Direkt zum Inhalt

VMP8 - canceled backup, vSphere says still in progress. Flattened snapshot, space still used up

Thread needs solution

I have a backup of a server that was frozen at 6% for hours in vmProtect 8. I canceled this job in vmProtect 8 web interface.

In vSphere client it still shows a task at the bottom... Acronis vmProtect Backup In Progress. It says this for hours even after cancelling the task in the gui.

For the server in question it still is taking up snapshot space. So I did a consolidate snapshots in vSphere client GUI. It shows no more snapshots, however there is still 21 GB of space on this VM used for snapshots.

I look and see the hard drive for this VM looks like it is pointed to a snapshot disk as it has -000001.vmdk at the end of its name... even though in snapshot manager I deleted it (was named funny like %20%something).

Now I look in vmProtect 8 and see there is a hard disk 2 attached to _datastorename_servername at 146.70349 GB.

What the heck is going on here? This was supposed to be fixed in vmProtect 8. I know in 7 this happened from time to time and we even had to run vmware converter on one of the servers because for awhile it was left unnoticed and had 57 snapshots (vmware only supports consolidating 32).

0 Users found this helpful
frestogaslorastaswastavewroviwroclolacorashibushurutraciwrubrishabenichikucrijorejenufrilomuwrigaslowrikejawrachosleratiswurelaseriprouobrunoviswosuthitribrepakotritopislivadrauibretisetewrapenuwrapi
Beiträge: 22
Kommentare: 3800

Hi KJSTech,

Such behavior is usually a result from accident crash of the backup daemon or the vmProtect Virtual Appliance itself. The disks main remain attached to appliance only when some kind of crash has occurred. To get rid of this you should go to the settings of the Virtual Appliance and "Delete from disk" that virtual disk which remain attached (this is in fact a link to the original production disk), after that create a simple snapshot of the original VM and delete it right away so that it could consolidate the deltas into the base disk. You can check more details at the following KB article: http://kb.acronis.com/content/32611 . In general there is no need to run VMware converter to get rid of these snapshots.

We investigate each such issue (by analyzing the core dumps) and typically the reasons are different in each particular case so we fix them one by one when they come. Please follow the instructions from the above mentioned article to collect the core dump (/tmp/core) and contact our support team so that we can investigate it. It would also make sense to update the Virtual Appliance to 8184 build (Update 1) where we've fixed several crashes, then delete the /tmp/core file (old dump) to see whether the problem is still reproduced with it - if you discover that the appliance still crashes then a new dump will be created and then we will be able to analyze it and prepare a fix accordingly.

Thank you.
--
Best regards,
Vasily
Acronis vmProtect Program Manager

Ok will do when I get a chance.

I was able to fix it by running a python script on a vmware esx host and then creating a snapshot on 2 affected virtual machines and then consolidating said snapshots. That got them off from running on -0000x.vmdk files.

http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cm…

The one time we did VMWare converter because with an earlier release of VMP7, we had a machine with over 50 snapshots. VMWare said to use the converter because they can't guarantee snapshot consolidation beyond 32 sequential snapshots. Even so, they said the temporary disk space and time involved to perform such a flattening of that many snapshots would be so enormous that VMWare converter was safer and quicker method to fix that issue.

Since then VMP8 has been MUCH more stable and I appreciate all the support over the last few months as well as an active beta program. I think all of the collaboration really helps improve the product and I look forward to a new year ahead with new product opportunities.

New case 01767290 has been created.

I have an archive of core and vmProtect folders and will await tech support to supply me with a link where I can upload it (60MB).

Thank you for your help.

frestogaslorastaswastavewroviwroclolacorashibushurutraciwrubrishabenichikucrijorejenufrilomuwrigaslowrikejawrachosleratiswurelaseriprouobrunoviswosuthitribrepakotritopislivadrauibretisetewrapenuwrapi
Beiträge: 22
Kommentare: 3800

Hi KJSTech,

Thank you for the update. Can you please clarify (just to double-check) whether the crash issue can be reproduced with the latest build (8184) and whether you have updated to this build already? It would make sense at this stage to remove the current /tmp/core file (to get rid of the old one) so that if this crash reoccurs we can pick up the new dump. On our side we will investigate the current dump to see whether the reasons for crash are new ones or something that we've fixed already. I'll get in touch with our support team to ensure proper processing of your ticket.

Thank you.
--
Best regards,
Vasily
Acronis vmProtect Program Manager