Coalesce failed, skipping - corrupted

gojjily

My Xen orchestra backup is failing for one specific VM with the error below.
How do I fix this?

Jul 28 01:28:51 milsrv01 SMGC: [30191] Coalesce candidate: *c5bf9937(350.000G/105.039G) (tree height 6)
Jul 28 01:28:51 milsrv01 SMGC: [30191] Coalescing *c5bf9937(350.000G/105.039G) -> *29153ed4(350.000G/214.587G)
Jul 28 01:28:51 milsrv01 SMGC: [30191] coalesce: EXCEPTION <class 'util.SMException'>, VHD *c5bf9937(350.000G/105.039G) corrupted
Jul 28 01:28:51 milsrv01 SMGC: [30191]   File "/opt/xensource/sm/cleanup.py", line 1753, in coalesce
Jul 28 01:28:51 milsrv01 SMGC: [30191]     self._coalesce(vdi)
Jul 28 01:28:51 milsrv01 SMGC: [30191]   File "/opt/xensource/sm/cleanup.py", line 1942, in _coalesce
Jul 28 01:28:51 milsrv01 SMGC: [30191]     vdi._doCoalesce()
Jul 28 01:28:51 milsrv01 SMGC: [30191]   File "/opt/xensource/sm/cleanup.py", line 764, in _doCoalesce
Jul 28 01:28:51 milsrv01 SMGC: [30191] Coalesce failed, skipping

olivierlambert

This is a storage issue, you VHD chain seems to be corrupted sadly.

gojjily

right, but what are we saying the options are to fix this?

olivierlambert

There's no universal answer. You might try with the vhd-util repair tool, but if it's very corrupted, restoring a backup is the best approach.

Corruption doesn't appear randomly, something went wrong at some point, but it's impossible to guess without more investigation.

gojjily

@olivierlambert I would guess it was a power failure. I had an APC UPS blow up, still working to troubleshoot the UPS issue. APC send a new warranty battery, and now also a new warranty UPS.

gojjily

@olivierlambert so I went to restore this VM from backup, and noticed that the monthly backup is currently running on this VM as we speak.

Will restoring from this newest backup fix my issue? Or does it restore the issue back into production?