#1 2020-10-12 11:37:50

AndersG
Member
Registered: 2020-01-16
Posts: 25

Backup of one VM fails, every other day.

This popped up as I am adding VMs to the backup job. It is always the same VM and the error ocurrs every other day:

Last error raised for the above VM:ERROR DIFQMSH4, details: [Filr-3.0] error: first 500M mistmatch [Filr-3.0_1-flat.vmdk]

• [ Mon Oct 12 05:11:14 UTC 2020 ] ERROR (DIFDELAL), details [Filr-3.0] error: error deleting all snapshots VM [Filr-3.0_XSIBAK] Id 15, error: Remove All Snapshots:• [ Mon Oct 12 05:11:22 UTC 2020 ] ERROR (DIFQMSH4), details [Filr-3.0] error: first 500M mistmatch [Filr-3.0-flat.vmdk]
• [ Mon Oct 12 05:11:30 UTC 2020 ] ERROR (DIFQMSH4), details [Filr-3.0] error: first 500M mistmatch [Novell Filr-2_1-flat.vmdk]
• [ Mon Oct 12 05:11:38 UTC 2020 ] ERROR (DIFQMSH4), details [Filr-3.0] error: first 500M mistmatch [Filr-3.0_1-flat.vmdk]
• Remove all snapshots failed Command: vim-cmd vmsvc/snapshot.removeall 15 2>&1 • [ Mon Oct 12 05:11:38 UTC 2020 ] WARNING (DIFRMVMX), details [Filr-3.0] measure DIFRMVMX: took measure, renamed remote .vmx file to reinitialize OneDiff

Any idea what may cause this?

XSIBACKUP-PRO 11.2.19
ESXi 6.7.0 build-14320388

I saw a post with a similar error and I checked that both machines run the same version of VMWare but the source has VMFS6 and the target VMFS5, could that be the issue?

Offline

#2 2020-10-13 13:47:24

admin
Administrator
Registered: 2017-04-21
Posts: 1,386

Re: Backup of one VM fails, every other day.

It could be a worn out disk raising some read error in some sector. The best part of post backup checksums is that you really know what's going on, the worst part is that when you know you start to wonder.

If you recover that VM you will, most probably, find out that it works. This is due to read errors being more frequent than write errors, thus the probability that you are getting a read error on some byte is higher than that of the byte having been written wrongly.

This might not justify changing the disk, so you have these options:

1 - Ignoring the error. You should first check the restored VM.
2 - Clone the disk that's causing the problem to a new virtual disk file.
3 - Change the disk.

Read this post to have an overview on why these errors are detected and how to deal with them:
https://33hops.com/xsibackup-disk-check … ption.html

Offline

#3 2020-10-14 06:00:30

AndersG
Member
Registered: 2020-01-16
Posts: 25

Re: Backup of one VM fails, every other day.

OK. It is a raidset, but I will read that document you linked to and see what can be done. I will change the datastore to VMFS6 anyway an incoming weekend. The target server is the old server and it will just be used for backup and archival purposes.

Offline

Board footer