sql/vecindex/vecstore: TestInMemoryLock failed #136958
Labels
branch-master
Failures and bugs on the master branch.
C-test-failure
Broken test (automatically or manually discovered).
O-robot
Originated from a bot.
release-blocker
Indicates a release-blocker. Use with branch-release-2x.x label to denote which branch is blocked.
T-sql-queries
SQL Queries Team
Comments
cockroach-teamcity
added
branch-master
|
sql/vecindex/vecstore.TestInMemoryLock failed on master @ d7ea85402dc35e36c6cc35520fa91f25fd5c999d: Parameters:
|
|
sql/vecindex/vecstore.TestInMemoryLock failed on master @ 74e83fa45e549680ed0e1ce8917468701bea7794: Parameters:
|
|
sql/vecindex/vecstore.TestInMemoryLock failed on master @ 5f83714a8111d0ea9ca8b34a72b2f5d97ea56b53: Parameters:
|
|
sql/vecindex/vecstore.TestInMemoryLock failed on master @ d18eb683b2759fd8814dacf0baa913f596074a17: Parameters:
|
|
sql/vecindex/vecstore.TestInMemoryLock failed on master @ 49cff91f3501494deaf038671bc643c194a0e3ca: Parameters:
|
|
sql/vecindex/vecstore.TestInMemoryLock failed on master @ fe048f6e72edca59b4367463db1d0d00783f08ab: Parameters:
|
|
sql/vecindex/vecstore.TestInMemoryLock failed on master @ 9354770c7c6eb5a89437068d8c6a4accf8031b67: Parameters:
|
|
sql/vecindex/vecstore.TestInMemoryLock failed on master @ a653b4e4e6483cec7d65808ba4d55d8c63747a6e: Parameters:
|
craig bot
pushed a commit
that referenced
this issue
Dec 17, 2024
137105: sql/parser: Add new SQL grammar for row-level security (RLS) r=spilchen a=spilchen
This change introduces new SQL grammar to support row-level security (RLS). All of the statements, with the exception of SHOW POLICIES match the postgres grammar. The added statements and options are as follows:
```
DROP POLICY [ IF EXISTS ] name ON table_name [ CASCADE | RESTRICT ]
CREATE POLICY name ON table_name
[ AS { PERMISSIVE | RESTRICTIVE } ]
[ FOR { ALL | SELECT | INSERT | UPDATE | DELETE } ]
[ TO { role_name | PUBLIC | CURRENT_USER | SESSION_USER } [, ...] ]
[ USING ( using_expression ) ]
[ WITH CHECK ( check_expression ) ]
ALTER POLICY name ON table_name RENAME TO new_name
ALTER POLICY name ON table_name
[ TO { role_name | PUBLIC | CURRENT_USER | SESSION_USER } [, ...] ]
[ USING ( using_expression ) ]
[ WITH CHECK ( check_expression ) ]
ALTER TABLE … DISABLE ROW LEVEL SECURITY
ALTER TABLE … ENABLE ROW LEVEL SECURITY
ALTER TABLE … FORCE ROW LEVEL SECURITY
ALTER TABLE … NO FORCE ROW LEVEL SECURITY
CREATE ROLE name [[ WITH ] [BYPASSRLS | NOBYPASSRLS]]
ALTER ROLE role_specification [ WITH ] [BYPASSRLS | NOBYPASSRLS]
SHOW POLICIES FOR <table_name>
```
Currently, attempting to use any of the new statements or options will return an unimplemented error. Documentation for this grammar is temporarily omitted and will be added closer to the feature's release, expected in version 25.2.
Epic: CRDB-11724
Closes: #136692
Release note: None
137363: teamcity: add script to run PGO build and collect profiles r=rail a=rickystewart
This script ties together the pieces of the workflow: start a TC job, wait for it to finish, downloads the artifacts, parses the profiles, merges them, and produces a final result.
Part of: CRDB-44692
Epic: CRDB-41952
Release note: None
137374: vecstore: disable deadlock linting for inMemoryLock r=drewkimball a=andy-kimball
Do not use syncutil.RWMutex in the inMemoryLock class, because deadlock detection reports spurious failures. Different partitions in the vector index can be locked in different orders by merge, split, format and other operations. In all these cases, we first acquire the in-memory store's structure lock to prevent deadlocks. But the deadlock detection package is not smart enough to realize this and reports false positives.
Epic: CRDB-42943
Fixes: #136958.
Fixes: #136960.
Release note: None
137448: sql/types, sql/stats: fix SQLStringFullyQualified for arrays of UDTs r=rytaft a=michae2
Starting in v24.2 we added `SQLStringFullyQualified`, but it looks like it didn't work for arrays of user-defined types. This is only used in a few places, including the output of `SHOW STATISTICS USING JSON`.
This commit:
1. fixes `SQLStringFullyQualified` for arrays of UDTs
2. adds the `t.TypeMeta.Name == nil` guard to `SQLString` for composite UDTs
3. sets a formatting flag in `stats.(*JSONStatistic).SetHistogram` so that we get fully-qualified type annotations on histogram upper bounds
No release note because the output of `SHOW STATISTICS USING JSON` isn't documented.
Fixes: #137443
Release note: None
137568: distsql: eliminate ctx allocation in setupFlow r=mgartner a=mgartner
The `ctx` parameter of `setupFlow` is no longer captured by closures,
preventing it from being heap allocated.
Epic: None
Release note: None
137613: orchestration: released CockroachDB version 24.3.1. Next version: 24.3.2 r=rail a=cockroach-teamcity
Release note: None
Epic: None
Release justification: non-production (release infra) change.
Co-authored-by: Matt Spilchen <[email protected]>
Co-authored-by: Ricky Stewart <[email protected]>
Co-authored-by: Andrew Kimball <[email protected]>
Co-authored-by: Michael Erickson <[email protected]>
Co-authored-by: Marcus Gartner <[email protected]>
Co-authored-by: Justin Beaver <[email protected]>
craig bot
pushed a commit
that referenced
this issue
Dec 17, 2024
137374: vecstore: disable deadlock linting for inMemoryLock r=drewkimball a=andy-kimball Do not use syncutil.RWMutex in the inMemoryLock class, because deadlock detection reports spurious failures. Different partitions in the vector index can be locked in different orders by merge, split, format and other operations. In all these cases, we first acquire the in-memory store's structure lock to prevent deadlocks. But the deadlock detection package is not smart enough to realize this and reports false positives. Epic: CRDB-42943 Fixes: #136958. Fixes: #136960. Release note: None 137448: sql/types, sql/stats: fix SQLStringFullyQualified for arrays of UDTs r=rytaft a=michae2 Starting in v24.2 we added `SQLStringFullyQualified`, but it looks like it didn't work for arrays of user-defined types. This is only used in a few places, including the output of `SHOW STATISTICS USING JSON`. This commit: 1. fixes `SQLStringFullyQualified` for arrays of UDTs 2. adds the `t.TypeMeta.Name == nil` guard to `SQLString` for composite UDTs 3. sets a formatting flag in `stats.(*JSONStatistic).SetHistogram` so that we get fully-qualified type annotations on histogram upper bounds No release note because the output of `SHOW STATISTICS USING JSON` isn't documented. Fixes: #137443 Release note: None 137568: distsql: eliminate ctx allocation in setupFlow r=mgartner a=mgartner The `ctx` parameter of `setupFlow` is no longer captured by closures, preventing it from being heap allocated. Epic: None Release note: None Co-authored-by: Andrew Kimball <[email protected]> Co-authored-by: Michael Erickson <[email protected]> Co-authored-by: Marcus Gartner <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
sql/vecindex/vecstore.TestInMemoryLock failed on master @ ccaed974f75032f1b908bca2d42145d21d0092ca:
Parameters:
attempt=1deadlock=truerun=1shard=1Help
See also: How To Investigate a Go Test Failure (internal)
This test on roachdash | Improve this report!
Jira issue: CRDB-45316
The text was updated successfully, but these errors were encountered: