home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.hp
- Path: sparky!uunet!stanford.edu!CSD-NewsHost.Stanford.EDU!seligman
- From: seligman@CS.Stanford.EDU (Scott Seligman)
- Subject: Re: /usr/adm/diag ?
- In-Reply-To: slu@paul.rutgers.edu's message of 22 Jul 92 14:23:00 GMT
- Message-ID: <1992Jul26.014653.7134@CSD-NewsHost.Stanford.EDU>
- Sender: news@CSD-NewsHost.Stanford.EDU
- Organization: Stanford University
- References: <Jul.22.10.23.00.1992.8293@paul.rutgers.edu>
- Distribution: usa
- Date: Sun, 26 Jul 1992 01:46:53 GMT
- Lines: 32
-
- In article <Jul.22.10.23.00.1992.8293@paul.rutgers.edu> slu@paul.rutgers.edu (shen lu) writes:
- >
- > I noticed that there were a lot of big files in the directory
- > /usr/adm/diag, by name of LOG00**. (Our systems is a HP817S
- > Running HP-UX 8.02. ) .
- > What are there files? Is there any reason to keep those files?
- > They sum up to a total about 20 megas.
-
-
- The same happens on our 705 & 720 systems running 8.07. Every 30
- minutes (exactly) some process wakes up on each machine, and does
- something that causes the LOG* files to grow by about 5 kilobytes
- each. This adds up quickly.
-
- DUI / LOGTOOL doesn't give much info. It shows a zillion repetitions of
- something to the effect of "unable to open device file ()" (note the
- empty parentheses).
-
- The culprit is one of the following processes. I'm not sure how to
- narrow it down much further. Note that "cron" is not running anything
- at 30 minute intervals.
-
- swapper init vhand statdaemon unhashdaemon lcsp syncdaemon gcsp MEMLOGP
- syslogd portmap nfsd inetd rpc.lockd snmpd cron ptydaemon DIAGMON DIAGINIT
-
- Our fix was to "chmod -x /usr/diag/bin/DIAGINIT". :-(
-
-
- Scott Seligman
-
- Internet: seligman@CS.Stanford.EDU
- UUCP: {uunet,decwrl,sun}!cs.stanford.edu!seligman
-