Interesting Crash During Backup

Erik    Feb 17 11:02PM 2017

[root@esxi:/opt/verticalbackup] ./vertical backup Gitlab
Vertical Backup 0.3.0
Licensed to (Beta Testing); expires on 2018-02-06
Storage set to /vmfs/volumes/vmware_nfs/backup
Listing all virtual machines
Backing up Gitlab, id: 17, vmx path: /vmfs/volumes/SSD/gitlab/gitlab.vmx, guest os: centos64Guest
No previous backup found.  Listing existing chunks
Virtual machine Gitlab is powered on
Removing all snapshots of Gitlab
Removing all snapshots of Gitlab
Traceback (most recent call last):
  File "", line 135, in 
  File "/root/verticalbackup/build/vertical_main/out00-PYZ.pyz/vertical_backup", line 504, in backupCommand
  File "/root/verticalbackup/build/vertical_main/out00-PYZ.pyz/vertical_backup", line 562, in backupVM
  File "/root/verticalbackup/build/vertical_main/out00-PYZ.pyz/vertical_backup", line 127, in findInFile
IOError: [Errno 16] Device or resource busy: '/vmfs/volumes/SSD/gitlab/gitlab_1.vmdk'

gchen    Feb 18 11:43AM 2017

This is because when vim-cmd is used to create a snapshot, it sometimes returns before the snapshot is ready (or more precisely, before the handle of the base vmdk disk file is released). On the next backup run usually this error will go away. If the problem persists, try to consolidate the snapshots in vSphere Client.

I'll roll out a fix in which it will wait until the vmdk file is readable.

Erik    Feb 18 1:16PM 2017

Makes sense. Thanks for the quick response!

gchen    Feb 24 10:42AM 2017

This bug has been fixed in version 0.3.3.

Log in to comment
Copyright © Acrosync LLC 2017