RX300S7, SLES11 SP2 utopisch hohe CPU Zeiten

PRIMERGY, SPARC Enterprise Server, PRIMEFLEX, PRIMEPower, BS2000

Moderator: ModTeam

guru
Posts: 2
Joined: Tue Dec 28, 2010 8:44
Product(s): Amilo M1437G

RX300S7, SLES11 SP2 utopisch hohe CPU Zeiten

Postby guru » Sat Apr 05, 2014 17:29

Hallo,

ich habe nach dem re-boot des servers sowohl in 'ps' als in 'top' bei vielen Prozessen eine utopisch hohe Angabe der CPU Zeit:

    # ps ax
    PID TTY STAT TIME COMMAND
    1 ? Ss 14332454:39 init [3]
    2 ? S 20233020:37 [kthreadd]
    3 ? S 0:00 [ksoftirqd/0]
    4 ? S 41347602:07 [kworker/0:0]
    6 ? S 0:00 [migration/0]
    7 ? S 0:00 [watchdog/0]
    8 ? S 0:00 [migration/1]
    10 ? S 300258:44 [ksoftirqd/1]
    11 ? S 20533279:22 [kworker/0:1]
    12 ? S 300258:44 [watchdog/1]
    13 ? S 0:00 [migration/2]
    14 ? S 0:00 [kworker/2:0]
    15 ? S 0:00 [ksoftirqd/2]
    16 ? S 300258:44 [watchdog/2]

    sunrise:~ # cat /etc/SuSE-release
    SUSE Linux Enterprise Server 11 (x86_64)
    VERSION = 11
    PATCHLEVEL = 2


die falschen Angaben stammen aus den /proc/PID/stat files; ich habe das hier mal fuer zwei Prozesse, einen mit normaler CPU-Zeitangabe und einen mit utopischer nachgesehen:

ps lines say:


    7109 pts/1 S+ 0:19 top
    7129 ? S 36843721:01 /data/opt/lib/sisis/avserver/bin/AVServer -p 8076

    /proc/pid/stat say:

    # cat /proc/7109/stat
    7109 (top) S 5427 7109 5427 34817 7109 8192 428 0 0 0 825 1142 0 0 20 0 1 0 1644813 9256960
    320 28562001920 4194304 4289772 140737488348448
    140737488347304 140737346743459 0 0 0 138047495 0 0 0 17 9 0 0 0 0 0

    # cat /proc/7129/stat
    7129 (AVServer) S 3981 3962 3962 0 -1 8256 2717 3582 0 0 1814163308025 1842872854930 0
    1801552443 20 0 1 0 1663231 81698816 8743 28562001920
    134512640 134754471 4294953328 4294918444 4294960176 0 0 4103 272632320 0 0 0 17 26 0 0 0 0
    0

es ist immer der 14. Wert, also 825 bzw. 1814163308025

Hat jemand dazu eine Idee? Danke


    Matthias

    devzero
    Posts: 6
    Joined: Tue Oct 26, 2010 20:20
    Product(s): RX300

    Re: RX300S7, SLES11 SP2 utopisch hohe CPU Zeiten

    Postby devzero » Tue Mar 24, 2015 11:31

    dazu gabs mal kernel bugs. im sles changelog find ich gerad nichts passendes, aber schau mal unter

    https://access.redhat.com/documentation ... ernel.html

    BZ#959930
    Due to the way the CPU time was calculated, an integer multiplication overflow bug could occur after several days of running CPU bound processes that were using hundreds of kernel threads. As a consequence, the kernel stopped updating the CPU time and provided an incorrect CPU time instead. This could confuse users and lead to various application problems. This update applies a patch fixing this problem by decreasing the precision of calculations when the stime and rtime values become too large. Also, a bug allowing stime values to be sometimes erroneously calculated as utime values has been fixed.

    BZ#908794
    When counting CPU time, the utime and stime values are scaled based on rtime. Prior to this update, the utime value was multiplied with the rtime value, but the integer multiplication overflow could happen, and the resulting value could be then truncated to 64 bits. As a consequence, utime values visible in the user space were stall even if an application consumed a lot of CPU time. With this update, the multiplication is performed on stime instead of utime. This significantly reduces the chances of an overflow on most workloads because the stime value, unlike the utime value, cannot grow fast.

    oder:
    http://bugs.centos.org/view.php?id=6421

    daher:
    einfach mal nen update machen oder es erstmal mit cold-boot versuchen


    Return to “Server Produkte”

    Who is online

    Users browsing this forum: No registered users and 1 guest