Hello, recently I have a problem where all the VMs in my pve register a high read and write load at the same time.
The times are random, sometimes it happens at 1-hour intervals, other times it takes a long time
Even VMs with basic functions in Linux or Windows at idle register the high load at exactly the same moment.
**Generally the io usage of all VMs is between 40k and 60k.
It happens both in vms with ssd storage in the pve and in vms with nvme storage scsi storage, which I have already ruled out being something on the disk.
The problem is that at the time of these loads, some processes running in the vm stop responding or simply shut down.
In my syslog I noticed that at the time of these loads it records the following log below on all vms:
"Oct 06 22:23:30 br209 pvestatd[2189]: VM 1007 qmp command failed - VM 1007 qmp command 'query-proxmox-support' failed - got timeout
Oct 06 22:23:35 br209 pvestatd[2189]: VM 1002 qmp command failed - VM 1002 qmp command 'query-proxmox-support' failed -
unable to connect to VM 1002 qmp socket - timeout after 51 retries"
Has anyone had something similar or have any idea what it could be?
My pve configuration is as follows:
Dual Xeon E5-2660 v2 @ 2.20GHz / 128GB RAM
Kernel 5.15.116-1-pve
PVE 7.4-16
Storage:
1 Micron M600 SSD = PVE and 1 Linux VM
ISCSI NVME storage on 10Gbps port = 10 VMs varying between Windows and Linux
The times are random, sometimes it happens at 1-hour intervals, other times it takes a long time
Even VMs with basic functions in Linux or Windows at idle register the high load at exactly the same moment.
**Generally the io usage of all VMs is between 40k and 60k.
It happens both in vms with ssd storage in the pve and in vms with nvme storage scsi storage, which I have already ruled out being something on the disk.
The problem is that at the time of these loads, some processes running in the vm stop responding or simply shut down.
In my syslog I noticed that at the time of these loads it records the following log below on all vms:
"Oct 06 22:23:30 br209 pvestatd[2189]: VM 1007 qmp command failed - VM 1007 qmp command 'query-proxmox-support' failed - got timeout
Oct 06 22:23:35 br209 pvestatd[2189]: VM 1002 qmp command failed - VM 1002 qmp command 'query-proxmox-support' failed -
unable to connect to VM 1002 qmp socket - timeout after 51 retries"
Has anyone had something similar or have any idea what it could be?
My pve configuration is as follows:
Dual Xeon E5-2660 v2 @ 2.20GHz / 128GB RAM
Kernel 5.15.116-1-pve
PVE 7.4-16
Storage:
1 Micron M600 SSD = PVE and 1 Linux VM
ISCSI NVME storage on 10Gbps port = 10 VMs varying between Windows and Linux
Last edited: