[SOLVED] Can't properly stop from booting into emergency mode

[LtPl4y3r]

I have a Proxmox 4 box that is up to date, and every time I have to reboot, it sends me into emergency mode, where it asks me to check out the journal and whatnot to find the problem. If I (controil-D) through the mode, then everything boots fine! Everything Just Works (tm), but my server is in a closet without a keyboard or monitor, so it makes that an awkward dance every time.

It seems to me that udev is reporting some failures which causes systemd to panic. I have found that if I run:

# systemctl mask systemd-udev-settle

That the machine will boot regularly, but then it no longer automounts my ZFS pool. This pool is only for storage and doesn't house any vm images/containers or other critical mount points.

My setup is a TS140, with a 500GB sata hard drive that holds proxmox and the vm containers, then the two mirrored zfs drives. I also have an intel quad port gigabit card ET.

I have tried changing the fstab entries from '/dev/pve/root' to 'uuid=XXXXX', to no success.

I'm hitting my head against a wall here, and any help would be appreciated.

Below is the most recent journal (attached) and a copy of fstab:

# <file system> <mount point> <type> <options> <dump> <pass>
#/dev/pve/root / ext4 errors=remount-ro 0 1
#/dev/pve/data /var/lib/vz ext4 defaults 0 1
#/dev/pve/swap none swap sw 0 0
#proc /proc proc defaults 0 0

UUID=3492d218-b8ca-4c12-aad0-596decc6f51b / ext4 errors=remount-ro 0 1
UUID=b0f1aa98-656c-4029-a2f8-3d64df78bc74 /var/lib/vz ext4 defaults 0 1
UUID=f0fa3c20-a8a8-40ae-af8b-eace7cb67b88 none swap sw 0 0
proc /proc proc defaults 0 0

 

[sdinet]

I don't know anyone who has good things to say about software RAID

 

[LtPl4y3r]

I don't know anyone who has good things to say about software RAID

Are you trying to say that if I blew away my ZFS mirror that it would solve the udev startup problem?

 

[sdinet]

I would say re-image proxmox and just use hardware raid. you can copy your VM disk images and config files to an external hard drive, then import. If you cant afford any downtime, idk...

 

[LtPl4y3r]

I took your advice and reimaged my host. It seems to have fixed the problem.

Before I did that, though, I destroyed my ZFS mirror, which had no effect on the whole situation. I've since recreated the mirror and restored from backups, and everything in lovely.

Thanks for your help.

 

[sdinet]

Great! You should mark your post as 'solved'