sql/copy: fix rare flake in TestLargeCopy by yuzefovich · Pull Request #160764 · cockroachdb/cockroach

yuzefovich · 2026-01-09T11:29:34Z

We have automatic retry mechanism for COPY but it can only be used for non-atomic COPY. If we have the atomic COPY and hit a txn retry error, it's bubbled up to the client. We now adjust TestLargeCopy to match this behavior fixing a rare flake where we'd fail the test on the txn retry error when we should've ignored it.

Fixes: #160537.
Release note: None

We have automatic retry mechanism for COPY but it can only be used for non-atomic COPY. If we have the atomic COPY and hit a txn retry error, it's bubbled up to the client. We now adjust `TestLargeCopy` to match this behavior fixing a rare flake where we'd fail the test on the txn retry error when we should've ignored it. Release note: None

cockroach-teamcity · 2026-01-09T11:29:48Z

This change is

yuzefovich · 2026-01-09T11:30:56Z

For context, the relevant block is

cockroach/pkg/sql/copy_from.go

Lines 1171 to 1193 in a10d701

    
           // It is currently only safe to retry if we are not in atomic copy 
        
           // mode & we are in an implicit transaction. 
        
           // 
        
           // NOTE: we cannot re-use the connExecutor retry scheme here as COPY 
        
           // consumes directly from the read buffer, and the data would no 
        
           // longer be available during the retry. 
        
           if c.implicitTxn && !c.p.SessionData().CopyFromAtomicEnabled && c.p.SessionData().CopyFromRetriesEnabled { 
        
           	log.SqlExec.Infof(ctx, "%s is retrying", c.copyFromAST.String()) 
        
           	if c.p.ExecCfg().TestingKnobs.CopyFromInsertRetry != nil { 
        
           		if err := c.p.ExecCfg().TestingKnobs.CopyFromInsertRetry(); err != nil { 
        
           			return err 
        
           		} 
        
           	} 
        
           	continue 
        
           } else { 
        
           	log.SqlExec.Infof( 
        
           		ctx, 
        
           		"%s is not retrying; "+ 
        
           			"implicit: %v; copy_from_atomic_enabled: %v; copy_from_retryable_enabled %v", 
        
           		c.copyFromAST.String(), c.implicitTxn, 
        
           		c.p.SessionData().CopyFromAtomicEnabled, c.p.SessionData().CopyFromRetriesEnabled, 
        
           	) 
        
           }

michae2

Thanks!

@michae2 reviewed 1 file and all commit messages, and made 1 comment.
Reviewable status: complete! 1 of 0 LGTMs obtained (waiting on @yuzefovich).

yuzefovich · 2026-01-09T22:56:51Z

TFTR!

bors r+

160632: sql/bulkmerge: reuse SST iterator across bulk merge tasks r=spilchen a=spilchen This change reduces overhead in the bulk merge processor by initializing a single iterator over all input SSTs at startup, rather than creating a new one per task. The iterator is reused across tasks, seeking only when needed. Informs #159414 Epic: CRDB-48845 Release note: none Co-authored by: `@jeffswenson` 160760: execbuilder: fix a stats-related flake in a new test r=yuzefovich a=yuzefovich Fixes: #160752. Fixes: #160753. Release note: None 160764: sql/copy: fix rare flake in TestLargeCopy r=yuzefovich a=yuzefovich We have automatic retry mechanism for COPY but it can only be used for non-atomic COPY. If we have the atomic COPY and hit a txn retry error, it's bubbled up to the client. We now adjust `TestLargeCopy` to match this behavior fixing a rare flake where we'd fail the test on the txn retry error when we should've ignored it. Fixes: #160537. Release note: None Co-authored-by: Matt Spilchen <matt.spilchen@cockroachlabs.com> Co-authored-by: Yahor Yuzefovich <yahor@cockroachlabs.com>

craig · 2026-01-10T00:02:04Z

Build failed (retrying...):

local_roachtest

160760: execbuilder: fix a stats-related flake in a new test r=yuzefovich a=yuzefovich Fixes: #160752. Fixes: #160753. Release note: None 160764: sql/copy: fix rare flake in TestLargeCopy r=yuzefovich a=yuzefovich We have automatic retry mechanism for COPY but it can only be used for non-atomic COPY. If we have the atomic COPY and hit a txn retry error, it's bubbled up to the client. We now adjust `TestLargeCopy` to match this behavior fixing a rare flake where we'd fail the test on the txn retry error when we should've ignored it. Fixes: #160537. Release note: None Co-authored-by: Yahor Yuzefovich <yahor@cockroachlabs.com>

craig · 2026-01-10T00:32:21Z

Build failed (retrying...):

unit_tests

160764: sql/copy: fix rare flake in TestLargeCopy r=yuzefovich a=yuzefovich We have automatic retry mechanism for COPY but it can only be used for non-atomic COPY. If we have the atomic COPY and hit a txn retry error, it's bubbled up to the client. We now adjust `TestLargeCopy` to match this behavior fixing a rare flake where we'd fail the test on the txn retry error when we should've ignored it. Fixes: #160537. Release note: None Co-authored-by: Yahor Yuzefovich <yahor@cockroachlabs.com>

craig · 2026-01-10T01:40:49Z

Build failed:

docker_image_amd64

yuzefovich · 2026-01-10T01:46:48Z

bors retry

craig · 2026-01-10T02:31:57Z

Build succeeded:

blathers-crl · 2026-01-10T02:32:05Z

Based on the specified backports for this PR, I applied new labels to the following linked issue(s). Please adjust the labels as needed to match the branches actually affected by the issue(s), including adding any known older branches.

