Repository navigation
DAOS-17861 cart: avoid sending RPC reply repeatedly - #16757
Merged
Merged
Conversation
|
Ticket title is 'Segmentation fault against UCX provider during CR test' |
Nasf-Fan
force-pushed
the
Nasf-Fan/DAOS-17861
branch
4 times, most recently
from
September 3, 2025 09:50
fa37d57 to
8e8dfb3
Compare
Nasf-Fan
force-pushed
the
Nasf-Fan/DAOS-17861
branch
2 times, most recently
from
September 8, 2025 09:14
9ba7691 to
9a773f1
Compare
Collaborator
|
Test stage Functional Hardware Medium MD on SSD completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net//job/daos-stack/job/daos/view/change-requests/job/PR-16757/9/execution/node/1468/log |
Nasf-Fan
force-pushed
the
Nasf-Fan/DAOS-17861
branch
from
September 18, 2025 03:47
9a773f1 to
83698cf
Compare
Contributor
Author
|
Resolve merge conflict. |
Collaborator
|
Test stage Functional Hardware Large MD on SSD completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net//job/daos-stack/job/daos/view/change-requests/job/PR-16757/10/execution/node/1307/log |
When handle collective RPC, some failure may happen before invoking RPC handler for local node process. Then crt_hg_reply_send() may be triggered. And then in subsequent process, crt_rpc_handler_common() will call crt_hg_reply_error_send() to reply the RPC repeatedly. It is observed that the latter one maybe failed with NA_BUSY and cause the callback for former reply to be blocked or lost. Then reference on the RPC cannot be released. Such RPC leaking may cause assertion in UCX environment when destroy related CaRT context. Allow-unstable-test: true Signed-off-by: Fan Yong <fan.yong@hpe.com>
Nasf-Fan
force-pushed
the
Nasf-Fan/DAOS-17861
branch
from
September 24, 2025 02:08
83698cf to
cb375f7
Compare
Nasf-Fan
marked this pull request as ready for review
September 25, 2025 04:50
frostedcmos
approved these changes
Sep 30, 2025
Contributor
Author
|
Ping reviewers, thanks! |
soumagne
approved these changes
Oct 13, 2025
gnailzenh
approved these changes
Oct 14, 2025
frostedcmos
added a commit
that referenced
this pull request
Oct 17, 2025
This reverts commit 69116f4.
mchaarawi
added a commit
that referenced
this pull request
Oct 17, 2025
This reverts commit 69116f4.
mchaarawi
pushed a commit
that referenced
this pull request
Oct 20, 2025
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
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.
When handle collective RPC, some failure may happen before invoking RPC handler for local node process. Then crt_hg_reply_send() may be triggered. And then in subsequent process, crt_rpc_handler_common() will call crt_hg_reply_error_send() to reply the RPC repeatedly. It is observed that the latter one maybe failed with NA_BUSY and cause the callback for former reply to be blocked or lost. Then reference on the RPC cannot be released. Such RPC leaking may cause assertion in UCX environment when destroy related CaRT context.
Allow-unstable-test: true
Steps for the author:
After all prior steps are complete: