sql/jobs: Add SHOW JOBS ... WITH RESOLVED TIMESTAMP #159068
Merged
+78
−2
Conversation
This file contains hidden or 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
Member
|
This change is |
Contributor
Author
|
msbutler
approved these changes
Dec 9, 2025
Collaborator
msbutler
left a comment
msbutler
left a comment
There was a problem hiding this comment.
Could you add a one liner test? could tack on to an LDR job test
Release note (sql change): The WITH RESOLVED TIMESTAMP option can be passed to SHOW JOBS or SHOW JOB to include the resolved timestamp, if any, for the job in the output columns. Epic: none.
Contributor
Author
Fun fact: |
Contributor
Author
|
known cdc flake. bors r+ |
craig bot
pushed a commit
that referenced
this pull request
Dec 9, 2025
155953: kvserver: simplify rewriteRaftState r=arulajmani a=pav-kv Keys in the RangeID-local unreplicated space have both state machine and raft state. With separated engines, these will reside in different engines, and are also interleaved in an unfortunate way. So we can't use one `ClearRawRange` to cover them all, or even one per engine. This PR clears all unreplicated keys in `rewriteRaftState` manually, so that we can control which goes to which engine/`Writer`. Additionally, this uncovers that half of these keys don't need to be cleared at all because they don't exist, or don't need to be changed. The end effect of the PR is that `rewriteRaftState` now only mutates the raft engine state, which is very convenient. Incidentally, this function finally lives up to its name. Part of #152845 Epic: CRDB-55220 158335: jobs/metricspoller: zero out metrics on exit r=dt a=dt Typically only the node actively running the poller is expected to export non-zero values for poller-maintained metrics; if a node is not *actively* running the poller it should export zeros, not stale values last set when it was last running it. This change adds on-exit calls to each of the tasks in the poller to let them zero out the metrics they each are responsible for updating when running. Release note: none. Epic: none. 158390: sql: support table storage params change in the declarative schema changer r=rafiss a=shghasemi ### sql: support table storage params change in the declarative schema changer Previously, storage parameters were set and reset using the legacy schema changer. With this change, setting most non-backfilling table storage params will use the declarative schema changer. Setting ttl_expire_after, schema_locked, and infer_rbr_region_col_using_constraint is not implemented yet. ### schemachanger: handle RESET storage param in declarative schema changer This reuses the logic that was added to handle SET. The only difference is that RESET always drops the StorageParam element. ### schemachanger: handle manual schema_locked setting in declarative This storage parameter is special, so this requires logic to use the pre-existing TableSchemaLocked element rather than the normal storage param element. As part of this, we partially revert bc3f228 and refactor the check to be a little simpler since we don't actually need to handle the mixed version case now. Epic CRDB-31281 Fixes #155990 Release note: None 159067: util/ioctx: add RandomAccessReader for cloud storage r=sravotto a=sravotto Add a general-purpose wrapper that provides ReadAt/Seek support for any cloud storage reader using an opener factory function pattern. This enables random access (required by formats like Parquet) without requiring the cloud storage implementation itself to natively support ReadAt/Seek. Instead, it opens new readers at specific offsets for each ReadAt call. Key features: - Thread-safe ReadAt: Each call opens a new connection, allowing concurrent reads from different offsets - Seek support: Tracks position logically without seeking underlying readers - Generic: Works with any ReadCloserCtx via OpenerAtFunc factory Release note: None Epic: CRDB-23802 159068: sql/jobs: Add SHOW JOBS ... WITH RESOLVED TIMESTAMP r=dt a=dt Release note (sql change): The WITH RESOLVED TIMESTAMP option can be passed to SHOW JOBS or SHOW JOB to include the resolved timestamp, if any, for the job in the output columns. Epic: none. Co-authored-by: Pavel Kalinnikov <pavel@cockroachlabs.com> Co-authored-by: David Taylor <davidt@davidt.io> Co-authored-by: Shadi Ghasemitaheri <shadi.ghasemitaheri@cockroachlabs.com> Co-authored-by: Rafi Shamim <rafi@cockroachlabs.com> Co-authored-by: Silvano Ravotto <silvano@cockroachlabs.com>
Contributor
|
Build failed (retrying...): |
craig bot
pushed a commit
that referenced
this pull request
Dec 9, 2025
158335: jobs/metricspoller: zero out metrics on exit r=dt a=dt Typically only the node actively running the poller is expected to export non-zero values for poller-maintained metrics; if a node is not *actively* running the poller it should export zeros, not stale values last set when it was last running it. This change adds on-exit calls to each of the tasks in the poller to let them zero out the metrics they each are responsible for updating when running. Release note: none. Epic: none. 158390: sql: support table storage params change in the declarative schema changer r=rafiss a=shghasemi ### sql: support table storage params change in the declarative schema changer Previously, storage parameters were set and reset using the legacy schema changer. With this change, setting most non-backfilling table storage params will use the declarative schema changer. Setting ttl_expire_after, schema_locked, and infer_rbr_region_col_using_constraint is not implemented yet. ### schemachanger: handle RESET storage param in declarative schema changer This reuses the logic that was added to handle SET. The only difference is that RESET always drops the StorageParam element. ### schemachanger: handle manual schema_locked setting in declarative This storage parameter is special, so this requires logic to use the pre-existing TableSchemaLocked element rather than the normal storage param element. As part of this, we partially revert bc3f228 and refactor the check to be a little simpler since we don't actually need to handle the mixed version case now. Epic CRDB-31281 Fixes #155990 Release note: None 159068: sql/jobs: Add SHOW JOBS ... WITH RESOLVED TIMESTAMP r=dt a=dt Release note (sql change): The WITH RESOLVED TIMESTAMP option can be passed to SHOW JOBS or SHOW JOB to include the resolved timestamp, if any, for the job in the output columns. Epic: none. Co-authored-by: David Taylor <davidt@davidt.io> Co-authored-by: Shadi Ghasemitaheri <shadi.ghasemitaheri@cockroachlabs.com> Co-authored-by: Rafi Shamim <rafi@cockroachlabs.com>
Contributor
|
Build failed (retrying...):
|
Contributor
|
Build succeeded: |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
Release note (sql change): The WITH RESOLVED TIMESTAMP option can be passed to SHOW JOBS or SHOW JOB to include the resolved timestamp, if any, for the job in the output columns.
Epic: none.