You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@cloudstack.apache.org by Nathan Gleason <n8...@gmail.com> on 2023/08/09 19:55:56 UTC
Linstor volume boot issues
Hello,
We are working to get Linstor integrated into our environment and have
noticed a few issues.
When we spawn a VM everything works as expected. Cloudstack reports
success, we see the volume(s) are "InUse" in Linstor. But when we attach
to the console we see the SeaBIOS screen with the message "Boot failed: Not
a bootable disk" "No bootable device". The strange thing is that sometimes
it works and the VM will boot. We have tried many different templates that
all work just fine on local storage. We also see that sometimes the drbd
primary volume is assigned to the "DfltDisklessStorPool", sometimes it is
assigned to the pool we created "cloudstack_pool"
Things we have checked include:
- Make sure the block device is indeed attached to the host machine
- virsh dumpxml the VM to make sure it's looking at the correct block device
- numerous "--place-count N" resource group configurations
Has anyone run into this boot issue?
Versions:
Ubuntu 22.04
Cloudstack 4.17.1.0
linstor-server-1.23.0
drbd-9.2.4
drbd-utils-9.25
Thanks,
Nathan
Re: Linstor volume boot issues
Posted by Nathan Gleason <n8...@gmail.com>.
Hi Antoine, the fix was an updated Cloudstack Linstor storage plugin. The
fix should be in the 4.18.1.0 version of the plugin.
Thanks,
Nathan
On Sat, Feb 3, 2024 at 6:14 PM Antoine Boucher <an...@haltondc.com>
wrote:
> Hello Nathan,
>
> We were experimenting with Linstor and we have the same issue when a VM is
> directly created from a Template on a Linstor Primary Storage but not when
> created on a non Linstor PS and then Migrated to Linstor PS. Have you
> found the culprit ?
>
> Regards,
> Antoine
>
> On Aug 9, 2023, at 15:56, Nathan Gleason <n8...@gmail.com> wrote:
>
> Hello,
>
> We are working to get Linstor integrated into our environment and have
> noticed a few issues.
>
> When we spawn a VM everything works as expected. Cloudstack reports
> success, we see the volume(s) are "InUse" in Linstor. But when we attach
> to the console we see the SeaBIOS screen with the message "Boot failed: Not
> a bootable disk" "No bootable device". The strange thing is that sometimes
> it works and the VM will boot. We have tried many different templates that
> all work just fine on local storage. We also see that sometimes the drbd
> primary volume is assigned to the "DfltDisklessStorPool", sometimes it is
> assigned to the pool we created "cloudstack_pool"
>
> Things we have checked include:
> - Make sure the block device is indeed attached to the host machine
> - virsh dumpxml the VM to make sure it's looking at the correct block
> device
> - numerous "--place-count N" resource group configurations
>
> Has anyone run into this boot issue?
>
> Versions:
> Ubuntu 22.04
> Cloudstack 4.17.1.0
> linstor-server-1.23.0
> drbd-9.2.4
> drbd-utils-9.25
>
> Thanks,
> Nathan
>
Re: Linstor volume boot issues
Posted by Antoine Boucher <an...@haltondc.com>.
Hello Nathan,
We were experimenting with Linstor and we have the same issue when a VM is directly created from a Template on a Linstor Primary Storage but not when created on a non Linstor PS and then Migrated to Linstor PS. Have you found the culprit ?
Regards,
Antoine
On Aug 9, 2023, at 15:56, Nathan Gleason <n8...@gmail.com> wrote:
Hello,
We are working to get Linstor integrated into our environment and have
noticed a few issues.
When we spawn a VM everything works as expected. Cloudstack reports
success, we see the volume(s) are "InUse" in Linstor. But when we attach
to the console we see the SeaBIOS screen with the message "Boot failed: Not
a bootable disk" "No bootable device". The strange thing is that sometimes
it works and the VM will boot. We have tried many different templates that
all work just fine on local storage. We also see that sometimes the drbd
primary volume is assigned to the "DfltDisklessStorPool", sometimes it is
assigned to the pool we created "cloudstack_pool"
Things we have checked include:
- Make sure the block device is indeed attached to the host machine
- virsh dumpxml the VM to make sure it's looking at the correct block device
- numerous "--place-count N" resource group configurations
Has anyone run into this boot issue?
Versions:
Ubuntu 22.04
Cloudstack 4.17.1.0
linstor-server-1.23.0
drbd-9.2.4
drbd-utils-9.25
Thanks,
Nathan