Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Hello, I have a primary/secondary node with 2 ressources. When I watch the i/O performance i look some think like this : When I watch the physical drive utilisation : root at relax:~# iostat -x /dev/sda 1 Linux 2.6.32-5-amd64 (relax) 01/19/2012 _x86_64_ (16 CPU) avg-cpu: %user %nice %system %iowait %steal %idle 0.02 0.00 1.09 0.02 0.00 98.87 Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util sda 220.78 700.91 555.52 374.88 20794.66 21227.05 45.16 1.54 1.66 0.42 38.87 avg-cpu: %user %nice %system %iowait %steal %idle 0.04 0.00 0.17 0.00 0.00 99.79 Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util sda 186.00 48.00 211.00 162.00 15392.00 2624.00 48.30 0.03 0.09 0.06 2.40 avg-cpu: %user %nice %system %iowait %steal %idle 0.00 0.00 0.29 0.00 0.00 99.71 Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util sda 197.00 9.00 379.00 117.00 19472.00 1172.00 41.62 0.15 0.31 0.19 9.60 When I watch the drbd device utilisation : root at relax:~# iostat -x /dev/drbd0 1 Linux 2.6.32-5-amd64 (relax) 01/19/2012 _x86_64_ (16 CPU) avg-cpu: %user %nice %system %iowait %steal %idle 0.02 0.00 1.09 0.02 0.00 98.87 Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util drbd0 0.00 0.00 775.16 1016.47 20729.96 21174.70 23.39 0.94 5.51 0.56 99.85 avg-cpu: %user %nice %system %iowait %steal %idle 0.00 0.00 0.56 0.00 0.00 99.44 Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util drbd0 0.00 0.00 426.00 104.00 15648.00 1262.00 31.91 5.19 0.15 1.87 99.20 avg-cpu: %user %nice %system %iowait %steal %idle 0.00 0.00 0.56 0.00 0.00 99.44 Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util drbd0 0.00 0.00 404.00 148.00 15888.00 1872.00 32.17 5.78 0.33 1.80 99.60 avg-cpu: %user %nice %system %iowait %steal %idle 0.00 0.00 0.31 0.00 0.00 99.69 Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util drbd0 0.00 0.00 398.00 117.00 15632.00 1736.00 33.72 5.28 0.10 1.94 100.00 avg-cpu: %user %nice %system %iowait %steal %idle 0.00 0.00 0.24 0.00 0.00 99.76 Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util drbd0 0.00 0.00 406.00 106.00 15376.00 1382.00 32.73 5.07 0.17 1.92 98.40 I don't understand why my drbd device is at 100 %util. This is my ressource configuration : There are a raid 0 hardware with 12 sas 15k drives. resource data { protocol C; startup { wfc-timeout 0; } disk { on-io-error detach; } syncer { rate 20M; verify-alg md5; } on surtax { device /dev/drbd0; disk /dev/sda; address 10.1.42.11:7788; meta-disk internal; } on relax { device /dev/drbd0; disk /dev/sda; address 10.1.42.12:7788; meta-disk internal; } } What's wrong on my configuration ? Thank's Matthieu