Bareos-dir outputs errors when connecting to S3 backend

[Dmitry Ponkin]

After about 2 months of perfect operation, Bareos-sd (s3 buckets via bareos-storage-droplet) started outputting errors during most jobs. The instance (CentOS 7, Bareos 19.2.7 packages from the official Bareos repository) was unattended and nothing preceded this issue.

bareos-sd JobId 1513: Please mount append Volume "VolumeName" or label a new one for:

Job: JobName

Storage: "StorageName" (AWS_S3_Storage)

Pool: PoolName

Media type: S3_Object1

After the above, it outputs the following line 15 times:

bareos-sd JobId 1513: Warning: stored/mount.cc:275 Open device "StorageName" (AWS_S3_Storage) Volume "VolumeName" failed: ERR=stored/dev.cc:747 Could not open: AWS_S3_Storage/VolumeName, ERR=Success

And then proceeds with the job without any issues. The volume gets written into, the bucket receives the data, nothing is out of order.

Non-fatal FD errors: 0
SD Errors: 0
FD termination status: OK
SD termination status: OK
Termination: Backup OK  

I did the rounds of restarting everything and checking the configuration, ensuring that there are no network issues. I also checked to see if disabling recycling (and thus forcing new volumes to be created) changed anything, it didn't. 

I don't understand the issue as everything seems to work just fine, apart from my email getting a dozen of "Intervention needed" messages every night. What am I missing here?

Thank you.

[Earbuds Audio]

Hi, 

When i Got this error, it was due to timezone difference between the AWS S3 storage and my server where bareos-sd is running. 

You can try to synchronize your server if possible with the timezone of S3.