Mar 21

NAS Server Fixes

A few problems that I had to solve while setting up the new NAS.

One of the ZFS pools dismounted for no reason

This was a problem I’d never had before. Even the pool was gone and I had to re-import it using zfs import –f [pool name].

Hopefully, this isn’t a sign of problems to come.

ZFS doesn’t mount on boot and dir gets mounted by other app first

– sudo rm –R [mountpoint name]

– sudo zfs mount –a

(Technically, this is wrong, since mountall is installed, but somehow, it isn’t working on the main volume)

Edit /etc/default/zfs

Change ZFS_MOUNT=’no’ to ZFS_MOUNT=’yes’

Routing Table is Fucked Up

– applications that should be going out on eth0 are stuck on tun0

– netstat –rn showed slightly different tables on old NAS and new NAS

– turned out it wasn’t the routing tables, but a typo in rc.local that for some reason had the old machine’s eth0 address in it. These lines should be in rc.local

ip rule add from [eth0 address] table 128
ip route add table 128 to [eth0 address] dev eth0
ip route add table 128 default via [eth0 gateway]

Crashplan Reports whole directories as missing when they are clearly there:

– stop service

– go to /usr/local/crashplan/cache and delete EVERYTHING

– restart service

– it will compare files (might take a while)

Leave a Reply