freenas 8.2 file system full error

[nismohks]

Hi,

Recently one of my boxes kept coming up with these errors:

nano kernel: pid 1843 (nmbd), uid inumber 24794 on /var: file system full
nano kernel: pid 1536 (syslogd), uid inumber 30911 on /var: file system full
nano kernel: pid 44908 (python), uid inumber 33982 on /var: file system full
nano kernel: pid 2353 (nginx), uid inumber 30902 on /var: file system full

and the errors keep repeating every few minutes.

the system specs are:

HP N36L
Intel CT gigabit ethernet adaptor
2x4gb Kingston ECC Ram
5x ST2000DL003 Seagate Barracuda Green
Freenas 8.2.0 Release p1 x64

I can still access the shares, but in the terminal on the host, these error keep showing non stop.

Any idea on what i should do?
It was running fine for 7 days, since the last reboot but after coming back to it after the weekend i was presented with all these errors.

Thanks

 

[cyberjock]

Your /var file system is full(duh). FreeNAS is a very tightly packed system and has very little freespace. For some reason your USB key is filling up. Backup your config, reinstall FreeNAS and then restore your config. If you have another USB key I'd use it.

More than likely something is corrupted in FreeNAS.

 

[nismohks]

Your /var file system is full(duh). FreeNAS is a very tightly packed system and has very little freespace. For some reason your USB key is filling up. Backup your config, reinstall FreeNAS and then restore your config. If you have another USB key I'd use it.

More than likely something is corrupted in FreeNAS.

ahh ok thanks for that.

ive never had this problem before with my previous freenas builds, so you think something may be corrupted?

Is there any way where i can try and prevent this from happening again?

It's quite annoying because due to these messages, i can even control the system locally as it locks up but the gui and network shares still work fine!

thanks

 

[cyberjock]

If you want to compile your own copy of FreeNAS you could make the partitions bigger. There really isn't any way to "prevent" it because it shouldn't happen normally.

 

[William Grzybowski]

First step is find out why /var is full, the rootcause... what is taking up so much space

 

[nismohks]

First step is find out why /var is full, the rootcause... what is taking up so much space

With some help, ive got this now

[root@nano ~]# du -h /var/log
512B /var/log/samba/cores/smbd
512B /var/log/samba/cores/nmbd
1.5K /var/log/samba/cores
140M /var/log/samba
512B /var/log/proftpd
140M /var/log
[root@nano ~]#

[root@nano /var/log/samba]# ls -l
total 286239
drwx------ 4 root wheel 512 Aug 3 17:37 cores
-rw-r--r-- 1 root wheel 146247680 Aug 13 09:28 log.nmbd
-rw-r--r-- 1 root wheel 4553 Aug 13 11:20 log.smbd
-rw-r--r-- 1 root wheel 10245 Aug 10 14:27 log.smbd.old
[root@nano /var/log/samba]#

How can i see what is causing the error?

Sorry im new to linux/bsd stuff... any help would be much much appreciated.

Cheers,

 

[William Grzybowski]

Wow, 140MB of samba logs? paste tail -n 80 /var/log/samba/log.nmbd

There is something strange going on...
Copy that file to your volume so we can analyze it, then remove it from /var/log and restart CIFS.

 

[nismohks]

update:

i did "more log.nmbd" and was presented with this:

[2012/08/11 08:08:46, 0] nmbd/nmbd_subnetdb.c:127(make_subnet)
nmbd_subnetdb:make_subnet()
Failed to open nmb bcast socket on interface 0.255.255.255 for port 137. Er
ror was Can't assign requested address
[2012/08/11 08:08:46, 0] lib/util_sock.c:667(open_socket_in)
bind failed on port 137 socket_addr = 0.255.255.255.
Error = Can't assign requested address

it seems to be a network related error, but im not sure what it is trying to do.

i only run a single CIFS share on the box, is there anything i can disable or tweak to stop this from happening?
Like i was saying before, the box still works fine.

i have another freenas system based off an old AMD system with the same build release of freenas, but that doesnt have this problem.

Any suggestions?

 

[ben]

/var/log/samba is 140 megs? that's quite a lot.

 

[nismohks]

Wow, 140MB of samba logs? paste tail -n 80 /var/log/samba/log.nmbd

There is something strange going on...
Copy that file to your volume so we can analyze it, then remove it from /var/log and restart CIFS.

Hi William,

Just did what you told me and here is the output:

[2012/08/11 08:09:41, 0] lib/util_sock.c:667(open_socket_in)
bind failed on port 137 socket_addr = 0.255.255.255.
Error = Can't assign requested address
[2012/08/11 08:09:41, 0] nmbd/nmbd_subnetdb.c:127(make_subnet)
nmbd_subnetdb:make_subnet()
Failed to open nmb bcast socket on interface 0.255.255.255 for port 137. Er
ror was Can't assign requested address

So same as what i posted just previously.

So much fiasco though, i was trying to use the system locally before and the errors kept popping up and i couldnt get back to the main Freenas console(on the local system not gui) and in frustration i defaulted to CTRL+ALT-DEL and the system rebooted.
After that, freenas wouldnt boot and i had to reinstall and the i auto-imported the volume again. lucky i had copied the log file to my mount just prior so i can have a look at the logs!

Can anyone shed any more light as to what is causing this error?

thanks

 

[William Grzybowski]

Sorry, I don't have a clue