Error!

[YaseenKamala]

Hello
I have a PVE node which has the bellow list of pools and the relevant disks:
root@dw-srv-p4:~# zfs list
NAME USED AVAIL REFER MOUNTPOINT
hddpool 5.04T 3.38T 153K /hddpool
hddpool/vm-100-cloudinit 6M 3.38T 121K -
hddpool/vm-100-disk-2 4.03T 4.31T 3.10T -
hddpool/vm-100-disk-3 1.01T 4.26T 128G -
nvmepool 95.0G 366G 96K /nvmepool
nvmepool/data 95.0G 366G 96K /nvmepool/data
nvmepool/data/vm-100-disk-0 95.0G 366G 95.0G -
rpool 175G 748G 104K /rpool
rpool/ROOT 8.61G 748G 96K /rpool/ROOT
rpool/ROOT/pve-1 8.61G 748G 8.61G /
rpool/data 166G 748G 96K /rpool/data
rpool/data/base-9000-disk-0 1.02G 748G 1.02G -
rpool/data/base-9001-disk-0 942M 748G 942M -
rpool/data/vm-100-disk-0 56.8G 748G 56.8G -
rpool/data/vm-1002-cloudinit 424K 748G 424K -
rpool/data/vm-1002-disk-0 35.5G 748G 20.4G -
rpool/data/vm-1002-state-swarm_backup_snapshot 904M 748G 904M -
rpool/data/vm-1004-cloudinit 424K 748G 424K -
rpool/data/vm-1004-disk-0 20.3G 748G 20.3G -
rpool/data/vm-1004-disk-1 50.5G 748G 50.5G -
rpool/data/vm-9000-cloudinit 56K 748G 56K -
rpool/data/vm-9001-cloudinit 56K 748G 56K -

When the node is off by any reason there is a VM which can't start on reboot due to the below error:

zfs error: cannot open 'hddpool': no such pool

TASK ERROR: zfs error: cannot create 'hddpool/vm-100-cloudinit': dataset already exists

However when I start the VM manually it works. could you please tell me why I am getting this error?



Thanks in advance for your help as usual

 

[mira]

It seems the pool requires a bit more time to become available.
Could you try adding a boot delay to the VM start? This should be enough time for the `hddpool` to become available.

 

[YaseenKamala]

Hello @mira ,
Thanks for you reply, Could you please tell me how do we make the boot delay?

I have another problem in this VM whenever the host goes down. the self assigned certificate on that VM changes, could it be related?

Thanks

 

[mira]

What exactly do you mean the self signed certificate changes? What errors do you get?

If you take a look at VM -> Options -> Start/Shutdown order you can find the delay option. There set a delay of 1 or 2 minutes and see if that resolves your issue.

 

[YaseenKamala]

Thanks a lot @mira I have seted to 3

so the VM is hosting a webserver and I use a SSL self assigned certificate. whenever the host goes down not the VM I have to regenerate the certificate.
so I am wandering if there a kind of setting or a change that made which can initialized the VM when the host is down or restarted. do you have any ideas about such kind of settings?

 

[mira]

Please note that the delay here is in seconds. I mentioned minutes based on your previous output when the VM start failed and when it succeeded.

It seems the certificate issue is a misconfiguration inside your guest and has nothing to do with the host. Unless you rollback a snapshot on start every time?

 

[YaseenKamala]

OK I seted to 180 seconds

Could you please tell me how can I check if the VM "rollback's a snapshot on start every time?"

 

[mira]

If you haven't scripted something yourself, then that's not the case.
So check inside your guest to see why new certificates get generated on every boot.

 

[YaseenKamala]

Know I have not scripted such kind of settings. again thank you so much for your help.

 

[YaseenKamala]

@mira I know it's not related to the ProxMox, Do you see any thing wrong with the bellow mounted points and disks?
?

 

[mira]

No, not really seeing anything wrong here.
Do you have issues, or anything concrete you're looking for?