OpenVZ Forum


Home » General » Support » Getting error when trying to restart VMs
Getting error when trying to restart VMs [message #50033] Tue, 23 July 2013 22:18 Go to next message
nolageek is currently offline  nolageek
Messages: 4
Registered: July 2013
Location: United States
Junior Member
While I don't know if it's related to r1soft, we're having an issue on openvz (using openvz web panel) after installing r1soft agent.

We are not able to restart or start any of the VMs without getting an popup error "Cannot change the virtual server state" and the error in the log is:

Jul 23 10:29:23 vz kernel: [370670.831802] CT: 5: stopped
Jul 23 10:29:23 vz kernel: [370670.832026] CT: 5: failed to start with err=-12



When starting the VM and creating a debug.log this seems to be the relevant part:

840074 ioctl(4, 0x40202e0a, 0x7fff63c36900) = -1 ENOMEM (Cannot allocate memory)
840074 write(2, "env_create error", 16) = -1 EBADF (Bad file descriptor)
840074 write(2, ": Cannot allocate memory", 24) = -1 EBADF (Bad file descriptor)
840074 write(2, "\n", 1) = -1 EBADF (Bad file descriptor)





R1Soft stopped the agent service and unloaded the hcpdriver that we use to perform backups. Afterwards, they attempted to run a simple list of the current running vznodes on the server and notice that it's returning "Cannot allocate memory" error messages.

[root@vz ~]# vzlist
Can't access file /vz/private/7/root.hdd/DiskDescriptor.xml: Cannot allocate memory
Can't access file /vz/private/8/root.hdd/DiskDescriptor.xml: Cannot allocate memory
CTID NPROC STATUS IP_ADDR HOSTNAME
2 70 running xx4.1xx.x3.204 server.xxxxx.com
4 72 running 1xx.xx4.2x.206 host.xxxxxxxxxxxxxxxxxx.com
6 19 running 1xx.1x.x3x.205 host.xxxxxxxxxx.com
7 68 running xx.x4.1xx.98 host.xx.com
8 66 running xx.154.1x9.x9 host.xxxxxxx.com
10 58 running xxx.xx.1x9.1xx host.xxxxxxxxxxxx.com


R1Soft says the problem doesn't look to be related to the cdp-agent process nor the driver as this issue wouldn't be a problem after stopping and unloading the driver. There is plenty of memory on the server so not sure why the cannot allocate memory is being returned. If this continues to be an issue we can uninstall the agent entirely to see if it makes a difference, but based on what I can find the agent installation is not the cause of this issue.


Any suggestions?
Re: Getting error when trying to restart VMs [message #50101 is a reply to message #50033] Fri, 26 July 2013 19:17 Go to previous message
Paparaciz
Messages: 302
Registered: August 2009
Senior Member
Hi,

first of all you didn't mentioned software and kernel versions.
please provide kernel, vzctl, ploop versions. what distribution and version runs in HN.

I don't know anything about "r1soft", but errors are very strange. maybe it's something related that CT ploop disk locking or smth. it is question how software backup files.

also there can be other limits, for example max number of open files or other.

if you know how to reproduce or at least get more help how to investigate problem and maybe found a bug just file a bug report at https://bugzilla.openvz.org/

p.s. ploop layout gives very good backup tools. please read more in http://wiki.openvz.org/Ploop/Backup
Previous Topic: kernel-firmware-2.6.32-358.14.1.el6.noarch is obsoleted by vzkernel-firmware-2.6.32-042stab078.28
Next Topic: how to migrate x86 container from x86 to x86_64
Goto Forum:
  


Current Time: Sun Jul 28 06:26:19 GMT 2024

Total time taken to generate the page: 0.02744 seconds