Details
-
Bug
-
Resolution: Fixed
-
Critical
-
6.6.0, 6.6.1
-
Untriaged
-
1
-
No
Description
What's the issue?
During a restore, 'cbbackupmgr' may encounter "connection reset by peer" errors when reading from the data store. At the moment these errors are causing the restore to fail because they are not actively retried by the AWS SDK.
What's the fix?
We should add some sort of retry reader functionality into restore which will keep track of the state of the download and retry the API request in the event of a failure due to the connection being reset by S3.