Hi Admins,
The OS Version is 4.0D with patch kit 2. I am getting the following after the sys-check output.
I would like to know which parameter to set in the vm subsystem right to reduce
the page-in activitiy.
Are there any good web documentation on this stuff ? I had been browsing a lot and could not find
something useful.
TIA,
Natesa.
OUTPUT OF SYS_CHECK
Operational: High free memory ( 1474200 KB ).
This may be caused by the system being idle or system parameters
incorrectly set.
Operational: There is excessive page-in activity on the system ( 4558 ).
Investigate this condition.
See
vmstat(1)
reference page for information.
Note: The top and syd system monitoring tools, which are available in the public domain or on the DIGITAL UNIX Freeware CD-ROM, may consume enough kernel memory to crash the system if they are being run continuously. Do not run these tools continuously, unless you apply patches to these tools that fix the problem.
Total memory = 2097152 KB ( 2048 MB ) ( 2058 MB )
Kernel size, console and gh-chunks = 47800 KB ( 46 MB )
gh-chunks = 0 ( 0 KB ) ( 0 MB )
Total vm managed memory after boot = 2049352 KB ( 2001 MB )
current ubc size = 278712 KB ( 272 MB ) ( 13 % )
current wired memory = 161416 KB ( 157 MB ) ( 7 % )
current vm memory = 1620288 KB ( 1582 MB ) ( 79 % )
vm-page-free-target = 512 ( 4096 KB )
vm-page-free-optimal = 266 ( 2128 KB )
vm-page-free-min = 20 ( 160 KB )
vm-page-free-reserved = 10 ( 80 KB )
ubc-maxpercent = 100 % ( 2049352 KB ) ( 2001 MB )
ubc-minpercent = 10 % ( 204935 KB ) ( 200 MB )
ubc-borrowpercent = 20 % ( 409870 KB ) ( 400 MB )
vm-ubcdirtypercent = 10 % ( 27871 KB ) ( 27 MB )
vm-ubcseqstartpercent = 30 % ( 614805 KB ) ( 600 MB )
vm-ubcseqpercent = 10 % of current UBC size, when UBC size
is greater than vm-ubcseqstartpercent ( 61480 KB ) ( 60 MB )
The namei cache hit rate is 98 %.
The UBC cache hit rate is 28 %.
Total Physical Memory = 2048.00 M = 262144 pages Physical Memory Clusters: start_pfn end_pfn type size_pages / size_bytes 0 256 pal 256 / 2.00M 256 262139 os 261883 / 2045.96M 262139 262144 pal 5 / 40.00k Physical Memory Use: start_pfn end_pfn type size_pages / size_bytes 256 280 unixtable 24 / 192.00k 280 284 scavenge 4 / 32.00k 284 814 text 530 / 4.14M 814 918 data 104 / 832.00k 918 1134 bss 216 / 1.69M 1134 1287 kdebug 153 / 1.20M 1287 1294 cfgmgmt 7 / 56.00k 1294 1295 locks 1 / 8.00k 1295 1825 unixtable 530 / 4.14M 1825 1838 pmap 13 / 104.00k 1838 5970 vmtables 4132 / 32.28M 5970 262139 managed 256169 / 2001.32M ============================ Total Physical Memory Use: 261883 / 2045.96M Managed Pages Break Down: free pages = 185042 active pages = 7650 inactive pages = 9498 wired pages = 21451 ubc pages = 34294 ================== Total = 257935 WIRED Pages Break Down: vm wired pages = 2381 ubc wired pages = 5757 meta data pages = 2618 malloc pages = 7003 contig pages = 584 user ptepages = 3038 kernel ptepages = 58 free ptepages = 8 ================== Total = 21447
Virtual Memory Statistics: (pagesize = 8192) 7590 active pages 9498 inactive pages 185096 free pages 21459 wired pages 2053607840 virtual memory page faults 652246754 copy-on-write page faults 547530215 zero fill page faults 16352 reattaches from reclaim list 649476446 pages paged in 751 pages paged out 4497686305 task and thread context switches 297946918 device interrupts 24631656187 system calls
34141060 forks 250 vforks
Virtual Memory Statistics: (pagesize = 8192) procs memory pages intr cpu r w u act free wire fault cow zero react pin pout in sy cs us sy id 3170 31 51K 185K 21K 2053M 652M 547M 16K 649M 751 88 7K 1K 5 9 86 3172 31 52K 184K 21K 2815 408 1499 0 635 0 288 17K 2K 7 17 76 3166 30 51K 184K 21K 1140 231 319 0 349 0 177 3K 2K 3 6 90 4166 30 51K 185K 21K 1188 222 253 0 162 0 263 31K 3K 10 24 66 3166 30 51K 185K 21K 3225 774 697 0 725 0 238 26K 4K 12 24 64
vmstat -f
34141116 forks 250 vforksReceived on Wed Jun 20 2001 - 20:23:35 NZST
This archive was generated by hypermail 2.4.0 : Wed Nov 08 2023 - 11:53:42 NZDT