Description of problem: While creating devspaces for MaxDB 7.5.0 oom kills MaxDB processes/threads after 30-50GB were written to disk. Enabling USE_OPEN_DIRECT on MaxDB works fine, although after 20GB the writing speed drops from 70MB/sec to 2MB/sec. Version-Release number of selected component (if applicable): kernel-smp-2.6.9-11.EL How reproducible: Everytime Steps to Reproduce: 1. Download MaxDB 7.5 from http://dev.mysql.com/downloads/maxdb/7.5.00.html 2. Install software wich SDBINST 3. Modify create_demo_db.sh script: replace "param_addvolume 1 DATA DISKD0001 F 2560" with "param_addvolume 1 DATA DISKD0001 F 8544921" (which is around 70GB in 8k blocks). 4. start the script Actual results: All MaxDB processes/threads were killed by oom Expected results: oom doing nothing... Additional info: Hardware information: Dell PowerEdge 6650 4 x 3Ghz Xeon HT RAID bus controller: LSI Logic / Symbios Logic MegaRAID (rev 01) external SCSI enclosure with 12x36GB disks (RAID 5) Software information: # lsmod Module Size Used by nfs 200869 1 lockd 65257 2 nfs vfat 16961 0 fat 44129 1 vfat md5 8001 1 ipv6 238817 28 autofs4 22085 0 i2c_dev 14273 0 i2c_core 25921 1 i2c_dev sunrpc 138789 4 nfs,lockd dm_mod 58949 0 button 10449 0 battery 12869 0 ac 8773 0 tg3 82373 0 floppy 58065 0 sg 38113 0 ext3 118729 3 jbd 59481 1 ext3 qla2300 126529 0 qla2xxx 109665 1 qla2300 scsi_transport_fc 11712 1 qla2xxx megaraid_mbox 38097 4 megaraid_mm 17905 1 megaraid_mbox aic7xxx 146553 0 sd_mod 20545 6 scsi_mod 116429 6 sg,qla2xxx,scsi_transport_fc,megaraid_mbox,aic7xxx,sd_mod #dmesg oom-killer: gfp_mask=0xd0 Mem-info: DMA per-cpu: cpu 0 hot: low 2, high 6, batch 1 cpu 0 cold: low 0, high 2, batch 1 cpu 1 hot: low 2, high 6, batch 1 cpu 1 cold: low 0, high 2, batch 1 cpu 2 hot: low 2, high 6, batch 1 cpu 2 cold: low 0, high 2, batch 1 cpu 3 hot: low 2, high 6, batch 1 cpu 3 cold: low 0, high 2, batch 1 cpu 4 hot: low 2, high 6, batch 1 cpu 4 cold: low 0, high 2, batch 1 cpu 5 hot: low 2, high 6, batch 1 cpu 5 cold: low 0, high 2, batch 1 cpu 6 hot: low 2, high 6, batch 1 cpu 6 cold: low 0, high 2, batch 1 cpu 7 hot: low 2, high 6, batch 1 cpu 7 cold: low 0, high 2, batch 1 Normal per-cpu: cpu 0 hot: low 32, high 96, batch 16 cpu 0 cold: low 0, high 32, batch 16 cpu 1 hot: low 32, high 96, batch 16 cpu 1 cold: low 0, high 32, batch 16 cpu 2 hot: low 32, high 96, batch 16 cpu 2 cold: low 0, high 32, batch 16 cpu 3 hot: low 32, high 96, batch 16 cpu 3 cold: low 0, high 32, batch 16 cpu 4 hot: low 32, high 96, batch 16 cpu 4 cold: low 0, high 32, batch 16 cpu 5 hot: low 32, high 96, batch 16 cpu 5 cold: low 0, high 32, batch 16 cpu 6 hot: low 32, high 96, batch 16 cpu 6 cold: low 0, high 32, batch 16 cpu 7 hot: low 32, high 96, batch 16 cpu 7 cold: low 0, high 32, batch 16 HighMem per-cpu: cpu 0 hot: low 32, high 96, batch 16 cpu 0 cold: low 0, high 32, batch 16 cpu 1 hot: low 32, high 96, batch 16 cpu 1 cold: low 0, high 32, batch 16 cpu 2 hot: low 32, high 96, batch 16 cpu 2 cold: low 0, high 32, batch 16 cpu 3 hot: low 32, high 96, batch 16 cpu 3 cold: low 0, high 32, batch 16 cpu 4 hot: low 32, high 96, batch 16 