home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.sgi
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!magnus.acs.ohio-state.edu!usenet.ins.cwru.edu!wsu-cs!igor.physics.wayne.edu!atems
- From: atems@igor.physics.wayne.edu (Dale Atems)
- Subject: Who gets killed?
- Message-ID: <1992Nov9.053507.5033@cs.wayne.edu>
- Sender: usenet@cs.wayne.edu (Usenet News)
- Organization: Wayne State University, Detroit, MI
- References: <1992Nov4.065527.4867@shannon.ee.wits.ac.za> <1992Nov4.163355.12112@odin.corp.sgi.com> <3112@contex.contex.com>
- Date: Mon, 9 Nov 1992 05:35:07 GMT
- Lines: 24
-
- In article <3112@contex.contex.com> frank@marvin.contex.com (Frank Perdicaro) writes:
- >In article <1992Nov4.163355.12112@odin.corp.sgi.com> martyi@sgi.com writes:
- >>Usually the message means that the system ran out of swap space, and killed
- >>the process. While it is typically the process with the high demand for
- >>swap that is killed, it is not always: the system is desperate, and strikes
- >>out at someone.
-
- I once watched helplessly while IRIX killed most of the processes on the system
- trying to get rid of a memory pig. The thing is, it already had, but it wouldn't
- quit. The system was a 16MB Indigo running 4.0.5F with 40 MB swap space, and I
- had tried to run a Fortran program that by mistake (mine, that is) would have
- needed over 300MB to run in. The offending program was killed within a minute.
- After that it was slow death, I tried to stop the process remotely but was
- thrown off. The last entry in the syslog said that klogpp was killed due to
- insufficient swap space.
-
- BTW, I was running as an ordinary user, not root. Should I have reported this
- to the TAC?
-
- ------
- Dale Atems
- Wayne State University, Detroit, MI
- Department of Physics and Astronomy
- atems@igor.physics.wayne.edu
-