从手册上看到的检查点触发执行的条件：
1 到达CKPTINTVL指定的时间并切有修改
2 物理日志占用超过75％
3 ...
另外一台没有承载业务的机子，配置都是一样的，如下
# Shared Memory Parameters
LOCKS 500000 # Maximum number of locks
NUMAIOVPS 3 # Number of IO vps
PHYSBUFF 64 # Physical log buffer size (Kbytes)
LOGBUFF 64 # Logical log buffer size (Kbytes)
CLEANERS 16 # Number of buffer cleaner processes
SHMBASE 0x44000000L # Shared memory base address
SHMVIRTSIZE 200000 # initial virtual shared memory segment size
SHMADD 81920 # Size of new shared memory segments (Kbytes)
EXTSHMADD 8192 # Size of new extension shared memory segments (Kbytes)
SHMTOTAL 0 # Total shared memory (Kbytes). 0=>unlimited
SHMVIRT_ALLOCSEG .80,4 # Values between 0 and .99 are %, values > 1 are
# KB - when this much virtual memory is used we
# try to get a new segment. 0 means "off". 2nd
# parameter is alarm level
CKPTINTVL 300 # Check point interval (in sec)
TXTIMEOUT 300 # Transaction timeout (in sec)
STACKSIZE 128 # Stack size (Kbytes)
执行onstat -g ckp结果如下，这个是到5分钟就坚决执行检查点的，但是不承载业务：
> onstat -g ckp
IBM Informix Dynamic Server Version 11.10.FC2 -- On-Line -- Up 134 days 23:38:57 -- 1375204 Kbytes
AUTO_CKPTS=On RTO_SERVER_RESTART=Off
Critical Sections Physical Log Logical Log
Clock Total Flush Block # CkptWaitLong# Dirty DskfluTotal Avg Total Avg
Interval Time Trigger LSN TimeTimeTimeWaitsTimeTimeTimeBuffers /Sec Pages /Sec Pages /Sec
66802 14:35:05CKPTINTVL247:0x4bc018 0.0 0.0 0.0 0 0.0 0.0 0.0 56 56 42 0 11 0
66803 14:40:05CKPTINTVL247:0x4bd23c 0.0 0.0 0.0 0 0.0 0.0 0.0 26 26 8 0 1 0
66804 14:45:06CKPTINTVL247:0x4c0018 0.0 0.0 0.0 0 0.0 0.0 0.0 32 32 19 0 3 0
66805 14:50:06CKPTINTVL247:0x4c103c 0.0 0.0 0.0 0 0.0 0.0 0.0 13 13 0 0 1 0
66806 14:55:06CKPTINTVL247:0x4c4018 0.0 0.0 0.0 0 0.0 0.0 0.0 32 32 19 0 3 0
66807 15:00:06CKPTINTVL247:0x4c503c 0.0 0.0 0.0 0 0.0 0.0 0.0 13 13 0 0 1 0
66808 15:05:06CKPTINTVL247:0x4c8018 0.0 0.0 0.0 0 0.0 0.0 0.0 32 32 19 0 3 0
66809 15:10:06CKPTINTVL247:0x4c903c 0.0 0.0 0.0 0 0.0 0.0 0.0 13 13 0 0 1 0
66810 15:15:06CKPTINTVL247:0x4cc018 0.0 0.0 0.0 0 0.0 0.0 0.0 32 32 19 0 3 0
66811 15:20:06CKPTINTVL247:0x4cd03c 0.0 0.0 0.0 0 0.0 0.0 0.0 13 13 0 0 1 0
66812 15:25:06CKPTINTVL247:0x4d0018 0.0 0.0 0.0 0 0.0 0.0 0.0 32 32 19 0 3 0
66813 15:30:07CKPTINTVL247:0x4d103c 0.0 0.0 0.0 0 0.0 0.0 0.0 13 13 0 0 1 0
66814 15:35:07CKPTINTVL247:0x4fa018 0.0 0.0 0.0 0 0.0 0.0 0.0 92 92 76 0 41 0
66815 15:40:07CKPTINTVL247:0x4fc308 0.0 0.0 0.0 0 0.0 0.0 0.0 61 61 50 0 2 0
66816 15:45:07CKPTINTVL247:0x4ff018 0.0 0.0 0.0 0 0.0 0.0 0.0 32 32 19 0 3 0
66817 15:50:07CKPTINTVL247:0x50003c 0.0 0.0 0.0 0 0.0 0.0 0.0 13 13 0 0 1 0
66818 15:55:07CKPTINTVL247:0x503018 0.0 0.0 0.0 0 0.0 0.0 0.0 32 32 19 0 3 0
66819 16:00:08CKPTINTVL247:0x50403c 0.0 0.0 0.0 0 0.0 0.0 0.0 13 13 0 0 1 0
66820 16:05:08CKPTINTVL247:0x507018 0.0 0.0 0.0 0 0.0 0.0 0.0 32 32 19 0 3 0
66821 16:10:08CKPTINTVL247:0x50803c 0.0 0.0 0.0 0 0.0 0.0 0.0 13 13 0 0 1 0
Max Plog Max Llog Max Dskflush Avg Dskflush Avg Dirty Blocked
pages/sec pages/sec Time pages/sec pages/sec Time
1408 2592 0 1 0 0
> onstat -R
IBM Informix Dynamic Server Version 11.10.FC2 -- On-Line -- Up 134 days 23:52:04 -- 1375204 Kbytes
Buffer pool page size: 2048
32 buffer LRU queue pairs priority levels
# f/m pair total % of length LOW HIGH
0 f 15625 100.0% 15625 15416 209
1 m 0.0% 0 0 0
2 f 15625 100.0% 15625 15406 219
3 m 0.0% 0 0 0
................................
60 f 15625 100.0% 15624 15429 195
61 m 0.0% 1 0 1
62 f 15625 100.0% 15625 15398 227
63 m 0.0% 0 0 0
31 dirty, 500000 queued, 500000 total, 524288 hash buckets, 2048 buffer size
start clean at 0.780% (of pair total) dirty, or 122 buffs dirty, stop at
0.390%
目前确实是不影响业务运行，但我真的很想搞明白到底是为什么，为什么没有承载业务的机子LRU dirty总是好几十，而业务量很大的LRU dirty在前5分钟几乎全是0，难道真的不往LRU里面写吗，呵呵 回复 11# houhuaw
检查下两个库是不是都使用的 buffered-log，再检查点logbuff的大小是不是一样。。 回复 12# liaosnet
都是buffered-log的库，LOGBUFF都是 64(Kbytes)
页:
1
[2]