memory leak?

vegas35 · 11-03-2007, 07:46 AM

Greetings,

I have a question regarding a possible memory leak on SuSE 9. My SuSE 9 server's used memory steadily increases from around 87MB right after reboot to around 580MB within a day or two. It stays at 580 consistently thereafter. When I do a "top", sorted by %MEM, I don't see anything hogging the memory. Are there other tools available that will show me what's using all the memory?

Here's some info from the server:

Code:

[root@srv03: ~ ]
$ uname -a
Linux srv03 2.4.21-243-athlon #1 Thu Aug 12 15:24:15 UTC 2004 i686 athlon i386 GNU/Linux

****************************************
TOP OUTPUT AFTER SERVER UP 2 DAYS...
****************************************

[root@srv03: ~ ]
$ top -b -n 1|more
top - 06:42:47 up 14:18,  1 user,  load average: 0.00, 0.00, 0.00
Tasks:  46 total,   1 running,  44 sleeping,   0 stopped,   1 zombie
Cpu(s):   0.0% user,   0.1% system,   0.0% nice,  99.9% idle
Mem:    741408k total,   579128k used,   162280k free,   263644k buffers
Swap:  1050800k total,        0k used,  1050800k free,    97076k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
11688 root      15   0  2872 2872 2568 S  0.0  0.4   0:00.12 smbd
 1727 root      15   0  2336 2336 2164 S  0.0  0.3   0:00.03 smbd
14139 root      15   0  2064 2064 1840 S  0.0  0.3   0:00.25 sshd
 1383 root      15   0  1780 1780 1228 S  0.0  0.2   0:00.10 nmbd
14140 mb        15   0  1712 1712 1292 S  0.0  0.2   0:00.10 bash
 1390 root      15   0  1644 1644 1512 S  0.0  0.2   0:00.24 sshd
14944 root      16   0  1644 1644 1224 S  0.0  0.2   0:00.04 bash
  890 root      15   0  1412 1412  468 S  0.0  0.2   0:00.08 klogd
 1704 postfix   15   0  1404 1404 1140 S  0.0  0.2   0:00.10 qmgr
 1700 root      15   0  1352 1352 1100 S  0.0  0.2   0:00.15 master
14268 postfix   15   0  1352 1352 1100 S  0.0  0.2   0:00.02 pickup
14943 root      16   0  1148 1148  840 S  0.0  0.2   0:00.02 su
 1602 root      15   0   868  868  732 S  0.0  0.1   0:00.02 xinetd
14996 root      16   0   840  840  676 R  2.0  0.1   0:00.02 top

***************************************
TOP OUTPUT SHORTLY AFTER A REBOOT
***************************************

$ top -b -n 1
top - 09:22:56 up 6 min,  1 user,  load average: 0.03, 0.06, 0.03
Tasks:  42 total,   1 running,  40 sleeping,   0 stopped,   1 zombie
Cpu(s):   1.7% user,   2.3% system,   0.0% nice,  96.0% idle
Mem:    741408k total,    65936k used,   675472k free,    15152k buffers
Swap:  1050800k total,        0k used,  1050800k free,    18608k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
 1726 root      16   0  2336 2336 1696 S  0.0  0.3   0:00.01 smbd
 1967 root      15   0  2064 2064 1840 S  0.0  0.3   0:00.03 sshd
 1383 root      15   0  1776 1776 1224 S  0.0  0.2   0:00.02 nmbd
 1968 mb        15   0  1688 1688 1276 S  0.0  0.2   0:00.04 bash
 1390 root      15   0  1644 1644 1512 S  0.0  0.2   0:00.12 sshd
  890 root      15   0  1412 1412  468 S  0.0  0.2   0:00.08 klogd
 1700 root      15   0  1352 1352 1100 S  0.0  0.2   0:00.02 master
 1705 postfix   15   0  1352 1352 1104 S  0.0  0.2   0:00.02 qmgr
 1704 postfix   15   0  1324 1324 1088 S  0.0  0.2   0:00.02 pickup
 1602 root      15   0   868  868  732 S  0.0  0.1   0:00.02 xinetd
 1989 mb        16   0   836  836  676 R  0.0  0.1   0:00.02 top

Thanks for any help!

Pearlseattle · 11-03-2007, 03:12 PM

You have 263MB buffers & 97MB cache - probably by some program reading or writing something - which will be freed as soon as any application needs them. Linux tries to cach e as much as possible, until that ram is needed. That amount should not increase after a few days.

gilead · 11-03-2007, 03:29 PM

Since you say that the usage peaks and then stabilises it doesn't sound like you're having a problem - it sounds normal. There's an article at http://gentoo-wiki.com/FAQ_Linux_Memory_Management which gives a useful overview of memory management under Linux.