Memory utilization issue

solaris_1977 · February 11, 2013, 10:48am

I have parallels container running on Suse. From top command, I am not able to see, what is eating up so big amount of memory.

top - 07:44:24 up 172 days, 18:52,  1 user,  load average: 0.01, 0.02, 0.00
Tasks:  44 total,   1 running,  43 sleeping,   0 stopped,   0 zombie
Cpu(s):  0.0%us,  0.0%sy,  0.0%ni, 99.9%id,  0.0%wa,  0.0%hi,  0.0%si,  0.0%st
Mem:    515228M total,   514604M used,      623M free,     1884M buffers
Swap:    64764M total,        0M used,    64763M free,   437525M cached
  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
    1 root      15   0   800  292  244 S    0  0.0   0:00.81 init
 3429 root      15   0 14840 5096 3524 S    0  0.0   0:05.73 rs
 3856 root      18   0 14844 5104 3528 S    0  0.0   0:06.01 rs
 3916 root      15   0 14844 5100 3524 S    0  0.0   0:05.46 rs
 5214 root      15   0 42288 3008 2348 S    0  0.0   0:00.02 sshd
 5216 root      15   0 12876 2328 1464 S    0  0.0   0:00.07 bash
 5256 root      15   0  9616 1288  972 R    0  0.0   0:00.02 top
 7805 root      11  -4  7672  604  328 S    0  0.0   0:00.00 udevd
 9617 bb        18   0  8040  660  280 S    0  0.0   0:00.00 runbb.sh
 9620 bb        18   0  8040  656  276 S    0  0.0   0:00.00 runbb.sh
 9623 bb        18   0  1572  492  420 S    0  0.0   0:00.25 bbrun
 9626 bb        20   0  8040  648  268 S    0  0.0   0:00.00 runbb.sh
 9627 bb        18   0  1572  496  420 S    0  0.0   0:00.03 bbrun
 9669 redwd8t   18   0  9168 1644 1272 S    0  0.0   0:00.02 platform-agent
 9709 redwd8t   15   0  7052 1660 1232 S    0  0.0  18:49.52 network-process
11526 root      15   0 14844 5100 3524 S    0  0.0   0:05.53 rs
11576 bb        18   0  1572  496  420 S    0  0.0   0:00.02 bbrun
13802 redwd8t   15   0  5440  448  160 S    0  0.0   0:00.59 network-process
13995 messageb  18   0 13100 1012  740 S    0  0.0   0:00.00 dbus-daemon
15463 root      15   0 14844 5100 3524 S    0  0.0   0:05.66 rs
17805 root      16   0 11912 1740 1380 S    0  0.0   0:00.01 hald
18217 root      15   0  3772  672  536 S    0  0.0   0:43.20 syslogd
18231 nobody    17   0  4776  452  340 S    0  0.0   0:00.00 portmap
19651 nobody    18   0  6976  664  528 S    0  0.0   5:45.87 rwhod
19653 nobody    15   0  6976  472  336 S    0  0.0   0:00.09 rwhod
19796 root      16   0 10808  960  748 S    0  0.0   0:00.05 xinetd
19986 root      16   0 29056 1444  924 S    0  0.0   7:58.98 sshd
20157 root      15   0 14844 5100 3524 S    0  0.0   0:05.52 rs
20370 endeca    18   0  110g 4.5g  35m S    0  0.9   6802:07 dgraph
20456 root      17   0  5996  320  180 S    0  0.0   0:00.00 rpc.mountd
21550 postfix   15   0 24196 2260 1800 S    0  0.0   0:00.01 proxymap
21733 root      15   0 23156 2288 1808 S    0  0.0   0:59.25 master
21747 postfix   15   0 24244 2440 1928 S    0  0.0   0:00.08 qmgr
21779 root      15   0 52412 2936 1520 S    0  0.0   0:04.31 httpd2-prefork
21789 wwwrun    18   0 52544 2908 1300 S    0  0.0   0:00.00 httpd2-prefork
22017 wwwrun    18   0 52544 2900 1300 S    0  0.0   0:00.00 httpd2-prefork
22057 root      18   0 11252  776  604 S    0  0.0   0:12.16 cron
22148 endeca    18   0 13.4g 6.0g  12m S    0  1.2   7557:38 java
23574 root      18   0 14932 5380 3700 S    0  0.0   5:32.62 rs
23581 root      15   0 14852 5156 3572 S    0  0.0   0:01.96 rs
root@vtldoms06:~> vmstat 5 5
procs -----------memory---------- ---swap-- -----io---- -system-- -----cpu------
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st
 0  0    704 642780 1930544 448026928    0    0     6   176    0    0  1  0 98  0  0
 0  0    704 641492 1930564 448027136    0    0     0     2    0 2834  0  0 100  0  0
 0  0    704 644192 1930600 448027128    0    0     0     2    0 3023  0  0 100  0  0
 0  0    704 644440 1930620 448027324    0    0     0     1    0 2853  0  0 100  0  0
 0  0    704 643592 1930656 448027292    0    0     0     1    0 2866  0  0 100  0  0
root@vtldoms06:~>

Can somebody give me a pointer, where to check.

DGPickett · February 11, 2013, 2:00pm

You can use 'ps -le' and sort on the SZ column to see big VM procs. Sort is now built in. You may have to specify the columns (-o) to see rss which is ram used, and sort on that. Sorts are now built into newer ps versions.

$ ps -le|(line;sort -nr +9 -10)|head
  F S        UID   PID  PPID  C PRI NI             ADDR   SZ            WCHAN TTY       TIME COMD
  1 S      50453  6739     1  0 155 20         e7201400 241737           e843b4 ?        1336:16 TMUSREVT
  1 R          0  2940  2757  0 152 20         a4d0c800 144262                - ?        250150:38 mhragent
  1 S      22049  6852     1  0 168 20        160d80b00 43720        138032980 ?        259:21 dequeueClientX
  1 S      22049  6851     1  0 168 20        1558c1700 43720         e54ec200 ?        258:38 dequeueClientX
  1 S      22049  6860     1  1 168 20        14b001a00 43716        172cc3a80 ?        261:35 dequeueClientX
  1 S      22049  6859     1  0 168 20         dab08d00 43716        11b204a40 ?        259:56 dequeueClientX
  1 S      22049  6858     1  0 168 20         b5b47500 43716        14ba4f780 ?        262:21 dequeueClientX
  1 S      22049  6857     1  1 168 20         d9049c00 43716         ab7fd380 ?        261:19 dequeueClientX
  1 S      22049  6856     1  0 168 20         a46e6900 43716        12bbc3180 ?        259:59 dequeueClientX
$