Weird... ?
After deleting ALL email records (as per Peter Chubbs' email, and after
downloading a backup to my own machine) I ran those commands and found the
log files.... 700mb for the largest, next in line not far behind... so I
downloaded, as a backup, 'combined_log' and then deleted it .
Then I used 'df -h', and *no difference* ??
Here is the report:
Filesystem Size Used Avail Use% Mounted on
/dev/sda1 4.1G 4.1G 0 100% /
none 251M 0 251M 0% /dev/shm
/dev/hda1 61M 42M 16M 72% /old/r
/dev/hda5 387M 294M 73M 80% /old/usr
/dev/hda6 243M 41M 190M 18% /old/var
/dev/hda8 152M 47M 97M 33% /old/usr/local
/dev/hda9 3.5G 2.9G 505M 86% /old/home
/dev/hda10 3.5G 2.9G 469M 87% /old/cache
the logs came from /raid/ssl/..... so it's definitely on 'sda1' right?
4.1G still taken up! weird...
does 'rm' completely remove it? I should think so... it just makes the
i-node free so it's as good as gone right? :(
odd.. :P
Can anyone shed some light on this phenomena? (read: user error)
> On Thu, 25 Sep 2003 10:39 am, [EMAIL PROTECTED] wrote:
>> Well - the problems are never-ending here ....
>>
>> Our waterexchange mail server is reporting that it is out of disk space.
>>
>> (Outlook message: )
>> "The message could not be sent because it rejected you as the sender.
Server Response: '452 4.4.5 Insufficient disk space; try again later'.
(Account: 'mail.mactel.net.au', SMTP Server: 'waterexchange.com.au',
Error Number: 0x800ccc78)."
>
> *snipped*
>
>> Any ideas?
>
> Hopefully you have root on the box concerned - coz it looks like either
your
> mail spool or mailq is maxing out one of your disks.
>
> "df -h" will show you sussinctly how much space is being used where on a
per-partition basis. "du -s" is handy to show a summary (-s) of the
current directory's disk usage (du). Here's a quick-and-dirty hack to
find
> the biggest disk hogs starting from the current directory:
>
> find . -xtype d -maxdepth 1 -exec du -s {} \; | sort -rn | less
>
> here's what you should see (add a pipe to "less" if you want one screen
at a
> time):
> james@<myhost> - ~
>>find . -xtype d -maxdepth 1 -exec du -s {} \; | sort -rn
> 840800 .
> 181340 ./.thumbnails
> 142528 ./Documents
> 130136 ./Mail
> 114000 ./.kde
> 33372 ./.mozilla
> 30804 ./Desktop
> 22520 ./Utilities
> 12252 ./bin
> .....
> 4 ./.gnome_private
> 4 ./.gnome2_private
> 4 ./.gconfd
>
> james@<myhost> - ~
>>
>
> So the process is this:
> 1. Find the partition that's maxed out (probably /var)
> 2. cd /var (assuming /var is maxed out)
> 3. Run the "find . -xtype...." and have a look the first few dirs. The
biggest disk usage will be first.
> 4. dig around in the big dirs and keep going (using the "fins .
-xtype..." etc) until you idetify the problem.
>
> Here's my guesses:
> 1. The current crop of viruses are putting big demands on mail servers -
check your mail queue. There's probably a bunch of un-sendable crud
that's
> choking up the queue. "mailq" is your friend.
> 2. Your users are being hammered with new viruses that I mentioned in
#1. Add some sort of filter to clean your incoming mail: procmail,
> spamassassin, amavis or a mixture of all three :)
>
> What mail server software are you running? Sendmai, Postfix, Qmail
(*cough*, choke), Exim?? I could offer some more specific help if I
knew :)
>
> Good luck.
>
> --James
> __________________________________
> A random quote of nothing:
>
> Cache:
> A very expensive part of the memory system of a computer that no one is
supposed to know is there.
>
--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
