![]() |
|
|
google unix.com
|
|||||||
| Forums | Register | Forum Rules | Links | Albums | FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
| UNIX for Advanced & Expert Users Expert-to-Expert. Learn advanced UNIX, UNIX commands, Linux, Operating Systems, System Administration, Programming, Shell, Shell Scripts, Solaris, Linux, HP-UX, AIX, OS X, BSD. |
More UNIX and Linux Forum Topics You Might Find Helpful
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| WolframAlpha a Google Killer? Not! | iBot | Complex Event Processing RSS News | 0 | 05-19-2009 09:45 AM |
| Using ZFS to Fight Data Rot, the Silent Killer | iBot | Solaris BigAdmin RSS | 0 | 01-29-2009 09:30 AM |
| Linux Foundation launches killer development tool | iBot | UNIX and Linux RSS News | 0 | 08-07-2008 01:40 PM |
| help rsh killer | gonzo | Shell Programming and Scripting | 3 | 04-23-2008 04:06 AM |
| Webinar: BAM: The Killer App for CEP | iBot | Complex Event Processing RSS News | 0 | 01-31-2008 09:50 AM |
![]() |
|
|
LinkBack | Thread Tools | Search this Thread | Rate Thread | Display Modes |
|
|
|
||||
|
OOM KIller INVOKE
Hi
I have written a code that will exhaust the memory completely by allocating it and not freeing it. The OOM killer as expected in invoked and it killed a few high memory using processes and their childs. But after that, though the system is not getting hung, it is not printing any msgs in console neither it is responding to ping or telnet. What is the expected behaviour after an OOM kill. ? Please comment on why the node went into an virtual hung state after OOM kill. ---------- Post updated at 05:46 AM ---------- Previous update was at 05:42 AM ---------- This is the console prints: ================================================================= available memory stated is 487MB And malloc'd memory is 0MB Enter percentage memory to be alloc'd (0 to quit) : 100 available memory stated is 487MB And malloc'd memory is 487MB malloc invoked oom-killer: gfp_mask=0x1200d2, order=0, oomkilladj=0 Call Trace: [9e913d20] [80006844] show_stack+0x3c/0x1bc (unreliable) [9e913d60] [80056a14] oom_kill_process+0x114/0x1bc [9e913d90] [80056e60] out_of_memory+0x1ac/0x1e8 [9e913dd0] [800598f0] __alloc_pages_internal+0x40c/0x440 [9e913e50] [8006424c] handle_mm_fault+0x26c/0x470 [9e913e80] [801ccb58] do_page_fault+0x314/0x510 [9e913f40] [8000d834] handle_page_fault+0xc/0x80 Mem-info: DMA per-cpu: CPU 0: hi: 186, btch: 31 usd: 175 Active:123608 inactive:31 dirty:0 writeback:0 unstable:0 free:689 slab:1244 mapped:1259 pagetables:311 bounce:0 DMA free:2756kB min:2860kB low:3572kB high:4288kB active:494432kB inactive:124kB present:512064kB pages_scanned:772531 all_unreclaimable? yes lowmem_reserve[]: 0 0 0 DMA: 1*4kB 0*8kB 0*16kB 0*32kB 1*64kB 1*128kB 0*256kB 1*512kB 0*1024kB 1*2048kB 0*4096kB = 2756kB 1292 total pagecache pages 129024 pages of RAM 1730 reserved pages 7284 pages shared 0 pages swap cached Out of memory: kill process 2021 (tn_init.d) score 162152 or a child Killed process 2023 (nm.d) malloc invoked oom-killer: gfp_mask=0x1200d2, order=0, oomkilladj=0 Call Trace: [9e913d20] [80006844] show_stack+0x3c/0x1bc (unreliable) [9e913d60] [80056a14] oom_kill_process+0x114/0x1bc [9e913d90] [80056e60] out_of_memory+0x1ac/0x1e8 [9e913dd0] [800598f0] __alloc_pages_internal+0x40c/0x440 [9e913e50] [8006424c] handle_mm_fault+0x26c/0x470 [9e913e80] [801ccb58] do_page_fault+0x314/0x510 [9e913f40] [8000d834] handle_page_fault+0xc/0x80 Mem-info: DMA per-cpu: CPU 0: hi: 186, btch: 31 usd: 176 Active:123603 inactive:36 dirty:0 writeback:0 unstable:0 free:783 slab:1240 mapped:1259 pagetables:311 bounce:0 DMA free:3132kB min:2860kB low:3572kB high:4288kB active:494412kB inactive:144kB present:512064kB pages_scanned:0 all_unreclaimable? no lowmem_reserve[]: 0 0 0 DMA: 1*4kB 3*8kB 4*16kB 5*32kB 3*64kB 1*128kB 0*256kB 1*512kB 0*1024kB 1*2048kB 0*4096kB = 3132kB 1292 total pagecache pages 129024 pages of RAM 1730 reserved pages 7284 pages shared 0 pages swap cached Out of memory: kill process 2039 (nm.d) score 13684 or a child Killed process 2039 (nm.d) ^C^C^C^C^C^C^C^Cc ================================================================ As seen its not responding to ^C . But on pressing enter , response is there. |
|
||||
|
There is no expected behavour once time Oom Killer starts, it's a lottery. He could kill ANY process in order to free memory while is trying to keep the server up...
You could find more informacion in next link OOM Killer - linux-mm.org Wiki I recommed you to setup netdump or diskdump and you could make a memory dump to analize after hung. Maybe you will discover a memory leak ![]() |
![]() |
| Bookmarks |
| Tags |
| oom killer |
| Thread Tools | Search this Thread |
| Display Modes | Rate This Thread |
|
|