CentOS VM crashes - vmtoolsd

[Josh26]

Hi,

I have a CentOS (effectively RedHat) server running ESXi 5.1. It crashes during backup (snapshot) operations with this error at console:

INFO: task vmtoolsd:1213 blocked for more than 120 seconds.

"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.

RedHat has a knowledge base article about it here: https://access.redhat.com/site/solutions/442043

For those with a login to see the official "resolution", it is: "Contact vendor of "vmtoolsd" utility ( VMware ) and take their opinion on this issue."

The host was a 5.1  GA however I've just applied all ESXi patches as of today and installed the latest tools available from that, and it does not appear to have resolved the problem. Any assistance on this issue that RedHat feel is VMware related appreciated.

[vlho]

Hi Josh26,

maybe this is solution for you:

http://blog.ronnyegner-consulting.de/2011/10/13/info-task-blocked-for-more-than-120-seconds

[anayrat01]

Hi,

I have the same issue with vmware ESX 5.5 U1, Centos 6.5 (up to date) and VEEAM Backup (7.0 R2 - 7.0.0.871).

When veeam do a snapshot with quiesce option the VM is freezed. It respond to ping but we can't login, as if the filesystem was freezed but not released. I have several vm with the same version of vmware tools (provided from vmware repository : vmware repos) and there is no error.

When I do a manual snapshot with the option "quiesce file system" the snapshot success.

Edit : When I do a manual snapshot WITHOUT "quiesce file system" the vm crash.

Did you solved your issue?

[IlyaSkorik]

Any solution there?

[IlyaSkorik]

Investigation of the problem revealed, that there is a problem between newer versions of vmware-tools and linux 2.6 kernel.

For example, it is about of all centos  6.x / redhat 6.x versions.

To me helped the following decision:

Add to file "/etc/vmware-tools/tools.conf" this:

[vmbackup]

enableSyncDriver=false

Or downgrade vmware-tools to 5.0

Or upgrade to Centos7.x

[anayrat01]

We try an to kernel 3.10 with elrepos,

[sguk]

I had a similar problem to this thread and enableSyncDriver=false stopped the vm crashing.

However, with enableSyncDriver=false does the quiescing still happen? Or do we need to use pre freeze / post thaw scripts to achieve application consistent backups. I can't find a definitive answer online or the scripts themselves.

Apologies for digging this thread up, I appreciate any advice offered.

Thanks