1. Community Tip: Please Give Thanks to Those Sharing Their Knowledge.
    If you receive helpful answer on this forum, please show thanks to the poster by clicking "LIKE" link for the answer that you found helpful.
    Click the link above to access ADSM.ORG Acceptable Use Policy and forum rules which should be observed when using this website. Violators may be banned from this website. This notice will disappear after you have made at least 3 posts.

TS3500 data path failover

Discussion in 'Tape / Media Library' started by adsmsuser, Nov 1, 2012.

  1. adsmsuser

    adsmsuser ADSM.ORG Member

    Jan 11, 2010
    Likes Received:
    teamleader tsm

    for now we have all our drives split over 2 fabrics.
    all even drives over fabric1
    all uneven dirves over fabric2

    that was fine as long as we had no library at a second location ;)

    now we depend on a WAN connection, administered by a local provider. (2 dark-fibre)
    some days ago we where told fibre1 would have a "downtime" ;)

    sadly the provider swapped the to cables ^^.

    so we took 50% of our drive manually offline. the other 50% _went_ offline.


    - i guess this KEY you have to enter is NOT free? Or do i really just drive to the 2nd location and check if its already in the L-frame as the documentation suggests?
    - is anyone using datapathfailover?
    - is anyone using datapathfailover on HP/UX?

    thanks! :)

    ps: http://www-01.ibm.com/support/docview.wss?uid=ssg1S7001206&aid=1

    documentation wit LTO example says:
    /home/root# ioscan -funC tape
    Class I H/W Path Driver S/W State H/W Type Description
    tape 0 0/1/0/ atdd CLAIMED DEVICE IBM ULT3580-TD3 /dev/rmt/0m
    tape 1 0/1/0/ atdd CLAIMED DEVICE IBM ULT3580-TD3 /dev/rmt/1m.altpath

    my output with 3592 is a litte different ;)
    /dev/rmt# ioscan -funC tape
    Class I H/W Path Driver S/W State H/W Type Description
    tape 104 0/2/1/ atdd_leg CLAIMED DEVICE IBM 03592E06
    /dev/rmt/104m /dev/rmt/104mn /dev/rmt/c172t0d0BEST /dev/rmt/c172t0d0BESTn
    /dev/rmt/104mb /dev/rmt/104mnb /dev/rmt/c172t0d0BESTb /dev/rmt/c172t0d0BESTnb
    tape 85 0/2/1/ atdd_leg CLAIMED DEVICE IBM 03592E06
    /dev/rmt/85m /dev/rmt/85mn /dev/rmt/c148t0d0BEST /dev/rmt/c148t0d0BESTn
    /dev/rmt/85mb /dev/rmt/85mnb /dev/rmt/c148t0d0BESTb /dev/rmt/c148t0d0BESTnb

    with symlinks to
    lrwxr-xr-x 1 root sys 24 Aug 27 16:57 j1a_06_b04 -> /dev/rtape/tape43_BESTnb
    Last edited: Nov 1, 2012
  3. moon-buddy

    moon-buddy ADSM.ORG Moderator

    Aug 24, 2005
    Likes Received:
    Electronics Engineer, Security Professional
    Somewhere in the US
    I am little confused.

    What exactly do you want to achieve? Data path failover or Device failover?

    You said data path failover on your post title but you go on describing what seems to be a device failover. These are very different scenarios.

    Data path failover generally means that you have two paths to a device, i.e., tape drive. If one path goes, the other takes over.

    Device failover literally means that you have a redundant set of devices that you maintain should the primary set of devices fail. The devices can be local or remote. If these are local, the devices should be registered and configured on the local system. What you do is simply to bring offline the bad device and bring online the good one. It may be necessary to reassign the good one to the right storage pool.

    If the device is remote, the configuration will be a little tricky especially if this involves tape drives and a library. For one, the tapes has be at the remote site, and the TSM DB has to be the right one IF you are using a remote site server.
    Last edited: Nov 2, 2012

Share This Page