Hello everyone!
We have encountered a problem with restoring data from backup. I will describe our actions:
-
Started restoring full backup on mongo 5.0.4 and pbm 1.6.1. Restore was successful, after that everything worked normally.
-
Next we ran the restor of PITR. The restor passed without errors, but after the restor we started getting read/write errors in fs.chunks.
Errors like:
“t“:{”$date“: ‘2024-06-20T17:54:51.526+00:00’}, ‘s’: ‘I’, ‘c’: ‘QUERY’, ‘id’:4625501, ‘ctx’: ‘conn43’, ‘msg’: ‘Unable to establish remote cursors’, ‘attr’:{”error“:{”code":13388, ‘codeName’: ‘StaleConfig’, ‘errmsg’: ‘sharding status of collection app_db.fs.chunks is not currently available for description and needs to be recovered from the config server’, ‘ns’: ”app_db. fs.chunks“,”vReceived“:{”$timestamp“:{”t“:0,”i“:0}},”vReceivedEpoch“ {”$oid“:”000000000000000000000000“},”shardId“:”rs03“},”nRemotes":0}}}
-
Tried restarting the whole cluster. It didn’t help.
-
Tried performing flushRouterConfig on all mongos. The errors remained.
-
Tried restoring full backup. Restore was successful, but the errors remained.
-
Updated mongo to 5.0.27, pbm to 1.7.0.
-
Restarted the entire cluster and ran restore full backup. The errors remained.
-
We updated pbm to 2.5.0. Made a new full backup. Restored this backup. The errors remained.
-
Restarted the whole cluster and performed flushRouterConfig. The errors remained.
What ideas do you have for solving this problem? Maybe we followed the wrong steps and left something out?