[HPADM] RE: SCSI: Read error

From: Sabrina Lautier (slautier_at_amadeus.net)
Date: 10/23/03

  • Next message: Sabrina Lautier: "[HPADM] RE: RE: SCSI: Read error"
    To: hpux-admin@DutchWorks.nl
    Date: Thu, 23 Oct 2003 17:32:30 +0200
    
    

    First of all, thanks a lot all for your quick answers.

    The concerned disk is the 2nd internal disk c2t2d0
    According to the following commands it seems ok:
    $ ioscan -kfnC disk
    PV Name /dev/dsk/c2t2d0
    VG Name /dev/vg00
    PV Status available
    Allocatable yes
    VGDA 2
    Cur LV 7
    PE Size (Mbytes) 4
    Total PE 4340
    Free PE 8
    Allocated PE 4332
    Stale PE 0
    IO Timeout (Seconds) default
    Autoswitch On

    $ ioscan -kfnC disk
    Class I H/W Path Driver S/W State H/W Type
    Description
    ==================================================================================
    disk 0 0/0/1/1.2.0 sdisk CLAIMED DEVICE
    SEAGATE ST318203LC
                                       /dev/dsk/c1t2d0 /dev/rdsk/c1t2d0
    disk 1 0/0/2/0.2.0 sdisk CLAIMED DEVICE
    SEAGATE ST318404LC
                                       /dev/dsk/c2t2d0 /dev/rdsk/c2t2d0
    ...

    PS: the 2 internal disks are mirrored

    When I try to query the drive with the support tools manager (xstm), I got
    an successful answer (although the disk status is not specified).

    Any idea of other (deeper) tests I can perform on it ?
    With SunOs I used to run the fsck command which test and repair the bad
    blocs.
    Is there an equivalent command on HP-UX ?

    Regards,

    Sabrina Lautier
    Amadeus SAS
    DEV-IIS-OAU-SYS
    +33 (0)4 97 23 09 56
    slautier@amadeus.net

    From: "Talavera, Jesus" <TalavJ@europe.stortek.com> on 23/10/2003 15:25

                                                                              
     To: "'Sabrina Lautier'"
            <slautier@amadeus.net>
                                                                              
                                                                              
     cc:
                                                                              
                                                                              
                                                                              
                                                                              
                                                                              
                                                                              
     Subjec RE: [HPADM] SCSI: Read error
     t:
                                                                              
                                                                              
                                                                              

    Hi Sabrina,

            You can locate the physical disk in which the failure is reported
    by doing the following:

            #cd /dev/dsk
            #ls -ltr | grep "0x022000"

            The number 0x022000 is the minor number the OS assings to the
    device file /dev/dsk/cXtXdX and it can be used to see which device has the
    error.

            Once you know which disk is, you can run ioscan -kfnC disk to see
    is the disk is on an state diferent than CLAIMED, you can see a descripcion
    off this statues in the ioscan man page. But remember the -k switch in the
    ioscan reads the current kernel tables, that means, the information could
    not be updated.

            Also you can do a "pvdisplay /dev/dk/cXtXdX", and check if there is
    any PE in a status different the current, only if the PE status is current
    the everyting is rigth otherwise change the disk as soon as possible. Other
    PE statues could be stale ( = status ) and ???? ( = very bad status) .

    I hope this helps. Regards.

    -----Mensaje original-----
    De: Sabrina Lautier [mailto:slautier@amadeus.net]
    Enviado el: jueves, 23 de octubre de 2003 15:34
    Para: hpux-admin@DutchWorks.nl
    Asunto: [HPADM] SCSI: Read error

    All,

    Since the 15th of September, we're getting the same error message in
    /var/adm/syslog/syslog.log file:
    <date> <server> vmunix: SCSI: Read error -- dev: b 31 0x022000, errno: 126,

    resid: 2048,
    <date> <server> vmunix: blkno: 8, sectno: 16, offset: 8192, bcount:

    2048.

    This error is reported more than every seconds.
    I'm trying to find where the Pb is, but I'm quite new to HP-UX (I know
    SunOS pretty well, but HP-UX not at all...).

    This HP server has 2 internal disks and a EMC disk array attached to it.
    This disk array is monitored by EMC and they told me that this Pb is not
    related to it.

    The Pb should then come from the internal disks (SEAGATE T318203LC).
    PS: how to check if these disks are SCSI or IDE one ?

    Does anybody have ever experienced a such Pb ?

    Thanks a lot for any help.

    Rgds

    Sabrina Lautier
    Amadeus SAS
    DEV-IIS-OAU-SYS
    +33 (0)4 97 23 09 56
    slautier@amadeus.net

    --
                 ---> Please post QUESTIONS and SUMMARIES only!! <---
            To subscribe/unsubscribe to this list, contact
    majordomo@dutchworks.nl
           Name: hpux-admin@dutchworks.nl     Owner:
    owner-hpux-admin@dutchworks.nl
     Archives:  ftp.dutchworks.nl:/pub/digests/hpux-admin       (FTP, browse
    only)
                http://www.dutchworks.nl/htbin/hpsysadmin   (Web, browse &
    search)
    --
                 ---> Please post QUESTIONS and SUMMARIES only!! <---
            To subscribe/unsubscribe to this list, contact majordomo@dutchworks.nl
           Name: hpux-admin@dutchworks.nl     Owner: owner-hpux-admin@dutchworks.nl
     
     Archives:  ftp.dutchworks.nl:/pub/digests/hpux-admin       (FTP, browse only)
                http://www.dutchworks.nl/htbin/hpsysadmin   (Web, browse & search)
    

  • Next message: Sabrina Lautier: "[HPADM] RE: RE: SCSI: Read error"

    Relevant Pages

    • Migrating Veritas logical volumes config to another internal disk
      ... We have a E45K with two internal disks. ... internal disks are mirrored with veritas but the file ... We have three disk arrays connected in daisy chain to ...
      (SunManagers)
    • [HPADM] SUMMARY: SCSI: Read error
      ... 2/ Address 0x022000, indicates disk c2t2d0 ... stopped responding to SCSI commands ... Before concluding the disk is nearly dead, I'm going to apply LVM and SCSI ... This HP server has 2 internal disks and a EMC disk array attached to it. ...
      (HP-UX-Admin)
    • Re: ZFS hot swapping and OS hangs
      ... A disk in A5200 or in the U60? ... failure in the A5200 would hang the OS - especially because the system ... If it didn't come up easily, then one of the two internal disks ... Were you running SVM ...
      (comp.unix.solaris)
    • Replacing disks in DiskSuite
      ... We have a Sun 480R with two 36GB internal disks and we need to replace them ... Apart from swap there are only two partitions on the ... Take the mirror disk offline ... Remove and replace the mirror disk ...
      (SunManagers)
    • Re: New HDD Installation
      ... "John John" wrote: ... a single partition disk. ... confirm or change the active flag of the partition using the FDISK command. ... There are other ways to mark the partition active, either with a Windows ...
      (microsoft.public.windowsxp.general)