hp-ux nettl tracing won't start

[Kevin Batter_4]

I am trying to start nettl tracing on a HP-UX 11.11 server and the command fails as follows:

# nettl -start
# tail /var/adm/syslog/syslog.log
Aug 31 10:25:18 myserver nettl[18937]: nettl starting up.

/usr/sbin/nettl -tn pduin pduout -e ns_ls_ip -tm 99999 -f /var/tmp/trace

nettl : Warning: Individual trace file size should be a multiple
of user buffer size. Hence, the maximum trace file size
is adjusted to 100096 Kbytes.

nettl : Unable to detect the daemon process which receives
trace and log information from kernel subsystems.

# tail /var/adm/syslog/syslog.log
Aug 31 10:26:09 myserver nktl_daemon[19036]: Fail to P semaphore
Aug 31 10:26:09 myserver nktl_daemon[19036]: Errno(28): No space left on device

Semaphores look OK (from glance -t):
--------------------------------------------------------------------------------
SYSTEM TABLES REPORT Users= 30

System Table Available Used Utilization High(%)
--------------------------------------------------------------------------------
Proc Table (nproc) 4116 308 7 7
File Table (nfile) 10010 2391 24 24
Shared Mem Table (shmmni) 200 17 9 9
Message Table (msgmni) 50 2 4 4
Semaphore Table (semmni) 64 37 58 58
File Locks (nflocks) 1000 75 8 8
Pseudo Terminals (npty) 200 11 6 6
Buffer Headers (nbuf) na 511224 na na

Any ideas what may be causing this?

[Peter Godron]

Kevin,
what is the result of:
ll /var/tmp

[Kevin Batter_4]

ll /var/tmp just displays all the files in /var/tmp

# ls -ld /var/tmp
drwxrwxrwx 15 bin bin 82944 Aug 31 10:47 /var/tmp

bdf /var/tmp
Filesystem kbytes used avail %used Mounted on
/dev/vg00/lvol8 1572864 1142631 404681 74% /var

Does this help at all?

I suspect this may be shared memory or semaphore related but the tables look ok.

[Peter Godron]

Hi,
please check the solutions in:
http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=991202
and also ( without solution )
http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=690836