You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This isn't necessarily a bug as such, but I'm trying to get a bit more info on the failing mechanism to see if there is any way I can at least work around it, or even better perhaps contribute a fix somewhere.
When using CNI plugins with Nomad, and docker running from a snap, it results in the following error pattern when creating the container via Nomad:
failed to setup alloc: pre-run hook "network" failed: failed to configure networking for alloc: failed to configure network: plugin type="bridge1" failed (add): failed to open netns "/run/snap.docker/netns/db65a7df22ec": unknown FS magic on "/run/snap.docker/netns/db65a7df22ec": 1021994
The bridge1 plugin referred to here is identical to the referenced in the workaround here: hashicorp/nomad#11085
I've tried this with both CNI plugin bundle versions 1.3.0 and 1.6.0
Nomad [ which is also running from a snap ], works fine for standard docker operations.
I have tried running the nomad process with the snap confinement disabled [ apparmor and seccomp ], but get the same error. I'm wondering if there is some kind of mount namespace issue going on, but it's not clear because the contents of /run/snap.docker/netns/ is isn't itself in a separate mount namespace.
Is anyone able to tell me where I should start looking ? What is the likely cause of unknown FS magic ?
It seems like there is some kind of mount info somewhere which can't be access from nomad,. Just not sure where.
Thanks very much!
Cheers,
Just
The text was updated successfully, but these errors were encountered:
Hi,
This isn't necessarily a bug as such, but I'm trying to get a bit more info on the failing mechanism to see if there is any way I can at least work around it, or even better perhaps contribute a fix somewhere.
When using CNI plugins with Nomad, and docker running from a snap, it results in the following error pattern when creating the container via Nomad:
The
bridge1
plugin referred to here is identical to the referenced in the workaround here: hashicorp/nomad#11085I've tried this with both CNI plugin bundle versions
1.3.0
and1.6.0
Nomad [ which is also running from a snap ], works fine for standard docker operations.
I have tried running the nomad process with the snap confinement disabled [ apparmor and seccomp ], but get the same error. I'm wondering if there is some kind of mount namespace issue going on, but it's not clear because the contents of
/run/snap.docker/netns/
is isn't itself in a separate mount namespace.Is anyone able to tell me where I should start looking ? What is the likely cause of
unknown FS magic
?It seems like there is some kind of mount info somewhere which can't be access from nomad,. Just not sure where.
Thanks very much!
Cheers,
Just
The text was updated successfully, but these errors were encountered: