-
Notifications
You must be signed in to change notification settings - Fork 3.8k
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
sql/opt/exec/execbuilder: TestExecBuild failed #72802
Comments
sql/opt/exec/execbuilder.TestExecBuild failed with artifacts on master @ 52a7a6f10a3ed02d7d4d0b7dc09406c86ba937dd:
HelpSee also: [How To Investigate a Go Test Failure \(internal\)](https://cockroachlabs.atlassian.net/l/c/HgfXfJgM)Parameters in this failure:
|
sql/opt/exec/execbuilder.TestExecBuild failed with artifacts on master @ 1a078671083946342ecf610a6fb899df2d1783d3:
HelpSee also: [How To Investigate a Go Test Failure \(internal\)](https://cockroachlabs.atlassian.net/l/c/HgfXfJgM)Parameters in this failure:
|
sql/opt/exec/execbuilder.TestExecBuild failed with artifacts on master @ 0b9208b8e293a475d5a4c940b493c8414dfa95ea:
HelpSee also: [How To Investigate a Go Test Failure \(internal\)](https://cockroachlabs.atlassian.net/l/c/HgfXfJgM)Parameters in this failure:
|
This has happened before: #70108 |
Based on the variation of the failures it seems most likely we're running these tests in metamorphic mode even though the test specifies !metamorphic. |
Near as I can tell this was the command from teamcity:
That should be engaging metamorphic and so the show_trace_nonmetamorphic test should be skipped I think. |
Okay the max batch size metamorphism should be high-jacked by forceProductionBatchSizes which should be true for these tests so again I don't see why the batch size limit is seemingly being randomized. |
sql/opt/exec/execbuilder.TestExecBuild failed with artifacts on master @ 28bb1ea049da5bfb6e15a7003cd7b678cbc4b67f:
HelpSee also: [How To Investigate a Go Test Failure \(internal\)](https://cockroachlabs.atlassian.net/l/c/HgfXfJgM)Parameters in this failure:
|
Previously because of Parallel usage we would be writing serverArgs.forceProductionBatchSizes from multiple go routines. Avoid that with a separate copy for the nonMetamorphic tests. Fixes cockroachdb#72802 Release note: None
72957: roachprod: introduce Node type r=RaduBerinde a=RaduBerinde There is a lot of confusion in the code around node "indexes". Sometimes these are a (0-based) index in the target list of nodes, sometimes they are a (1-based) node ID. This makes the code harder to follow and more error-prone (in fact, there are a couple of places where we convert from one to the other incorrectly). This change introduces the types `Node` and `Nodes` and switches code to use `Node` (rather than the 0-based index) whenever possible. This enlists the compiler's help, as it is no longer legal to implicitly convert from one to the other. Release note: None 72967: logic: fix racy forceProductionBatchSizes bool r=cucaroach a=cucaroach Previously because of Parallel usage we would be writing serverArgs.forceProductionBatchSizes from multiple go routines. Avoid that with a separate copy for the nonMetamorphic tests. Fixes #72802 Release note: None 72994: bazel: have bazelisk pull bazel binaries from our fork r=rail a=rickystewart This will result in Bazel binaries being pulled from https://github.com/cockroachdb/bazel/releases/tag/4.2.1 rather than upstream. Release note: None Co-authored-by: Radu Berinde <[email protected]> Co-authored-by: Tommy Reilly <[email protected]> Co-authored-by: Ricky Stewart <[email protected]>
Temporary fix for cockroachdb#72802 Release note: None
Temporary fix for cockroachdb#72802 Due to cockroachdb#73876 these tests have become flakey. With the disable-span-configs option 70 runs of make stress on the opt logic tests pass. Release note: None
I didn't mean for this to be closed. Should only close after #75282 is resolved. |
75261: row: fetcher cleanup and improvements r=RaduBerinde a=RaduBerinde #### rowenc: remove deprecated return from DecodeIndexKey Release note: None #### row: minor cleanup around foundNull This moves around some code to make it clear that it's only relevant in a specific case. Release note: None #### row: clean up ReadIndexKey Renaming to DecodeIndexKey and removing return value which is no longer useful. Release note: None #### row: more Fetcher cleanup - improve comment for `indexKey`; - unexport NextKey; - slightly change the return value of nextKey to simplify the logic (the semantic difference is what the first call returns, which is not used); - use numKeysPerRow instead of counting the total families; this enables the faster paths for more cases. Release note: None 75272: kvserver: de-flake TestStoreSplitRangeLookupRace r=irfansharif a=irfansharif Fixes #75198. This test was a bit brittle in expecting only one kind of range lookup request in a testing filter -- it was always possible to intercept a ReverseScanRequest, and after enabling span configs (#73876), we now have an internal query ("validate-span-cfgs") that makes use of it. See #75198 for more details. Release note: None 75281: sql: disable span-config on flakey 5node tests r=cucaroach a=cucaroach Temporary fix for #72802 and 5node/distsql_enum CI failures. Due to #73876 these tests have become flakey. With the disable-span-configs option 70 runs of make stress on the opt logic tests pass. Release note: None Co-authored-by: Radu Berinde <[email protected]> Co-authored-by: irfan sharif <[email protected]> Co-authored-by: Tommy Reilly <[email protected]>
Temporary fix for cockroachdb#72802 Due to cockroachdb#73876 these tests have become flakey. With the disable-span-configs option 70 runs of make stress on the opt logic tests pass. Release note: None
@cucaroach BTW, I think you meant to reference #74933 earlier. The failure in this issue looks like something else (though still TestExecBuild), and before #73876. |
Yes I got my signals crossed. This issue was fixed by #72967 |
sql/opt/exec/execbuilder.TestExecBuild failed with artifacts on master @ 6e288820935624d0839406a0a8bbacaff8fe960d:
Help
See also: [How To Investigate a Go Test Failure \(internal\)](https://cockroachlabs.atlassian.net/l/c/HgfXfJgM)Parameters in this failure:
This test on roachdash | Improve this report!
The text was updated successfully, but these errors were encountered: