pl help out .Server is up but of no use !!!need expert like u 2 advice

[withu]

Dear All,

I have a Sun solaris server with sun OS 5.8. My server is configured as mailserver + dns server.
The DNS parts work fine but the mailservices are totally down.My observation are as follows :-

syslog messege /var/log/syslog :-

"sendmail[26884]: [ID 801593 mail.alert] NOQUEUE: low on space (have 0, SMTP-DAEMON needs 101 in /var/spool/mqueue"

So many times this messege is there..round about 100000 times ...No idea what it means and how it came.

And when I tried to c the /var/adm/messege

it is filled with soooooooo many messegestarting from 29th 2.00p.m onward for each second it has this messege throughout:-
#######
29 14:43:35 inetd[18944]: [ID 455825 daemon.error]getpwnam: nobody: No such user
May 29 14:43:35 stpc inetd[337]: [ID 858011 daemon.warning]
/usr/openwin/lib/fs.auto: Hangup
May 29 14:43:35 stpc inetd[18945]: [ID 455825 daemon.error] getpwnam: nobody:No such user
######

After going through google and searching for prob i.e need 101 in var/spool/mqueue > I got something and did the mount and then reboot.After reboot system used to hang up at ctrl d i.e single user only. And while rebotting it gave this
"cannot create /var/adm/utmpx"
Somehow with continuous "fsck" able to come to dt session
but problem was as it is.No mail transaction possible and syslog was giving same messege. :-(

when gave df -k it says the following:-

Filesystem kbytes used avail capacity Mounted
/dev/dsk/c0t0d0s0 1018191 990493 0 100% /
/dev/dsk/c0t0d0s3 1489367 1047297 382496 74% /usr
/proc 0 0 0 0% /proc
fd 0 0 0 0% /dev/fd
mnttab 0 0 0 0% /etc/mnttab
swap 1326632 0 1326632 0% /var/run
swap 1327016 384 1326632 1% /tmp
/dev/dsk/c0t0d0s4 4131866 313407 3777141 8% /opt
/dev/dsk/c0t0d0s5 4922590 273379 4599986 6% /user1
/dev/dsk/c0t0d0s7 4776538 1120036 3608737 24% /export/home

while seeing the ls -l for root (/) I saw one archive11629.tar file of about 1Gb.So I deleted this but then also the df -k showing 100% for /.

Now I am really fed up and don't know how to go ahead. Please help to come out from this.

With best regards and hope to hear from u,
Withu is signing out from ur screen.

 

[bigdaddy01]

Hmmm thats a tough one, clear out root for a start ! the error cannot create /var/adm/utmpx"
is definately a space issue I have had that before. Thats all I am sure of.

 

[bi]

I would recommend killing your sendmail daemon while you get this fixed. I suspect it keeps filling up any space you have in / with messages to syslog and perhaps its own log.

Also, stop the sylogd and move the current message file and any others to another filesystem. It must be huge by now. Then restart syslog.

Then continue to clean out root.

At /, use du -sk * | sort -n to see what files and directories are taking up the most room. I suspect your /var/spool/mqueue directory has a lot in there.

It would be good, if possible, to add more space to /. You have a huge amount for /opt, /user1 and /export/home that isn't being used. If you could rebuild /user1 or /export/home (causing downtime, unfortunately) to make them smaller and allocate the freed space to /, you may be able to avoid this in the future.

(I would have made /var its own filesystem, but I have noticed a lot of Sun sys admins don't for some reason.)

Good luck.