BENT 8.5, NT4Domain, Win2K RAID Svr crash - Veritas Backup Exec

This is a discussion on BENT 8.5, NT4Domain, Win2K RAID Svr crash - Veritas Backup Exec ; Have the follwing; Network: + Complete Fiber switched using Catalyst 4006's - Only one segment - No Routers or RSM's installed - All Servers mentioned have 3COM Gigabit NIC's Backup Server: + BENT/2000 8.5 running on 2K Server SP1. PII-266 ...

+ Reply to Thread
Results 1 to 3 of 3

Thread: BENT 8.5, NT4Domain, Win2K RAID Svr crash

  1. BENT 8.5, NT4Domain, Win2K RAID Svr crash


    Have the follwing;

    Network:
    + Complete Fiber switched using Catalyst 4006's
    - Only one segment
    - No Routers or RSM's installed
    - All Servers mentioned have 3COM Gigabit NIC's

    Backup Server:
    + BENT/2000 8.5 running on 2K Server SP1. PII-266
    - HP SureStore DDS4 DAT 40x6.

    NT4 Servers:
    + Backups up NT4 PDC, BDC's and all Member Svs FINE all SP6a

    WIN2K Servers:
    The problems start when the job gets to the (2) Win2K Servers
    + Each Server is a Dual PIII800 gateway
    - each has a 170GB H/W RAID 5 array
    - System/Boot are Mirrored
    - drives are LVD UltraSCSI 160
    - each server has the BE 8.5 Agent running
    - NO ACTIVE DIRECTORY

    Problem:
    The Win2K servers have about 56Gb of data in each array. BE starts the job
    and at any point after that it could fail. It has in alomost every case backed
    up around 30%-70% of the array and then failed. Lastnight it backed up one
    of the servers in its entirety, however; the second server failed about 85%
    into the backup (this varies). NOTE: the servers stay running but become
    unavailabe to the network. The same amount of data is being backed up on
    the NT4 Servers no problem.

    LOG:
    Drive is not responding. Backup set aborted.
    ^ ^ ^ ^ ^

    NOTE:
    BENT Multi-Server ED 7 backuped up all data on all servers without a single
    problem(Of course I needed the module for an Autoloader feature). This was
    not an upgrade the Server was rebuilt with JUST a new drive.



  2. Re: BENT 8.5, NT4Domain, Win2K RAID Svr crash

    Are the raid controllers different in the 2000 machines vs. the 4.0? We
    have seen something similar but it only affects our computers that have
    ami raid controllers.

    David Cross wrote:

    >Have the follwing;
    >
    >Network:
    > + Complete Fiber switched using Catalyst 4006's
    > - Only one segment
    > - No Routers or RSM's installed
    > - All Servers mentioned have 3COM Gigabit NIC's
    >
    >Backup Server:
    > + BENT/2000 8.5 running on 2K Server SP1. PII-266
    > - HP SureStore DDS4 DAT 40x6.
    >
    >NT4 Servers:
    > + Backups up NT4 PDC, BDC's and all Member Svs FINE all SP6a
    >
    >WIN2K Servers:
    >The problems start when the job gets to the (2) Win2K Servers
    > + Each Server is a Dual PIII800 gateway
    > - each has a 170GB H/W RAID 5 array
    > - System/Boot are Mirrored
    > - drives are LVD UltraSCSI 160
    > - each server has the BE 8.5 Agent running
    > - NO ACTIVE DIRECTORY
    >
    >Problem:
    >The Win2K servers have about 56Gb of data in each array. BE starts the job
    >and at any point after that it could fail. It has in alomost every case backed
    >up around 30%-70% of the array and then failed. Lastnight it backed up one
    >of the servers in its entirety, however; the second server failed about 85%
    >into the backup (this varies). NOTE: the servers stay running but become
    >unavailabe to the network. The same amount of data is being backed up on
    >the NT4 Servers no problem.
    >
    >LOG:
    >Drive is not responding. Backup set aborted.
    >^ ^ ^ ^ ^
    >
    >NOTE:
    >BENT Multi-Server ED 7 backuped up all data on all servers without a single
    >problem(Of course I needed the module for an Autoloader feature). This was
    >not an upgrade the Server was rebuilt with JUST a new drive.
    >
    >



  3. Re: BENT 8.5, NT4Domain, Win2K RAID Svr crash


    YES the controllers are different. In fact the ones in the gateway 7400 servers
    appear in Device Manger as LSI Logic ULTRA3 controllers But on Gateways web
    site the are refered to as AMI Mega Raid controllers.
    The NT4 machines have an Adaptec 39160 doing S/W Raid. It's not a real H/W
    RAID controller.


    Mack Sessoms wrote:
    >Are the raid controllers different in the 2000 machines vs. the 4.0? We


    >have seen something similar but it only affects our computers that have


    >ami raid controllers.
    >
    >David Cross wrote:
    >
    >>Have the follwing;
    >>
    >>Network:
    >> + Complete Fiber switched using Catalyst 4006's
    >> - Only one segment
    >> - No Routers or RSM's installed
    >> - All Servers mentioned have 3COM Gigabit NIC's
    >>
    >>Backup Server:
    >> + BENT/2000 8.5 running on 2K Server SP1. PII-266
    >> - HP SureStore DDS4 DAT 40x6.
    >>
    >>NT4 Servers:
    >> + Backups up NT4 PDC, BDC's and all Member Svs FINE all SP6a
    >>
    >>WIN2K Servers:
    >>The problems start when the job gets to the (2) Win2K Servers
    >> + Each Server is a Dual PIII800 gateway
    >> - each has a 170GB H/W RAID 5 array
    >> - System/Boot are Mirrored
    >> - drives are LVD UltraSCSI 160
    >> - each server has the BE 8.5 Agent running
    >> - NO ACTIVE DIRECTORY
    >>
    >>Problem:
    >>The Win2K servers have about 56Gb of data in each array. BE starts the

    job
    >>and at any point after that it could fail. It has in alomost every case

    backed
    >>up around 30%-70% of the array and then failed. Lastnight it backed up

    one
    >>of the servers in its entirety, however; the second server failed about

    85%
    >>into the backup (this varies). NOTE: the servers stay running but become
    >>unavailabe to the network. The same amount of data is being backed up on
    >>the NT4 Servers no problem.
    >>
    >>LOG:
    >>Drive is not responding. Backup set aborted.
    >>^ ^ ^ ^ ^
    >>
    >>NOTE:
    >>BENT Multi-Server ED 7 backuped up all data on all servers without a single
    >>problem(Of course I needed the module for an Autoloader feature). This

    was
    >>not an upgrade the Server was rebuilt with JUST a new drive.
    >>
    >>

    >



+ Reply to Thread