USB Drive Mounts Fail on Boot

Not sure even where to begin with this one. After countless years of automatically mounting USB drives at boot time via /etc/fstab entries, it no longer works. Other drives seem to mount OK. If I mount manually after boot (like: mount /mnt/usbdrive) it works perfectly, so I know my fstab file is configured properly.

Example entry looks like: UUID=dc38f3dd-d5ab-4c86-b15c-cb158cef54e1 /mnt/WD22TB01 xfs noauto,rw,suid,dev,exec,nouser,async 0 0

There’s got to be something simple that I’m missing. Any suggestions where to look? Did behavior change in some version update?

Thanks.

The man mount did write already in el7:

noauto Can only be mounted explicitly (i.e., the -a option will not cause the filesystem to be mounted).

I use noauto,nofail,x-systemd.automount,x-systemd.idle-timeout=300 that generate systemd automounter, which mounts on access.

Thanks.

I bounced around with “noauto” for a while… The reason I put it in was I have instances where the USB drive may not be present, and without the “noauto” the entire boot process would hang and the system would never boot.

I need to find a way to account for those cases.

Isn’t the ‘nofail’ for that?
man mount writes:

nofail
Do not report errors for this device if it does not exist.

As I did show, I do use ‘noauto,nofail’ too, and the automount to do the mounting.

Not worried about reporting errors… I’m worried about not booting!! LOL!!

I’ll have to play around with this some time when I can physically get to the server and shut down/reboot multiple times. Probably won’t be for a while.

You may not be worried, but

The mount -a is called by some SystemD unit. When mount “reports error”, the unit knows that something did fail and therefore it has to report to SystemD that it failed to start. There are many other units that are started on boot only after the filesystem unit has started successfully. (They have no way to know that the error was in non-vital filesystem.) Therefore, AFAIK, that error reporting prevents the boot.