samba - Apr 2008 - Printing Problem with Samba ' Failed to allocate a print job'

I=92m using Samba 3.0.24 on Ubuntu Feisty.  I=92m managing my printers with
CUPS. =20

=20

Just yesterday, one printer stopped being able to print through Samba.  This
printer still prints fine directly from CUPS.  When I try to send any print
job to the print via Samba, I get this:

=20

[2008/04/08 11:15:58, 0] printing/printing.c:allocate_print_jobid(2262)

  allocate_print_jobid: failed to allocate a print job for queue
truss_hp4050_2

[2008/04/08 11:15:58, 3] printing/printing.c:print_job_start(2431)

  print_job_start: returning fail. Error =3D No space left on device

=20

At first glance, I would think the disk was full.  It=92s actually not:

root@oink:~# df -h

Filesystem            Size  Used Avail Use% Mounted on

/dev/sdf1             106G   41G   60G  41% /

varrun               1006M  844K 1005M   1% /var/run

varlock              1006M  4.0K 1006M   1% /var/lock

procbususb           1006M  140K 1006M   1% /proc/bus/usb

udev                 1006M  140K 1006M   1% /dev

devshm               1006M     0 1006M   0% /dev/shm

/dev/sda1             1.4T  293G 1014G  23% /data

/dev/md0              1.2T  770G  380G  67% /backup

/data/home            1.4T  293G 1014G  23% /home

=20

Then I checked to make sure the permissions on the spool directory were OK:

root@oink:~# ls -ld /data/samba/spool

drwxrwxrwt 3 root Domain Users 49152 2008-04-08 11:17 /data/samba/spool

=20

I don=92t use /var/spool/samba but here are its permissions anyway:

root@oink:~# ls -ld /var/spool/samba

drwxrwxrwt 2 root root 4096 2007-05-22 13:53 /var/spool/samba

=20

Thinking it might just be a temporary thing with Samba, I restarted it last
night after everyone left.  The problem still persists.  I have not
restarted the entire server yet.  There are no errors in dmesg or
/var/log/messages.

=20

I checked this morning to see if there was a newer Samba version for Feisty,
but there isn=92t.  I found a few instances of this type of problem on Google,
but no real answers.  Plus, they all seemed to be old.

=20

Thanks for any help,

Misty

=20


No virus found in this outgoing message.
Checked by AVG.=20
Version: 7.5.519 / Virus Database: 269.22.9/1365 - Release Date: 4/8/2008
7:30 AM
=20

Hi Bill,

That would be fabulous (in a bad way) but it is not the case.  Here is the
equivalent part of dumpe2fs for /dev/sda1 (/data):
Inode count:              183156736
Block count:              366286008
Reserved block count:     18314300
Free blocks:              287497221
Free inodes:              181068686

And here is for /dev/sdf1 (/):
Inode count:              14057472
Block count:              28103701
Reserved block count:     1405185
Free blocks:              17311944
Free inodes:              13849218

Any other ideas would be appreciated!

--Misty

-----Original Message-----
From: William Jojo [mailto:jojowil@hvcc.edu] 
Sent: Tuesday, April 08, 2008 1:49 PM
To: Misty Stanley-Jones
Subject: Re: [Samba] Printing Problem with Samba ' Failed to allocate a
print job'



This will sound stupid, I am certain, but is this an inode issue? Does the FS
type you are using auto-extend (like JFS2 on AIX) the number of available
inodes? Or fragmentation perhaps?

I only ask since you clearly have enough space, but a lack of inode may appear
as no space left on device and a high level of fragmentation in JFS (not jfs2)
on AIX will throw strange errors.


Cheers,
Bill

(PS: I know you are not running AIX. :-) :-) )



No virus found in this outgoing message.
Checked by AVG. 
Version: 7.5.519 / Virus Database: 269.22.9/1365 - Release Date: 4/8/2008 7:30
AM