Troubleshooting
Should the need arise to troubleshoot a flexFS mount, there are a few tips that can help.
Log files
When running as root in the background (daemonized), or via fstab, debug logging is written to /root/.flexfs/mount/logs/<name>-<pid>.log unless otherwise specified. Note that <name> is the name of the volume and <pid> is the process ID of the mount process and will be replaced dynamically by the mount client when it starts. When not running in foreground mode, this is the first place you should look for answers when something isn't working as expected.
Verbose logging
By default, the flexFS mount client only logs high-level configuration data and serious error events. If more detailed logging is needed, you can specify the -v / --verbose flag in the start command or fstab options. Specifying this flag will dramatically increase the level of logging information.
Enabling this logging level has a negative impact on performance and should only be used for debugging purposes.
Transport endpoint not connected
If a flexFS mount client is killed or crashes, you can be left with a broken mount point. To recover from this condition, first try unmounting the broken mount point:
sudo umount -f <mount-point>
If unmounting fails, it may be necessary to first stop or kill processes that are using the mount point before unmounting. Those processes can generally be identified by running:
sudo lsof <mount-point>
or
sudo fuser -v -m <mount-point>
Killing all processes using a mount point can also be accomplished as follows:
sudo fuser -k -M -m <mount-point>
Once the mount point has been unmounted, you can then mount the volume on it again as specified in the mounting instructions.