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
refactor(dynamic mount): introduce dynamic server manager
We found that when a pod container is recreated, there's a chance the
dynamically mounted CSI directory gets unmounted. Since this directory
is shared among multiple pod containers via mount propagation from
/var/run/model-csi (a tmpfs), it introduces instability and security
isolation risks, for example, if one pod learns another pod's volume
name, it could maliciously call APIs to mount or unmount that submount
of volume.
To ensure secure isolation for each dynamic mount and avoid unstable
mount propagation, an independent csi.sock is currently created under
each dynamic mount directory instead of using a shared csi.sock, these
individual csi.sock servers are managed by the DynamicServerManager.
Signed-off-by: imeoer <[email protected]>
err:=handleError(echoCtx, grpcStatus.Error(codes.ResourceExhausted, errors.Wrap(errors.Wrapf(errors.Wrapf(syscall.ENOSPC, "model image is , but only of disk quota is available"), "pull model failed"), "pull model for dynamic volume").Error()))
0 commit comments