linux服务器上运行的informix数据库性能问题
系统运行的是informix数据库，目前发现IO等待非常严重，就是不知道从哪里下手开始查起，请大侠们指教。-bash-3.1$ uname -a
Linux tnms213 2.6.18-53.el5 #1 SMP Wed Oct 10 16:34:19 EDT 2007 x86_64 x86_64 x86_64 GNU/Linux
top - 15:26:08 up 1 day,1:56,6 users,load average: 17.25, 17.16, 17.57
Tasks: 182 total, 3 running, 179 sleeping, 0 stopped, 0 zombie
Cpu(s): 26.8%us,9.2%sy,0.0%ni,6.4%id, 55.7%wa,0.2%hi,1.7%si,0.0%st
Mem: 8174472k total,7942416k used, 232056k free, 48428k buffers
Swap: 20482864k total, 474436k used, 20008428k free,7379776k cached
PID USER PRNIVIRTRESSHR S %CPU %MEM TIME+COMMAND
16916 informix11 -10 1187m 392m 389m R 864.958:02.21 oninit
16937 root 5 -10 1186m 242m 241m D 73.0 1:17.54 oninit
16939 root 5 -10 1186m 241m 240m S 73.0 1:22.63 oninit
17205 root 5 -10 1186m 234m 233m D 52.9 1:10.48 oninit
3842 root 10-5 0 0 0 D 40.010:02.02 kjournald
16936 root 6 -10 1186m 240m 239m D 43.0 1:12.98 oninit
16933 root 6 -10 1186m 241m 240m R 33.0 1:08.76 oninit
16938 root 5 -10 1186m 241m 240m D 33.0 1:18.65 oninit
16927 root 5 -10 1186m 243m 241m D 33.0 1:18.78 oninit
17233 root 5 -10 1186m 231m 230m S 32.9 1:00.97 oninit
16920 root 5 -10 1186m 249m 248m D 33.1 2:36.14 oninit
16925 root 5 -10 1186m 8832 7480 S 30.1 0:55.82 oninit
16934 root 6 -10 1186m 241m 240m S 23.0 1:13.44 oninit
16924 root 5 -10 1186m 246m 245m D 23.1 1:38.15 oninit
16932 root 5 -10 1186m 241m 240m D 23.0 1:10.93 oninit
16935 root 5 -10 1186m 243m 241m D 23.0 1:12.90 oninit
17262 root 5 -10 1186m 228m 227m S 22.9 0:45.16 oninit
16921 root 5 -10 1186m 8568 7200 S 10.1 1:03.22 oninit
16928 root 5 -10 1186m 243m 242m D 13.1 1:15.85 oninit
16926 root 5 -10 1186m 245m 244m D 13.1 1:26.32 oninit
16931 root 5 -10 1186m 242m 240m D 13.0 1:11.12 oninit
17605 informix15 0 12716 1120792 R 10.0 0:02.25 top
16929 root 5 -10 1186m 243m 241m D 03.0 1:16.23 oninit
16930 root 5 -10 1186m 242m 240m D 03.0 1:11.67 oninit
16977 informix16 0 12716 1120792 S 00.0 0:25.51 top
17306 root 5 -10 1186m 226m 225m S 02.8 0:29.27 oninit
17654 root 15 0 0 0 0 D 00.0 0:00.01 pdflush
17657 root 15 0 0 0 0 D 00.0 0:00.02 pdflush
1 root 15 0 10316676568 S 00.0 0:01.16 init
2 root RT 0 0 0 0 S 00.0 0:00.07 migration/0
3 root 3419 0 0 0 S 00.0 0:00.00 ksoftirqd/0
4 root RT 0 0 0 0 S 00.0 0:00.00 watchdog/0
5 root RT 0 0 0 0 S 00.0 0:02.64 migration/1
6 root 3419 0 0 0 S 00.0 0:00.36 ksoftirqd/1
7 root RT 0 0 0 0 S 00.0 0:00.00 watchdog/1
8 root RT 0 0 0 0 S 00.0 0:00.03 migration/2
9 root 3419 0 0 0 S 00.0 0:00.01 ksoftirqd/2
10 root RT 0 0 0 0 S 00.0 0:00.00 watchdog/2
11 root RT 0 0 0 0 S 00.0 0:00.04 migration/3
12 root 3419 0 0 0 S 00.0 0:00.00 ksoftirqd/3
13 root RT 0 0 0 0 S 00.0 0:00.00 watchdog/3
14 root 10-5 0 0 0 S 00.0 0:00.16 events/0
15 root 10-5 0 0 0 S 00.0 0:00.62 events/1
16 root 10-5 0 0 0 S 00.0 0:00.15 events/2
17 root 10-5 0 0 0 S 00.0 0:00.14 events/3
-bash-3.1$ sar 1 10
Linux 2.6.18-53.el5 (tnms213) 2012年10月17日
15时26分42秒 CPU %user %nice %system %iowait %steal %idle
15时26分43秒 all 20.60 0.00 8.44 62.78 0.00 8.19
15时26分44秒 all 19.40 0.00 9.20 58.96 0.00 12.44
15时26分45秒 all 19.35 0.00 7.04 65.83 0.00 7.79
15时26分46秒 all 20.25 0.00 9.75 59.00 0.00 11.00
15时26分47秒 all 17.41 0.00 7.71 59.45 0.00 15.42
15时26分48秒 all 16.79 0.00 8.27 65.16 0.00 9.77
15时26分49秒 all 21.50 0.00 8.50 68.75 0.00 1.25
15时26分50秒 all 20.95 0.00 7.48 70.07 0.00 1.50
15时26分51秒 all 19.45 0.00 10.47 69.08 0.00 1.00
15时26分52秒 all 21.80 0.00 10.53 67.42 0.00 0.25
Average: all 19.75 0.00 8.74 64.64 0.00 6.87
在这里我发现有几个内存段状态一直是locked状态，重启服务器，过会还是这个样子，不知道怎么回事？
-bash-3.1$ ipcs -m
------ Shared Memory Segments --------
key shmid owner perms bytes nattch status
0x00000000 1015809 root 660 1064308736 28 dest locked
0x00000000 1048578 root 660 33439744 28 dest locked
0x52564803 1081347 informix660 8388608 28 locked
0x52564804 1114116 informix660 8388608 28 locked
0x52564805 1146885 informix660 8388608 28
0x52564806 1179654 informix660 8388608 28
0x52564807 1212423 informix660 8388608 28
0x52564808 1245192 informix660 8388608 28
0x52564809 1277961 informix660 8388608 28
0x5256480a 1310730 informix660 8388608 28
0x5256480b 1343499 informix660 8388608 28
0x5256480c 1376268 informix660 8388608 28
0x5256480d 1409037 informix660 8388608 28
0x5256480e 1441806 informix660 8388608 28
0x5256480f 1474575 informix660 8388608 28
0x52564810 1507344 informix660 8388608 28
0x52564811 1540113 informix660 8388608 28
0x52564812 1572882 informix660 8388608 28 可以查看一下onstat -g act|grep "sqlexec"|grep "running"里的这些SESSION，看看是不是有全表扫描。
页:
[1]