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
If we wish to use the mq server in a nonpersistent mode (eg. for integration testing our applications) we don't mount the data directory and we rely on the overlayfs provided by docker.
This works fine and dandy if there are no other filesystems mounted, but we sometimes run these integration tests on a Mesos cluster, which by default mounts a sandbox directory under /mnt/mesos. In this case the container fails to start.
The logic to check if we are running in non-persistent mode only works if there are no other file systems under /mnt:
If we wish to use the mq server in a nonpersistent mode (eg. for integration testing our applications) we don't mount the data directory and we rely on the overlayfs provided by docker.
This works fine and dandy if there are no other filesystems mounted, but we sometimes run these integration tests on a Mesos cluster, which by default mounts a sandbox directory under /mnt/mesos. In this case the container fails to start.
The logic to check if we are running in non-persistent mode only works if there are no other file systems under /mnt:
mq-container/cmd/runmqserver/mqconfig.go
Line 134 in 4d40513
Perhaps this should verify the exact path /mnt/mqm is mounted before performing the checkFs call.
The text was updated successfully, but these errors were encountered: