CARP setup : CPU load

  • Hi,

    We setup CARP+VIPs one month ago and it is working great. The only concern I have is that CPU load is 30% steady. Before implementing, it was only 2-3%. Could it be considerated as normal? Anyone could confirm?

    Thanks in advance.

  • You did not setup carp sync on both hosts did you?  Only firewall 1 should be syncing to firewall 2.  #2 should NOT sync back to #1.

  • Sync is enabled (checked) on both master and slave. Isn't correct? I've followed the tutorial…

  • I forgot to say that only the CARP master is having the 30% CPU issue… The slave is still running at 3%.

  • Take a screenshot of each of the hosts CARP setting screens.

  • Thanks for your quick reply. It is really appreciated!

    Screenshot of the CARP master :

    Screenshot of the CARP slave :

  • I can confirm with :

    # tcpdump -i fwe0

    that only CARP master is syncing to slave (not reverse)

  • Your setup looks correct.  Not sure why you would be seeing this added load.  My CARP hosts are not showing this.

    Can you try running a few commands:

    top -I
    vmstat -i
    vmstat -s

  • # top -I
    last pid: 28232;  load averages:  0.66,  0.34,  0.27  up 53+12:19:21  09:15:58
    42 processes:  1 running, 40 sleeping, 1 zombie
    CPU states:  0.4% user,  0.0% nice,  0.0% system,  2.3% interrupt, 97.3% idle
    Mem: 77M Active, 17M Inact, 51M Wired, 38M Buf, 286M Free
    Swap: 1024M Total, 1024M Free

    5756 root        1  8  20  6588K  6096K wait  579:03  1.86% sh

    # vmstat -i
    interrupt                          total      rate
    irq0: clk                      330038995        71
    irq1: atkbd0                          71          0
    irq3: sio1                            1          0
    irq4: sio0                            1          0
    irq8: rtc                      591744555        127
    irq10: fwohci0 re*            2122065738        458
    irq11: re0 atapci0            134476563        29
    irq14: ata0                      6560429          1
    Total                        3184886353        688

    # vmstat -s
    3996661449 cpu context switches
    3086919878 device interrupts
    159397933 software interrupts
    2324797307 traps
    1958236698 system calls
          39 kernel threads created
    10128798  fork() calls
      484718 vfork() calls
            0 rfork() calls
            0 swap pager pageins
            0 swap pager pages paged in
            0 swap pager pageouts
            0 swap pager pages paged out
          469 vnode pager pageins
        3167 vnode pager pages paged in
      2003234 vnode pager pageouts
      2034058 vnode pager pages paged out
            0 page daemon wakeups
            0 pages examined by the page daemon
          265 pages reactivated
    1701373918 copy-on-write faults
      885999 copy-on-write optimized faults
    415813077 zero fill pages zeroed
    414164367 zero fill pages prezeroed
        2874 intransit blocking page faults
    2329291089 total VM faults taken
            0 pages affected by kernel thread creation
    2926195887 pages affected by  fork()
    255613759 pages affected by vfork()
            0 pages affected by rfork()
    2179053141 pages freed
            0 pages freed by daemon
    1936954364 pages freed by exiting processes
        19535 pages active
        5476 pages inactive
            0 pages in VM cache
        12961 pages wired down
        72043 pages free
        4096 bytes per page
    774325484 total name lookups
              cache hits (89% pos + 9% neg) system 0% per-directory
              deletions 0%, falsehits 0%, toolong 0%

  • It may be the firewire interface. My experience is that firewire is usable for a sync interface, but it hasn't been 100% stable…

  • @dotdash:

    It may be the firewire interface. My experience is that firewire is usable for a sync interface, but it hasn't been 100% stable…

    Yep, that appears to be the culprit indeed.

    If you are not using firewire, try turning it off in the BIOS.

  • Unfortunately I'm using the 1394 int. for the CARP sync. I thought it was stable…

Log in to reply