Very Slow VMs on 5.3

[digitalbots]

Hello All,

I am not sure what I am doing wrong. But I have tried to build 3 different vm's all of which use very little resources but each time I build a VM it takes over 15 mins to boot, and hours to install.

Now to be very clear when I am doing the install no other VMs are on at the time. and the OS I am trying to load is Lite Linux and OpenMediaVault.

here are some stats

root@pve:~# pveperf
CPU BOGOMIPS: 57597.84
REGEX/SECOND: 1386937
HD SIZE: 27.19 GB (/dev/mapper/pve-root)
BUFFERED READS: 307.49 MB/sec
AVERAGE SEEK TIME: 0.12 ms
FSYNCS/SECOND: 547.31
DNS EXT: 25.85 ms
DNS INT: 0.53 ms (levyhome.local)

config of the OMV server
root@pve:~# qm config 100
bootdisk: scsi0
cores: 1
ide2: local:iso/openmediavault_4.0.14-amd64.iso,media=cdrom
memory: 1024
name: OMV
net0: virtio=A2:3E:A4:32:FA:7B,bridge=vmbr0
numa: 0
ostype: l26
scsi0: wade:vm-100-disk-0,cache=writeback,size=120G
scsihw: virtio-scsi-pci
smbios1: uuid=c364a39e-2383-4e64-b696-cc4537ab736f
sockets: 2
usb0: host=2-3
vmgenid: fa026614-bca8-4e20-afe4-230467e9a868

The wade drive is a Zpool which was created like this.
zpool create -f -o ashift=12 wade /dev/sda /dev/sdb

My machine has 32gb of ram and the CPU is a 8x AMD FX 8150.

Any help in trying to fix this would be greatly appreciated.

 

[sb-jw]

Do you already checked your Disks with SMART? You are able to check the disks performance itself with hdparm or something?

 

[digitalbots]

Do you already checked your Disks with SMART? You are able to check the disks performance itself with hdparm or something?

how would I do that?

 

[sb-jw]

You can use smartctl via shell or check it via PVE (disk tab on every Node).

Could you send us the dmesg log here?

 

[digitalbots]

ok they are attached.

 

[sb-jw]

[39383.508704] ata5: lost interrupt (Status 0x58)
[39383.508728] ata5.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen
[39383.508737] ata5.01: failed command: READ DMA
[39383.508744] ata5.01: cmd c8/00:01:80:07:10/00:00:00:00:00/f0 tag 0 dma 512 in
                        res 40/00:01:01:4f:c2/00:00:00:00:00/10 Emask 0x4 (timeout)
[39383.508748] ata5.01: status: { DRDY }
[39383.508766] ata5: soft resetting link
[39383.685696] ata5.01: configured for UDMA/33
[39383.685815] ata5: EH complete

This could be the problem. Can you post the output for "dmesg - T" so we will see the timestamp.

 

[digitalbots]

[39383.508704] ata5: lost interrupt (Status 0x58)
[39383.508728] ata5.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen
[39383.508737] ata5.01: failed command: READ DMA
[39383.508744] ata5.01: cmd c8/00:01:80:07:10/00:00:00:00:00/f0 tag 0 dma 512 in
                        res 40/00:01:01:4f:c2/00:00:00:00:00/10 Emask 0x4 (timeout)
[39383.508748] ata5.01: status: { DRDY }
[39383.508766] ata5: soft resetting link
[39383.685696] ata5.01: configured for UDMA/33
[39383.685815] ata5: EH complete

This could be the problem. Can you post the output for "dmesg - T" so we will see the timestamp.

Done!

 

[sb-jw]

Are you able to reboot your system and check immediately the disk performance? Or what disks do you have currently plugged into the system, only the two hard disks? The disks itself seems okay, so there might be a broken sata cable or its another disk, which are not related to this issue.

 

[digitalbots]

i have one USB drive with 4 HD's in it that is currently not plugged in at the moment but here is the logs once I did the reboot.

do you think that maybe I made the Zpool wrong?

 

[sb-jw]

Do you mean, you have 4 hard disks directly connected to the Mainboard and an USB stick, which currently not plugged in.

If you use sda and sdb as a zpool, where is the PVE itself installed?

It might be good if you can give us an complete overview of your setup, including the hardware and how the disks are connected etc.