-
Notifications
You must be signed in to change notification settings - Fork 305
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
DAOS-1946 md: fix initial solution for metadata full handling #2057
Merged
Conversation
This file contains 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
The original solution for rdb_raft_state checking to handle -DER_NOSPACE condition was to trigger log compaction and step down as the service leader. A problem with this implementation in commit a614cb1 is that it compacts to the current index. With this fix, it compacts to the committed index. Test-tag-hw-large: pr,hw,large metadatafill metadata_free_space Signed-off-by: Ken Cain <kenneth.c.cain@intel.com>
liw
approved these changes
Mar 10, 2020
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks.
Functional failed to start daos_agent... That's strange. |
Created a bug for this https://jira.hpdd.intel.com/browse/DAOS-4313 |
johannlombardi
approved these changes
Mar 10, 2020
kccain
added a commit
that referenced
this pull request
Mar 10, 2020
Cherry picked: commit a614cb1 PR #1956 and commit 0bb952e808906c1575e33a345a6d95a4c3f5bc2 PR #2057 from daos master branch to release/0.9 branch. Change rdb_raft state checking code so that when -DER_NOSPACE condition is observed when appending to the raft log, it is handled like the -DER_NOMEM case (become follower, step down). Also trigger rdb log compaction aggressively seeking to reclaim space. Before this change, stopping the service may leave it "dead" impacting subsequent resource destroy operations (e.g., pool destroy). Re-enable the metadatafill test and run it with multiple (4) servers and pool service replicas (3). Adjust the maximum number of containers to approximately 98% of what can be accommodated in a metadata capacity of 128MB. Test-tag-hw-large: pr,hw,large metadatafill metadata_free_space Signed-off-by: Ken Cain <kenneth.c.cain@intel.com>
kccain
added a commit
that referenced
this pull request
Mar 18, 2020
Cherry picked: commit a614cb1 PR #1956 and commit 0bb952e808906c1575e33a345a6d95a4c3f5bc2 PR #2057 from daos master branch to release/0.9 branch. Change rdb_raft state checking code so that when -DER_NOSPACE condition is observed when appending to the raft log, it is handled like the -DER_NOMEM case (become follower, step down). Also trigger rdb log compaction aggressively seeking to reclaim space. Before this change, stopping the service may leave it "dead" impacting subsequent resource destroy operations (e.g., pool destroy). The metadatafill test is being disabled again. While it passes frequently with the above change to DAOS, intermittently it fails with different symptoms when metadata storage is exhausted. Test-tag-hw-large: pr,hw,large metadatafill metadata_free_space Signed-off-by: Ken Cain <kenneth.c.cain@intel.com>
jolivier23
pushed a commit
that referenced
this pull request
Mar 18, 2020
Cherry picked: commit a614cb1 PR #1956 and commit 0bb952e808906c1575e33a345a6d95a4c3f5bc2 PR #2057 from daos master branch to release/0.9 branch. Change rdb_raft state checking code so that when -DER_NOSPACE condition is observed when appending to the raft log, it is handled like the -DER_NOMEM case (become follower, step down). Also trigger rdb log compaction aggressively seeking to reclaim space. Before this change, stopping the service may leave it "dead" impacting subsequent resource destroy operations (e.g., pool destroy). The metadatafill test is being disabled again. While it passes frequently with the above change to DAOS, intermittently it fails with different symptoms when metadata storage is exhausted. Signed-off-by: Ken Cain <kenneth.c.cain@intel.com>
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.
The original solution for rdb_raft_state checking to handle
-DER_NOSPACE condition was to trigger log compaction and step down
as the service leader. A problem with this implementation in commit
a614cb1 is that it compacts to the
current index. With this fix, it compacts to the committed index.
Test-tag-hw-large: pr,hw,large metadatafill metadata_free_space
Signed-off-by: Ken Cain kenneth.c.cain@intel.com