Issue #160537: branch-release-25.3, branch-release-25.4, branch-release-26.1.

_{🦉 Hoot! I am a Blathers, a bot for CockroachDB. My owner is dev-inf.}

blathers-crl · 2026-01-10T02:32:17Z

Encountered an error creating backports. Some common things that can go wrong:

The backport branch might have already existed.
There was a merge conflict.
The backport branch contained merge commits.

You might need to create your backport manually using the backport tool.

error creating merge commit from ae6e38c to blathers/backport-release-25.3-160764: POST https://api.github.com/repos/yuzefovich/cockroach/merges: 409 Merge conflict []

you may need to manually resolve merge conflicts with the backport tool.

Backport to branch 25.3.x failed. See errors above.

error creating merge commit from ae6e38c to blathers/backport-release-25.4-160764: POST https://api.github.com/repos/yuzefovich/cockroach/merges: 409 Merge conflict []

you may need to manually resolve merge conflicts with the backport tool.

Backport to branch 25.4.x failed. See errors above.

_{🦉 Hoot! I am a Blathers, a bot for CockroachDB. My owner is dev-inf.}

yuzefovich requested review from a team and michae2 January 9, 2026 11:29

yuzefovich added backport-25.3.x Flags PRs that need to be backported to 25.3 backport-25.4.x Flags PRs that need to be backported to 25.4 backport-26.1.x Flags PRs that need to be backported to 26.1 labels Jan 9, 2026

michae2 approved these changes Jan 9, 2026

View reviewed changes

craig bot merged commit 7403428 into cockroachdb:master Jan 10, 2026
25 checks passed

celeste-cockroachdb bot added the target-release-26.2.0 label Jan 10, 2026

blathers-crl bot mentioned this pull request Jan 10, 2026

pkg/sql/copy/copy_test: TestLargeCopy failed #160537

Closed

yuzefovich mentioned this pull request Jan 10, 2026

release-26.1: sql/copy: fix rare flake in TestLargeCopy #160807

Merged

blathers-crl bot added the backport-failed label Jan 10, 2026

yuzefovich deleted the fix-copy-flake branch January 10, 2026 02:42

yuzefovich removed backport-failed backport-25.3.x Flags PRs that need to be backported to 25.3 backport-25.4.x Flags PRs that need to be backported to 25.4 labels Jan 10, 2026

DrewKimball mentioned this pull request Jan 28, 2026

pkg/sql/copy/copy_test: TestLargeCopy failed #161631

Closed

celeste-cockroachdb bot added v26.2.0-prerelease and removed target-release-26.2.0 labels Mar 11, 2026

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

sql/copy: fix rare flake in TestLargeCopy#160764

sql/copy: fix rare flake in TestLargeCopy#160764
craig[bot] merged 1 commit intocockroachdb:masterfrom
yuzefovich:fix-copy-flake

yuzefovich commented Jan 9, 2026

Uh oh!

cockroach-teamcity commented Jan 9, 2026

Uh oh!

yuzefovich commented Jan 9, 2026

Uh oh!

michae2 left a comment

Uh oh!

yuzefovich commented Jan 9, 2026

Uh oh!

craig bot commented Jan 10, 2026

Uh oh!

craig bot commented Jan 10, 2026

Uh oh!

craig bot commented Jan 10, 2026

Uh oh!

yuzefovich commented Jan 10, 2026

Uh oh!

craig bot commented Jan 10, 2026

Uh oh!

Uh oh!

blathers-crl bot commented Jan 10, 2026

Uh oh!

blathers-crl bot commented Jan 10, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

yuzefovich commented Jan 9, 2026

Uh oh!

cockroach-teamcity commented Jan 9, 2026

Uh oh!

yuzefovich commented Jan 9, 2026

Uh oh!

michae2 left a comment

Choose a reason for hiding this comment

Uh oh!

yuzefovich commented Jan 9, 2026

Uh oh!

craig bot commented Jan 10, 2026

Uh oh!

craig bot commented Jan 10, 2026

Uh oh!

craig bot commented Jan 10, 2026

Uh oh!

yuzefovich commented Jan 10, 2026

Uh oh!

craig bot commented Jan 10, 2026

Uh oh!

Uh oh!

blathers-crl bot commented Jan 10, 2026

Uh oh!

blathers-crl bot commented Jan 10, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants