatory but encouraged)