OCPBUGS-60273: Ensure revision.json persists on ungraceful shutdown #1494
Conversation
openshift-ci-robot
added
jira/valid-reference
|
@clobrano: This pull request references Jira Issue OCPBUGS-60273, which is invalid:
Comment The bug has been updated to refer to the pull request using the external bug tracker. In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the openshift-eng/jira-lifecycle-plugin repository. |
openshift-ci
bot
added
the
do-not-merge/work-in-progress
|
Skipping CI for Draft Pull Request. |
clobrano
changed the title
WalkthroughUpdated atomic temporary-file save in Changes
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Pre-merge checks and finishing touches❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
clobrano
force-pushed
the
OCPBUGS-60273-ensure-revision-json-is-written-on-disk
branch
from
a3d111a to
bbc1c24
Compare
clobrano
changed the title
openshift-ci
bot
removed
the
do-not-merge/work-in-progress
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (2)
pkg/cmd/rev/rev.go (2)
192-192: Approve with suggestion: Consider explicitfile.Sync()for clarity.Adding
os.O_SYNCcorrectly ensures that theWriteoperation on line 203 blocks until data is physically written to disk, preventing empty files on ungraceful shutdown. This addresses the issue described in the PR.However, consider using an explicit
file.Sync()call after the write instead, which is more conventional and makes the intent clearer:- file, err := os.OpenFile(tmpPath, os.O_CREATE|os.O_WRONLY|os.O_TRUNC|os.O_SYNC, 0644) + file, err := os.OpenFile(tmpPath, os.O_CREATE|os.O_WRONLY|os.O_TRUNC, 0644) if err != nil { klog.Errorf("error opening file: %v", err) return } defer func() { if err = file.Close(); err != nil { klog.Errorf("error closing file: %v", err) } }() _, err = file.Write(jsonOutput) if err != nil { klog.Errorf("error writing result to file: %v", err) return } + + if err = file.Sync(); err != nil { + klog.Errorf("error syncing file to disk: %v", err) + return + } if err = os.Rename(tmpPath, outputFile); err != nil {Both approaches provide the same durability guarantee, but explicit
Sync()is more idiomatic and slightly more efficient (one sync operation after all writes vs synchronous mode for each write). Given the 10-second polling interval, performance impact is minimal either way.
209-212: Optional: Consider syncing the parent directory for complete durability.While the temp-file-then-rename pattern ensures atomic updates, for strict durability guarantees after ungraceful shutdown, you may also want to sync the parent directory after the rename to ensure the directory entry update is persisted:
if err = os.Rename(tmpPath, outputFile); err != nil { klog.Errorf("error during rename to destination file: %v", err) return } // Optional: sync parent directory to ensure rename is durable if dir, err := os.Open(filepath.Dir(outputFile)); err == nil { dir.Sync() dir.Close() }However, this is often omitted in practice as most filesystems handle this reasonably well, and the risk is low for this use case (if the rename doesn't complete, the old file remains valid). This is mentioned for completeness.
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
Cache: Disabled due to data retention organization setting
Knowledge base: Disabled due to Reviews -> Disable Knowledge Base setting
📒 Files selected for processing (1)
pkg/cmd/rev/rev.go(1 hunks)
clobrano
force-pushed
the
OCPBUGS-60273-ensure-revision-json-is-written-on-disk
branch
from
bbc1c24 to
0d67b4e
Compare
There was a problem hiding this comment.
Actionable comments posted: 1
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
Cache: Disabled due to data retention organization setting
Knowledge base: Disabled due to Reviews -> Disable Knowledge Base setting
📒 Files selected for processing (1)
pkg/cmd/rev/rev.go(1 hunks)
🔇 Additional comments (1)
pkg/cmd/rev/rev.go (1)
192-192: Good addition of O_SYNC for durability.Adding
os.O_SYNCensures writes are flushed to disk immediately, which is exactly what's needed to prevent empty or corrupted files on ungraceful shutdown.
Add os.O_SYNC flag to OpenFile call in trySaveRevision to force synchronous writes to disk. This prevents the revision.json file from being empty if the process terminates unexpectedly before buffered data is flushed.
clobrano
force-pushed
the
OCPBUGS-60273-ensure-revision-json-is-written-on-disk
branch
from
0d67b4e to
03de23b
Compare
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (1)
pkg/cmd/rev/rev.go (1)
197-211: Critical issue resolved: close error now gates rename.The defer function correctly prevents renaming if either write or close fails. The early-return pattern (lines 199-200) ensures that a failed close prevents the rename, addressing the critical issue from the previous review.
The current implementation achieves the same safety as the previously suggested combined check (
if err != nil || closeErr != nil) but uses idiomatic early returns instead, which is a valid Go pattern.Minor: The comment at line 202 could be more precise—consider "Only rename if write succeeded" since close is already verified above.
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
Cache: Disabled due to data retention organization setting
Knowledge base: Disabled due to Reviews -> Disable Knowledge Base setting
📒 Files selected for processing (1)
pkg/cmd/rev/rev.go(1 hunks)
🔇 Additional comments (1)
pkg/cmd/rev/rev.go (1)
192-192: LGTM: O_SYNC ensures durability on ungraceful shutdown.The addition of
os.O_SYNCforces synchronous writes, ensuring that data reaches the disk immediately rather than being buffered. This directly addresses the issue whererevision.jsoncould be empty after unexpected process termination.
|
/lgtm thanks @clobrano! |
|
/cherry-pick release-4.20 release-4.19 release-4.18 |
|
/jira refresh |
|
@tjungblu: once the present PR merges, I will cherry-pick it on top of In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. |
openshift-ci-robot
added
jira/valid-bug
|
@tjungblu: This pull request references Jira Issue OCPBUGS-60273, which is valid. The bug has been moved to the POST state. 3 validation(s) were run on this bug
No GitHub users were found matching the public email listed for the QA contact in Jira ([email protected]), skipping review request. The bug has been updated to refer to the pull request using the external bug tracker. In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the openshift-eng/jira-lifecycle-plugin repository. |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: clobrano, tjungblu The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
openshift-ci
bot
added
the
approved
|
/retest-required |
1 similar comment
|
/retest-required |
openshift-ci-robot
added
the
verified
|
@jaypoulz: This PR has been marked as verified by In response to this: Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the openshift-eng/jira-lifecycle-plugin repository. |
|
@clobrano: The following test failed, say
Full PR test history. Your PR dashboard. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. I understand the commands that are listed here. |
openshift-merge-bot
bot
merged commit c0663ed
into
openshift:main
|
@clobrano: Jira Issue Verification Checks: Jira Issue OCPBUGS-60273 Jira Issue OCPBUGS-60273 has been moved to the MODIFIED state and will move to the VERIFIED state when the change is available in an accepted nightly payload. 🕓 In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the openshift-eng/jira-lifecycle-plugin repository. |
|
Fix included in accepted release 4.21.0-0.nightly-2025-10-15-101012 |
|
/cherry-pick release-4.20 release-4.19 release-4.18 |
|
@jaypoulz: new pull request created: #1501 In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. |
6 participants
Add
os.O_SYNCflag to OpenFile call intrySaveRevisionto force synchronous writes to disk. This prevents therevision.jsonfile from being empty if the process terminates unexpectedly before buffered data is flushed