[PVE-User] hrtimer errors

[lst_hoe02 at kwsoft.de]

Zitat von David Lawley <davel at upilab.com>:

> thanks, I was looking thru the logs and found this;
>
> Found a thread in the forum on this warning..  but not sure its in  
> the same context as I'm following but it might.  As far a memory,  
> well guess stranger things have happened..  I have upped memory  
> allocation to my each of my VMs.
>
> (I'm thinking this is just errors from the underlying issue.)
>
> Nov 23 18:31:45 proliant02 pmxcfs[3128]: [status] notice: received log
> Nov 23 18:46:45 proliant02 pmxcfs[3128]: [status] notice: received log
> Nov 23 19:01:45 proliant02 pmxcfs[3128]: [status] notice: received log
> Nov 23 19:16:45 proliant02 pmxcfs[3128]: [status] notice: received log
> Nov 23 19:17:01 proliant02 /USR/SBIN/CRON[261314]: (root) CMD (   cd  
> / && run-parts --report /etc/cron.hourly)
> Nov 23 19:17:24 proliant02 pvestatd[3666]: WARNING: unable to  
> connect to VM 124 socket - timeout after 31 retries
> Nov 23 19:17:27 proliant02 pvestatd[3666]: WARNING: unable to  
> connect to VM 104 socket - timeout after 31 retries
> Nov 23 19:17:30 proliant02 pvestatd[3666]: WARNING: unable to  
> connect to VM 142 socket - timeout after 31 retries
> Nov 23 19:17:33 proliant02 pvestatd[3666]: WARNING: unable to  
> connect to VM 139 socket - timeout after 31 retries
> Nov 23 19:17:36 proliant02 pvestatd[3666]: WARNING: unable to  
> connect to VM 107 socket - timeout after 31 retries
> Nov 23 19:17:39 proliant02 pvestatd[3666]: WARNING: unable to  
> connect to VM 116 socket - timeout after 31 retries
> Nov 23 19:17:42 proliant02 pvestatd[3666]: WARNING: unable to  
> connect to VM 154 socket - timeout after 31 retries
> Nov 23 19:17:45 proliant02 pvestatd[3666]: WARNING: unable to  
> connect to VM 106 socket - timeout after 31 retries
> Nov 23 19:17:48 proliant02 pvestatd[3666]: WARNING: unable to  
> connect to VM 100 socket - timeout after 31 retries
> Nov 23 19:17:51 proliant02 pvestatd[3666]: WARNING: unable to  
> connect to VM 138 socket - timeout after 31 retries
> Nov 23 19:17:54 proliant02 pvestatd[3666]: WARNING: unable to  
> connect to VM 111 socket - timeout after 31 retries
> Nov 23 19:17:57 proliant02 pvestatd[3666]: WARNING: unable to  
> connect to VM 153 socket - timeout after 31 retries
> Nov 23 19:18:00 proliant02 pvestatd[3666]: WARNING: unable to  
> connect to VM 109 socket - timeout after 31 retries
> Nov 23 19:18:03 proliant02 pvestatd[3666]: WARNING: unable to  
> connect to VM 171 socket - timeout after 31 retries
> Nov 23 19:18:06 proliant02 pvestatd[3666]: WARNING: unable to  
> connect to VM 113 socket - timeout after 31 retries
> Nov 23 19:18:15 proliant02 kernel: BUG: soft lockup - CPU#22 stuck  
> for 67s! [ksmd:265]
>

We have also seen this but in our case the node crashed after some  
time. At least in our case bad RAM should not be the root case because  
it happend at random times at three different nodes with ECC RAM. We  
suspect it is related to the virtio we are using for all Windows VMs...
Do you also use virtio for Windows guests?

Regards

Andreas