Michael Butler
Michael Butler
Backport 1/1 commits from #87312. /cc @cockroachdb/release --- Currently RESTORE may restore invalid backup data from a backed up table that underwent an IMPORT rollback. See https://github.com/cockroachdb/cockroach/issues/87305 for a detailed...
To reproduce, stress the test on top of https://github.com/cockroachdb/cockroach/pull/88271, with DisableTenantTestTenant set to false. See test failure artifacts [here](https://teamcity.cockroachdb.com/viewLog.html?buildId=6543458&tab=buildResultsDiv&buildTypeId=Cockroach_BazelEssentialCi). Fresh build reveals more subtests hit the same panic: https://teamcity.cockroachdb.com/viewLog.html?buildId=6576285&buildTypeId=Cockroach_BazelEssentialCi Jira...
Michael Butler (msbutler) commented: Clarify that MR databases can only be restored in MR enabled serverless clusters. Note that the `skip_localities_check` restore flag is not a workaround. (The reason this...
Release note: none Epic: none
Epic: none Release note: none
This patch adds a rangefeed client level configuration option that configures the rangefeed to emit events originally written by clusters with a matching OriginID during LogicalDataReplication. Currently rangefeeds only support...
Michael Butler (msbutler) commented: Currently there isn't any warning in the docs around decomissioning multiple nodes at once. It seems like best practice drain/decomission nodes one at a time. (...
Clarify in the Backup Validation section that regular restore runs checksums as well. We seem to suggest h[ere](https://www.cockroachlabs.com/docs/stable/backup-validation#validate-backup-table-data-is-restorable) that `verify_with_table_data` runs a special checksum. we should clarify that regular old...
We observed in #147759 that backup's use of an RDI can degrade the foreground workload. We suspect that integrating elastic admission control into the RDI could prevent this. Additional slack...