Performance degrade suddenly
Environment: HP 11.11 IDS 9.30fc4w3
The performance is degrade suddenly in a very heavy OLTP environment, when the
issue is occured we can see more mutex wait on systables, the average responce
time of the per transcation is slow from 0.2 sec to 0.6 sec.
We didn't see any obvious error, the only feel is the system change to slow
and the cpu's usage is high.
We didn't find the root cause yet, but the problem can be relaxed if we bounce
the instance regularly.
the following information is from onstat -g wmx:
Informix Dynamic Server Version 9.30.FC4W3 -- On-Line -- Up 36 days 02:53:27
-- 8083776 Kbytes
Mutexes with waiters:
mid addr name holder lkcnt waiter waittime
31933 c00000019f945c18 pt_900023 9711376 0 10067034 0
9711419 0
10067033 0
9711428 0
9711380 0
9711424 0
10067035 0
9711432 0
9711408 0
9711410 0
9863376 0
9711395 0
9711409 0
10042145 0
9711413 0
9711399 0
9781199 0
9711369 0
9711406 0
9711425 0
9711378 0
9711426 0
9711372 0
9711368 0
9711384 0
9711375 0
9711377 0
9814662 0
9711381 0
10066709 0
9711407 0
9711412 0
9711462 0
9711429 0
pt_900023 is the systables of one database
TBLspace Report for cbs:informix.systables
Physical Address 1700026
Creation date 04/12/2003 03:24:31
TBLspace Flags 902 Row Locking
TBLspace contains VARCHARS
TBLspace use 4 bit bit-maps
Maximum row size 469
Number of special columns 3
Number of keys 2
Number of extents 15
Current serial value 50881
First extent size 8
Next extent size 8
Number of pages allocated 120
Number of pages used 115
Number of data pages 62
Number of rows 997
Partition partnum 9437219
Partition lockid 9437219
Extents
Logical Page Physical Page Size
0 17d7788 8
8 17d8c70 8
16 17d9ea0 8
24 17dc4f8 8
32 17ea818 8
40 17f0a98 8
48 186323b 8
56 1865adc 8
64 17d6661 8
72 170bdb4 8
80 170efb4 8
88 1702528 8
96 170e1f3 8
104 17085c5 8
112 1708cfd 8
If you encountered the problem before and can give me some help, thanks!