cpu 4 cold: low 0, high 32, batch 16 cpu 5 hot: low 32, high 96, batch 16 cpu 5 cold: low 0, high 32, batch 16 cpu 6 hot: low 32, high 96, batch 16 cpu 6 cold: low 0, high 32, batch 16 cpu 7 hot: low 32, high 96, batch 16 cpu 7 cold: low 0, high 32, batch 16 Free pages: 15136kB (1600kB HighMem) Active:285350 inactive:1668227 dirty:219220 writeback:113476 unstable:0 free:3784 slab:49928 mapped:283285 pagetables:1789 DMA free:12640kB min:16kB low:32kB high:48kB active:0kB inactive:0kB present:16384kB pages_scanned:20983 all_unreclaimable? yes protections[]: 0 0 0 Normal free:896kB min:928kB low:1856kB high:2784kB active:8252kB inactive:615108kB present:901120kB pages_scanned:817047 all_unreclaimable? yes protections[]: 0 0 0 HighMem free:1600kB min:512kB low:1024kB high:1536kB active:1133148kB inactive:6057800kB present:7471104kB pages_scanned:0 all_unreclaimable? no protections[]: 0 0 0 DMA: 4*4kB 4*8kB 3*16kB 4*32kB 4*64kB 1*128kB 1*256kB 1*512kB 1*1024kB 1*2048kB 2*4096kB = 12640kB Normal: 0*4kB 104*8kB 4*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 896kB HighMem: 98*4kB 39*8kB 22*16kB 1*32kB 0*64kB 0*128kB 0*256kB 1*512kB 0*1024kB 0*2048kB 0*4096kB = 1600kB Swap cache: add 21638, delete 21560, find 413/618, race 0+0 Free swap: 10190220kB 2097152 pages of RAM 1802208 pages of HIGHMEM 83433 reserved pages 1695143 pages shared 78 pages swap cached Out of Memory: Killed process 3505 (kernel). Out of Memory: Killed process 3506 (kernel). Out of Memory: Killed process 3507 (kernel). Out of Memory: Killed process 3508 (kernel). Out of Memory: Killed process 3509 (kernel). Out of Memory: Killed process 3510 (kernel). Out of Memory: Killed process 3511 (kernel). Out of Memory: Killed process 3512 (kernel). Out of Memory: Killed process 3513 (kernel). Out of Memory: Killed process 3514 (kernel). Out of Memory: Killed process 3515 (kernel). Out of Memory: Killed process 3516 (kernel). Out of Memory: Killed process 3517 (kernel). Out of Memory: Killed process 3518 (kernel). Out of Memory: Killed process 3519 (kernel). Out of Memory: Killed process 3520 (kernel). Out of Memory: Killed process 3521 (kernel). Out of Memory: Killed process 3522 (kernel). Out of Memory: Killed process 3523 (kernel). Out of Memory: Killed process 3524 (kernel). Out of Memory: Killed process 3525 (kernel). Out of Memory: Killed process 3526 (kernel). Out of Memory: Killed process 3527 (kernel). Out of Memory: Killed process 3528 (kernel). Out of Memory: Killed process 3535 (kernel). Out of Memory: Killed process 3536 (kernel). Out of Memory: Killed process 3539 (kernel). Out of Memory: Killed process 3540 (kernel). Out of Memory: Killed process 3737 (kernel). Out of Memory: Killed process 3738 (kernel). Fixed up OOM kill of mm-less task
Hannes, please try the latest RHEL4-U2 beta kernel and see if this problem is solved. We have made several changes to improve the OOM kill problems as well as adding more debug output when it does happen so we can beter determine exactly why this is happening. Thanks, Larry Woodman
Hi Larry, kernel-smp-2.6.9-15.EL works fine. No kills of oom killer anymore. Cheers Hannes