OMSA memory hogging - pe6850 - RHEL4 - dsm_sa_datamgr32 process growth
Dennis_Muldoon at Dell.com
Dennis_Muldoon at Dell.com
Mon Mar 12 06:07:12 CST 2007
Hello,
Does anybody have any insight into the following issue?
OMSA51 is gobbling up memory even though restarted twice daily, leaving
little space for a database app.
The memory decrease does not happen if OMSA is not started in the first
place.
The daily memory decrease exceeds the memory used by OMSA - leading one
to suspect the kernel simply doesn't reclaim any pages used by OMSA.
Best regards
Dennis Muldoon
======================
Technical details:
======================
6850 installed from scratch with Redhat U4 Iso.
dsm_sa_datamgr32 use on the 6850:
4414 root 17 0 346m 130m 3680 S 0 1.6 1:53.49
dsm_sa_datamgr3
=======================================================
meminfo and ipcs on the server with the problem.
=======================================================
[root at SYSTEM2 ~]# cat /proc/meminfo
MemTotal: 8158444 kB
MemFree: 2165984 kB
Buffers: 314168 kB
Cached: 3066724 kB
SwapCached: 0 kB
Active: 3415620 kB
Inactive: 139760 kB
HighTotal: 0 kB
HighFree: 0 kB
LowTotal: 8158444 kB
LowFree: 2165984 kB
SwapTotal: 6265340 kB
SwapFree: 6265340 kB
Dirty: 180 kB
Writeback: 0 kB
Mapped: 198544 kB
Slab: 315040 kB
CommitLimit: 10344560 kB
Committed_AS: 883388 kB
PageTables: 2212 kB
VmallocTotal: 536870911 kB
VmallocUsed: 265568 kB
VmallocChunk: 536605187 kB
HugePages_Total: 0
HugePages_Free: 0
Hugepagesize: 2048 kB
[root at SYSTEM2 ~]# ipcs -m
------ Shared Memory Segments --------
key shmid owner perms bytes nattch status
0x0001ffb8 983040 root 666 76 5
0x00025990 1015809 root 666 8308 5
0x00027cb9 1048578 root 666 132256 1
0x00027cba 1081347 root 666 132256 1
0x00027cbb 1114116 root 666 132256 1
=======================================================
meminfo and ipcs on a server without problem:
=======================================================
[root at SYSTEM1 ~]# cat /proc/meminfo
MemTotal: 8158500 kB
MemFree: 7405960 kB
Buffers: 118056 kB
Cached: 478508 kB
SwapCached: 0 kB
Active: 239580 kB
Inactive: 402052 kB
HighTotal: 0 kB
HighFree: 0 kB
LowTotal: 8158500 kB
LowFree: 7405960 kB
SwapTotal: 6265340 kB
SwapFree: 6265340 kB
Dirty: 192 kB
Writeback: 0 kB
Mapped: 68672 kB
Slab: 84928 kB
CommitLimit: 10344588 kB
Committed_AS: 405808 kB
PageTables: 1748 kB
VmallocTotal: 536870911 kB
VmallocUsed: 265536 kB
VmallocChunk: 536605187 kB
HugePages_Total: 0
HugePages_Free: 0
Hugepagesize: 2048 kB
[root at SYSTEM1 ~]# ipcs -m
------ Shared Memory Segments --------
key shmid owner perms bytes nattch status
0x0001ffb8 0 root 666 76 4
0x00025990 32769 root 666 8308 4
0x00027cb9 65538 root 666 132256 1
0x00027cba 98307 root 666 132256 1
0x00027cbb 131076 root 666 132256 1
=================================
Memory information
=================================
system has a total of 8Go of physical memory.
free
total used free shared buffers
cached
Mem: 8158444 3511136 4647308 0 282308
841868
-/+ buffers/cache: 2386960 5771484
Swap: 6265340 0 6265340
========================================================================
===
On System1, a similarly loaded 6800 that it's not experiencing the
memory leak issue the free value on the second line is more or less
equal to 8Go even while filling cache memory by running a script to read
eveything on disk.
========================================================================
===
free
total used free shared buffers
cached
Mem: 8162092 3163148 4998944 0 288428
2501136
-/+ buffers/cache: 373584 7788508
Swap: 6265340 176 6265164
=========== End ============================================
More information about the Linux-PowerEdge
mailing list