[plug] File System Errors - Take II

[Andrew Barbara]

Hi,
Thanks for the ideas, i'm no sure if there is any problems with this, but here is the output of df -h and df -hi.
Any problems?


[root at gateway /]# df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/hda1             2.9G  100M  2.6G   4% /
/dev/hda6             2.9G   20k  2.7G   1% /cache
/dev/hda9             486M  100M  361M  22% /home
/dev/hda7             2.9G  691M  2.0G  25% /usr
/dev/hda5             2.9G  180M  2.5G   7% /var
[root at gateway /]# df -hi
Filesystem            Inodes   IUsed   IFree IUse% Mounted on
/dev/hda1               376k     18k    358k    5% /
/dev/hda6               376k      11    375k    1% /cache
/dev/hda9               125k    3.1k    122k    3% /home
/dev/hda7               376k     47k    328k   13% /usr
/dev/hda5               376k     17k    359k    5% /var
[root at gateway /]#

It looks like thers stackes of inodes and free space...
What else could cause my problems?
Thabks
Andrew

>>> steveg at calm.wa.gov.au 08/17/01 16:30 PM >>>
Sorry, my previous was probably a bit brief. To verify inode or space on 
partitions:

df -hi 

will tell you your inode status for each partition while

df -h

will tell you your diskspace allocation/availability for each partition.

Generally, with respect to /var if you're short on inodes, but have tons of 
free space, suspect miriads of small files (eg. error messages sent to root - 
either fix the thing(s) generating the messages, or clear out the messages 
periodically). If you're short on space but have tons of inodes, suspect huge 
logs. Similarly, check the logs to find the source of messages & fix that, or 
move/clear old logs periodically.

Regards,
Steve

On Friday 17 August 2001 15:48, Steve Grasso wrote:
> Andrew,
>
> This sounds more lke the /var partition either running out of inodes or
> diskspace. My suggestion for a quick-fix would be to set up a cron job to
> periodically move old log files to another partition/machine, or delete
> them, and to check in /var/spool/mqueue for runaway error messages to root.
> Set up a cron to move or clear these out also.
>
> Regards,
> Steve
>
> On Friday 17 August 2001 15:51, Andrew Barbara wrote:
> > Hi,
> > One of our client's linux RedHat 7.1 server is Jamming up.  When I
> > Re-Boot it takes me to a file system repair shell, I run fsck on the
> > troubled file system (/var) and reboot then all is fine for a couple of
> > weeks.  This is the second time its happened.  Is it a faulty Hard Drive?
> > Or a linux issue?
> >
> > Thanks
> > Andrew Barbara