importccl: remove protected timestamp from IMPORT INTO #68043
Merged
Conversation
This file contains 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
adityamaru
requested review from
a team,
pbardea and
dt
and removed request for
a team
|
This change is |
pbardea
reviewed
Jul 26, 2021
| @@ -1940,18 +1891,6 @@ func (r *importResumer) Resume(ctx context.Context, execCtx interface{}) error { | |||
| details := r.job.Details().(jobspb.ImportDetails) | |||
| files := details.URIs | |||
| format := details.Format | |||
| ptsID := details.ProtectedTimestampRecord | |||
There was a problem hiding this comment.
Do we need to keep the cleanup around to free protections reserved by jobs on older binaries?
There was a problem hiding this comment.
good catch, readded the release logic. The helper we use to release has a nil check for details.ProtectedTimestampRecord, so for a 21.2 node this will just be a noop.
Previously, IMPORT INTO would lay a protected timestamp on the table span of the table being imported into. This would prevent all keys in this span from being GC'ed during IMPORT execution. IMPORT INTO however does not allow ingesting keys that shadow already existing keys. Additionally, during the IMPORT the table is taken offline and so it does not serve any online traffic. These two properties of IMPORT INTO ensure that the old keys (not ingested during IMPORT) will never become gc'eable during the IMPORT job. Thus, writing a protected timestamp was redundant. If the IMPORT job fails we issue a RevertRange to revert all the newly ingested keys. This RevertRange could attempt to revert to a time below the GC threshold but following cockroachdb#61004 this is alright for the same reason explained above. Fixes: cockroachdb#67924 Release note: None
IMPORT INTO no longer relies on protected timestamps and so it can be run within a tenant. Release note: None
adityamaru
force-pushed
the
remove-protected-ts-from-import
branch
from
9b68682
to
9ee5229
Compare
pbardea
approved these changes
Jul 28, 2021
|
TFTR! bors r=pbardea |
|
Build succeeded: |
rmloveland
added a commit
to cockroachdb/docs
that referenced
this pull request
Mar 13, 2023
rmloveland
added a commit
to cockroachdb/docs
that referenced
this pull request
Apr 13, 2023
mdlinville
pushed a commit
to cockroachdb/docs
that referenced
this pull request
Apr 14, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Previously, IMPORT INTO would lay a protected timestamp
on the table span of the table being imported into.
This would prevent all keys in this span from being GC'ed
during IMPORT execution.
IMPORT INTO however does not allow ingesting keys that
shadow already existing keys. Additionally, during the
IMPORT the table is taken offline and so it does not serve
any online traffic. These two properties of IMPORT INTO
ensure that the old keys (not ingested during IMPORT) will
never become gc'eable during the IMPORT job. Thus, writing
a protected timestamp was redundant.
If the IMPORT job fails we issue a RevertRange to revert
all the newly ingested keys. This RevertRange could
attempt to revert to a time below the GC threshold but
following #61004 this is alright for the same reason
explained above.
Fixes: #67924
IMPORT INTO no longer relies on protected timestamps and
so it can be run within a tenant.