Re: System hang due to high iowait

From: Carlos Antonio Gomez (cgomez_at_ceis.cujae.edu.cu)
Date: 10/04/04

  • Next message: Mike Burger: "Re: USB Flash drive ??"
    To: "General Red Hat Linux discussion list" <redhat-list@redhat.com>
    Date: Mon, 4 Oct 2004 08:11:34 -0700
    
    

    I do not speak English :-)

    RedHAt AS3 Update 1

    The problem is that the server hangs. The daemons start to die and the
    server only functions at the kernel level. You can remove the network
    cable and it notifies you on the console. You insert the network cable
    and it also notifies you. If you browse to Webmin, it responds with the
    authorization page, but authentication fails because the daemon died.
    You can telnet to the server on different ports and it connects, but it
    waits for the login banner, etc.

    I made a script which logs the load on the server each minute and when I
    look at this I realize mailman is the problem. The load average goes up
    to 60% (?) when normally it's only 4%. I browsed the mailman logs and
    for the last few hangs, log entries coincide with mail sent to lists,
    but it happens with small or large lists.

    The hanging of the server is happening whether mail is sent to the small
    or large lists. I've tried troubleshooting the hardware and through
    testing I'm able to send mail to the lists and the server hangs, but not
    every time. This makes me thinkg it's not postfix or mailman, but
    possibly some other daemon.

    ----- Original Message -----
    From: <csu4@fedex.com>
    To: <redhat-list@redhat.com>
    Sent: Monday, October 04, 2004 1:46 AM
    Subject: System hang due to high iowait

    > Hi there,
    >
    > Could anyone help to find the system hung problem? The box
    > is running RHEL 3AS Update3. Whenever I do a massive data
    > transfer (such as rcp from another box to this box), after
    > certain time (around six to seven hours), this box would
    > hang up. I could see the top output even it's hung, and it
    > showed that the iowait is 99% and system contributed another
    > 1% making the server moving nowhere.
    >
    > The box connected to a HP MSA500 Smart Array Storage, thru
    > multipath (I enabled it following the user guide) connection.
    > The filesystem is ext3 format. Any problem here?
    >
    >
    >
    >
    >
    > Thx & Rgds,
    >
    >
    > SCHYU
    >
    >
    >
    >
    >
    >
    >
    > --
    > redhat-list mailing list
    > unsubscribe mailto:redhat-list-request@redhat.com?subject=unsubscribe
    > https://www.redhat.com/mailman/listinfo/redhat-list
    >
    >
    >

    __________________________________________

    Aniversario 40 de la CUJAE
     
    Visite:
     
    XII Convencion de Ingenieria y Arquitectura
    http://www.cujae.edu.cu/eventos/convencion/

    -- 
    redhat-list mailing list
    unsubscribe mailto:redhat-list-request@redhat.com?subject=unsubscribe
    https://www.redhat.com/mailman/listinfo/redhat-list
    

  • Next message: Mike Burger: "Re: USB Flash drive ??"

    Relevant Pages

    • help: howto to make the machine secure!!
      ... I'm about to loose my nerves because the data center where my server ... I commented out all rpc services as it is said to be insecure. ... # pass the given daemon an AF_INET6 socket. ... # RLOGIND - rlogin daemon (BSD protocols) ...
      (comp.unix.solaris)
    • RE: Three Errors on Server 2003 SB Any idea?
      ... occurring on your SBS 2K3 Server: ... expand the Recipients object -> Offline ... Address Lists. ... Ensure a valid Public Folder Store is set for the Offline Address ...
      (microsoft.public.windows.server.sbs)
    • Re: redhat-list Digest, Vol 2, Issue 1
      ... end-of-life of Red Hat Linux 9.0.... ... In this case, if this is an important server, then it may as well be ... The problem with mailing lists like this one is that the from ...
      (RedHat)
    • Re: DNS poisoning block list?
      ... I have checked lists, and as an individual user they are really so far ... site for DNS poisoning, I don't know why I shouldn't try to explicitly ... that I was served from a poisoning DNS server like this that I had taken ...
      (comp.os.linux.security)
    • Re: fedora-list Digest, Vol 46, Issue 104
      ... server can't probe the monitor (but this isn't the real ... F8 Network Woe ... bridge-wlan0: enabling the bridge ... Most other lists I follow have this feature and it help a lot when trying ...
      (Fedora)