Hey guys and gals,
Yesterday I had one of my scientists kill one of my servers when his
program ran amok and gobbled up all the memory, or forked too many
processes, or I'm just not exactly sure what to be honest.
Is there something I can run manually in cron to look for rampant
programs and kill them? I know that may be hard to discern but I
could also include a list if "known good" programs not to kill, as
well as a list of "known suspect" user IDs
Anyone ever done this? Searching the list on "OOM" does not bring up
much.
thanks
-Alan
--
?Don't eat anything you've ever seen advertised on TV?
- Michael Pollan, author of "In Defense of Food"
* Alan McKay <alan.mckay at gmail.com> [07/25/2012 09:36]:> Yesterday I had one of my scientists kill one of my servers when his > program ran amok and gobbled up all the memory, or forked too many > processes, or I'm just not exactly sure what to be honest. > > Is there something I can run manually in cron to look for rampant > programs and kill them? I know that may be hard to discern but I > could also include a list if "known good" programs not to kill, as > well as a list of "known suspect" user IDs > > Anyone ever done this? Searching the list on "OOM" does not bring up much.I would look into disabling memory overcommit. On our compute cluster nodes we use the following sysctls: # disable overcommit (allocation_limit=swap + overcommit_ratio/100 * phys_mem) vm.overcommit_ratio=70 vm.overcommit_memory=2 This will prevent programs to allocate more than allocation_limit of memory. In this way the program that tries to allocate the memory that would put the memory usage over the limit will be prevented to do so. The malloc will return NULL. For the forking too many processes you can look into man limits.conf where you can limit the maximum number of processes per user. Daniel.
Alan McKay wrote:> Hey guys and gals, > > Yesterday I had one of my scientists kill one of my servers when his > program ran amok and gobbled up all the memory, or forked too many > processes, or I'm just not exactly sure what to be honest. > > Is there something I can run manually in cron to look for rampant > programs and kill them? I know that may be hard to discern but I > could also include a list if "known good" programs not to kill, as > well as a list of "known suspect" user IDs > > Anyone ever done this? Searching the list on "OOM" does not bring up > much.Yeah, we've had that a few times, even on 64 core systems with a ridiculous amount of memory. One thing we did was to tell them to limit the number of cores they were using in the parallel processing threads. There is some kind of limit you can set up - I forget exactly what it is, but I'm not sure it will limit memory usage. Someone suggested yesterday, in another context, giving 'em a VM of their own. Doing that, you can limit how many cores and how much memory they have, and then if they crash, it's more their problem than yours. Or have them buy another server and make it a cluster. We use the torque package. mark
From: Alan McKay <alan.mckay at gmail.com>> Yesterday I had one of my scientists kill one of my servers when his > program ran amok and gobbled up all the memory, or forked too many > processes, or I'm just not exactly sure what to be honest. > > Is there something I can run manually in cron to look for rampant > programs and kill them?? I know that may be hard to discern but I > could also include a list if "known good" programs not to kill, as > well as a list of "known suspect" user IDsWhat about setting limits...? /etc/security/limits.conf JD
On 7/25/2012 7:36 AM, Alan McKay wrote:> > Is there something I can run manually in cron to look for rampant > programs and kill them?You appear to be under the impression that you have a technical problem. What you actually have is a people problem. Go now, and kneel at the feet of the Bastard Operator From Hell (http://bofh.ntk.net/) to learn how to deal with such matters.