Skip to:

  1. Skip to Jira Navigation
  2. Skip to Side Navigation
  3. Skip to Main Content

Resync fails with duplicate keys

Description

In K8SPSMDB-975 our user complains about the following error:

Is it possible to do something about it?

Environment

None

Activity

Show:

Aaditya Dubey February 6, 2024 at 5:26 PM

Hi

Thank you for the report.
Closing the ticket as stated.

Thank you connecting Percona!

Jonathan Lynch September 28, 2023 at 3:26 PM

This ended up being a configuration issue with multiple mongodb clusters backing up to the same S3 bucket. When prefix is omitted the backups overlap, essentially corrupting them. Please close this ticket.

Jonathan Lynch September 20, 2023 at 9:18 PM

The pbmBackups table in the source cluster does not have duplicate entries.

I've tried a few workarounds:

  • drop the index then restore: percona operator re-creates it

  • recreate the index as non-unique then restore: restore immediately fails with IndexOptionsConflict

  • start the restore, wait for it to create the index, then recreate it as non-unique: the restore goes an hour then fails with IndexOptionsConflict. I guess it restores that table twice?

(Between each run I'm deleting and recreating the restore cluster.)

Not a Bug

Details

Assignee

Reporter

Needs QA

Priority

Smart Checklist

Created September 19, 2023 at 6:34 AM
Updated February 6, 2024 at 5:26 PM
Resolved February 6, 2024 at 5:26 PM