I've just purchased a QNAP TS-459 Pro II NAS to replace an aging (and creaking) Apple Time Capsule. Unfortunately, my 64-Bit Oneiric build is prone to a complete kernel-panic system hang when using the NAS, something it never exhibited with the TC.
My system comprises ubuntu 11:10 running on top of a Core i7 Quad system. I have added KDE to run as my GUI environment, on top of the default Window Manager.
My system currently has a working Time Capsule connection. Here is my connection string for that:-
//192.168.1.40/Data /media/capsule cifs username=####,password=####,rw,iocharset=utf8,file _mode=0777,dir_mode=0777
Since this worked, I thought it might be relatively straightforward to tweak the above to produce a new mount statement for the QNAP, so I produced one that looks like this:-
//192.168.1.41/Public /media/nas cifs username=####,password=####,rw,iocharset=utf8,file _mode=0777,dir_mode=0777
It worked, beautifully, and I was very smug for at least 20 minutes or so. At that point, my machine hung. We're talking a full system hang: screen, keyboard and mouse locking, and my Rhythmbox audio player catching a 1/5th snippet of a song in an endless loop. It took the hardware reset button to cure that problem. This has *NEVER* happened on ubuntu for me before...
Had a chat with the hard-core tekkies at work and they recommended that I replace the CIFS connection with something using NFS (which should be faster, more stable, etc). This took a little bit of tweaking, but at the second attempt I came to this:-
192.168.1.41ublic /media/nas nfs defaults 0 2
which also connects just perfectly and appears to behave flawlessly. Obviously I had to add nfs-common and portmap to get that to work... Except it then exhibits the same problems as the CIFS connection, dropping dead after a short period of time. To give an indication of the "death time"...
After getting the NFS mount to work, I flushed the original config of my Rhythmbox player [wiped /home/{me}/.local/share/Rhythmbox ] and re-launched, then directed the "folder import" across the NFS mount to my iTunes library now relocated to the QNAP... The player was able to scan 153 tracks before the interface hung and the system crashed. I can make it do this pretty consistently.
One more point of note: when I use Rhythmbox now [selecting from the 150ish tracks] it seems to work fine... but a check in /var/log/syslog shows me this sort of thing:
Mar 7 21:27:13 voyager kernel: [ 3093.635159] r8169 0000:03:00.0: eth0: link up
Mar 7 21:27:22 voyager kernel: [ 3102.820707] r8169 0000:03:00.0: eth0: link up
Mar 7 21:27:26 voyager kernel: [ 3107.379628] r8169 0000:03:00.0: eth0: link up
Mar 7 21:27:56 voyager kernel: [ 3136.627584] r8169 0000:03:00.0: eth0: link up
Mar 7 21:28:15 voyager kernel: [ 3155.716853] r8169 0000:03:00.0: eth0: link up
Mar 7 21:28:21 voyager kernel: [ 3161.906999] r8169 0000:03:00.0: eth0: link up
Mar 7 21:28:26 voyager kernel: [ 3166.952556] r8169 0000:03:00.0: eth0: link up
Mar 7 21:28:37 voyager kernel: [ 3177.773495] r8169 0000:03:00.0: eth0: link up
Mar 7 21:28:48 voyager kernel: [ 3188.530705] r8169 0000:03:00.0: eth0: link up
Mar 7 21:28:53 voyager kernel: [ 3194.166439] r8169 0000:03:00.0: eth0: link up
Mar 7 21:28:58 voyager kernel: [ 3198.972636] r8169 0000:03:00.0: eth0: link up
Mar 7 21:29:04 voyager kernel: [ 3204.540730] r8169 0000:03:00.0: eth0: link up
Mar 7 21:29:28 voyager kernel: [ 3228.783041] r8169 0000:03:00.0: eth0: link up
I can't swear, but I'm pretty sure that I didn't get this to the Time Capsule!
A couple more points. First, using *exactly the same hardware* and a different bootable OS drive [I use caddies] I can run up a copy of Windows 7 and it will connect to both the QNAP and the Time Capsule *flawlessly*. In other words, I am pretty sure I don't have a hardware, network, cable, or NAS problem. To verify, I also have 2 Mac Minis here, and both are quite comfortable working with the NAS and the TC.
Every instinct leads me to suspect that I am just being a numpty and using a suspect connection profile for the CIFS or NFS links with Oneiric, but try as I might, I can't find it.
I would be very happy to scrape logs, provide other config information, etc, to anyone willing to share their experience in the hope that we can fix this. Not sure if it's relevant, but on Saturday I also patched up the NAS to it's latest Firmware. I read the release notes, but I couldn't see anything that would suggest it fixes this problem...
Any ideas please folks? [ Thanks in advance ... ]
Bookmarks