• 
    $ top
    last pid: 48536;  load averages:  0.02,  0.03,  0.08  up 2+16:18:30    12:16:31
    96 processes:  1 running, 90 sleeping, 5 zombie
    
    Mem: 78M Active, 1503M Inact, 160M Wired, 112M Buf, 260M Free
    Swap: 2048M Total, 2048M Free
    
      PID USERNAME  THR PRI NICE   SIZE    RES STATE    TIME   WCPU COMMAND
      511 root        1  -8    0 24388K 23164K piperd   0:28  0.78% php
     1047 proxy      19  20    0 36012K 32960K kserel   3:00  0.00% squid
      672 root        1  96    0  6388K  6144K select   0:57  0.00% bsnmpd
      729 root        1   8   20  1932K  1372K wait     0:38  0.00% sh
     1158 zabbix      1   8    5  2592K  1292K nanslp   0:14  0.00% zabbix_agentd
      491 root        1   4    0  3724K  2792K kqread   0:08  0.00% lighttpd
      200 root        1  96    0  1468K  1104K select   0:05  0.00% syslogd
      313 root        1 -58    0  4240K  2532K bpf      0:04  0.00% tcpdump
    24045 root        1 -58    0  4100K  2052K bpf      0:04  0.00% bandwidthd
    24043 root        1 -58    0  4224K  2176K bpf      0:04  0.00% bandwidthd
     1128 root        1   8    0  1720K  1160K wait     0:04  0.00% sh
    24042 root        1 -58    0  4224K  2176K bpf      0:04  0.00% bandwidthd
    24044 root        1 -58    0  4224K  2176K bpf      0:04  0.00% bandwidthd
      811 root        1  96    0  1472K  1092K select   0:03  0.00% routed
    64010 root        1   8   20  1272K   720K nanslp   0:02  0.00% check_reload_status
      314 root        1  -8    0  1276K   728K piperd   0:01  0.00% logger
     1100 proxy       1  96    0  1604K  1120K select   0:01  0.00% pinger
      554 root        1  96    0  1280K   716K select   0:01  0.00% choparp
    
    

    Ntop seems to be causing a memory leak of some sort. I have 2gb of ram.. and I stopped ntop and still it shows 260mb of ram left.. I can't seem to find where the used memory is being sent to? Maybe the zombie proc?

    Rebooting the firewlal solves the problem, but obviously I would love to use ntop. To be honest I'm not sure it's the ntop proc, but when viewing the rdd graphs there it seems to happen (I think?)

    How do I get the missing memory back without rebooting? I don't want to reboot the firewall during working hours…


  • another screenshot from phpsys. All the memory seems to be located in the inactive part.. and It seems that it's not taking memory out of there like it's suppose to??


  • A bit off topic
    It's my impression that the NTOP package is out of date and shouldn't be used before a dedicated maintainer shows up and fix it.


  • That's true, but I need to figure out if it's bad or not when you have so much inactive memory and if it's being caused by ntop.

    I would expect it to be transfered to the free –> physical memory.


  • Inactive memory is not 'bad'.  When a program ends it's memory is placed into inactive.  That is memory that is free to other programs (if they need it) but is ready to go for the ended program should you restart it.


  • The system now shows only 25mb free and about 2mb of swap being used.. while there is still almost 1600mb in the inactive..

    Could it be that it's not freeing up or able to use this memory?