Search:

Type: Posts; User: TonyB

Page 1 of 9 1 2 3 4

Search: Search took 0.01 seconds.

  1. Replies
    4
    Views
    717

    Remember to check the show inactive versions box...

    Remember to check the show inactive versions box (or use the -inactive commandline option)...
  2. Replies
    13
    Views
    5,616

    We're running on RHEL5 x86_64 with ext3 -...

    We're running on RHEL5 x86_64 with ext3 - admittedly DB2 9.5 (TSM 6.1) rather than 9.7 for 6.2 - no unusual I/O load. Load values typically <1. A lot of threads, of course.

    Is your I/O load...
  3. Replies
    10
    Views
    1,642

    Might be a good idea to hit the big red switch on...

    Might be a good idea to hit the big red switch on the entire library, give it a minute then start 'er up again
  4. Replies
    7
    Views
    2,141

    Try adding /validate:yes to your dsmcutil and let...

    Try adding /validate:yes to your dsmcutil and let us know what the output looks like...

    EDIT: oh, and perhaps "dsmcutil install sched"?
  5. Replies
    10
    Views
    2,376

    So if everything is the same between 300 MiB/s...

    So if everything is the same between 300 MiB/s and 100 MiB/s hosts except the SAN connection and the host...I'd be looking at the SAN connection and the host...not TSM.
  6. Replies
    7
    Views
    2,141

    In polling mode the client contacts the server......

    In polling mode the client contacts the server...

    What dsmcutil command(s) are you using to install the service(s)?
  7. Replies
    5
    Views
    2,242

    Remember though - after you update the threshold...

    Remember though - after you update the threshold to 100% it won't actually stop existing processes - they will continue until completion, so you might want to leave yourself a buffer of time to let...
  8. Replies
    13
    Views
    4,766

    FYI - our 3500 is at 8930...this level stabilised...

    FYI - our 3500 is at 8930...this level stabilised some frame control issues we were having.
  9. Replies
    10
    Views
    1,642

    Ahah - sorry just saw your other thread - you...

    Ahah - sorry just saw your other thread - you also have a 3500. I guess its vaguely possible that the internal frame bus isn't terminated...the boards on the top of each frame have a daisy chained...
  10. Replies
    10
    Views
    1,642

    Sounds like the frame bus within the ATL is...

    Sounds like the frame bus within the ATL is chucking a big old wobbly. Do you have multiple network ports in your library? Another one might provide access - had an MCP problem with a ts3500 a while...
  11. Replies
    10
    Views
    1,642

    A bounce oftens get you a reset - if you're...

    A bounce oftens get you a reset - if you're running with slave instances (and this may include storage agents, although I've not experienced it) you typically need to bounce whichever slave owns the...
  12. Haha I didn't even look - that'll learn me. Yeah,...

    Haha I didn't even look - that'll learn me. Yeah, could be a support call I think. Looks like the instance has chucked a wobbly somewhere between 2010-09-24 13:37:24 and now - the series number is...
  13. Replies
    2
    Views
    1,867

    hmm...it does look reasonable - however I use...

    hmm...it does look reasonable - however I use rete and reto instead of vere and verd...

    Here's some output from my environment, which admittedly doesn't use diffs:



    TYPE CLASS_NAME ...
  14. Replies
    1
    Views
    1,082

    Doesn't work AFAIK - you'll need to schedule an...

    Doesn't work AFAIK - you'll need to schedule an admin session and run it from there (i.e. use a command shell of some sort and a windows scheduler).
  15. Replies
    8
    Views
    703

    I guess its possible you're running into an o/s...

    I guess its possible you're running into an o/s limit - have you run something like "netstat -na" to check connection states? If there's a whole bunch queued at the LISTENing port, or a shedload in...
  16. This feature is by design...no can delete the...

    This feature is by design...no can delete the last one.
  17. Replies
    3
    Views
    633

    You're hitting some hidden tables with the dsmc...

    You're hitting some hidden tables with the dsmc query ... its faster by a long shot too.

    You might want to wrap up a script with some proxy logic - create a dummy node, allow it proxy access to...
  18. Replies
    1
    Views
    1,339

    In terms of the backup scripting there's no real...

    In terms of the backup scripting there's no real difference between running via DSMC SCHED and cron - both just call a script which invokes RMAN with appropriate environment settings.

    To be honest...
  19. Replies
    5
    Views
    2,892

    Hi, yes no problem - at the point you break the...

    Hi, yes no problem - at the point you break the relationship the objects become manageable again. The scripts will need a rewrite where they touch on SQL, the v6 interface is quite different in some...
  20. Replies
    5
    Views
    2,892

    The way I achieved this was by joining the new...

    The way I achieved this was by joining the new instance to the existing v5 instance as a configuration slave, adding all the config I wanted to a profile and subscribing it on the v6 instance. After...
  21. The scan process during a filesystem level...

    The scan process during a filesystem level incremental uses the client object modification date to build its backup list. I think the client-side mtime gets stored in (hidden) backups table at the...
  22. Replies
    3
    Views
    2,204

    Key 03 = "Medium Error" (i.e. the tape and/or...

    Key 03 = "Medium Error" (i.e. the tape and/or drive head)
    ASC 09, ASCQ 00 = "Track following error" (i.e. the tape and/or drive head)

    Assuming you've cleaned the drive and the error isn't just...
  23. Replies
    4
    Views
    2,289

    Aye, that's the one - you need the IP of the...

    Aye, that's the one - you need the IP of the NetApp controller in the IP Address field...
  24. Replies
    4
    Views
    651

    Aye, it'll be archive logs. Frankly I've found...

    Aye, it'll be archive logs.

    Frankly I've found it useful to add this option to the dsmserv.opt: "AllowReorgTable No"
  25. Replies
    4
    Views
    1,456

    Why do you want to clone a node with all of its...

    Why do you want to clone a node with all of its data within TSM? A backupset might be useful, perhaps?
  26. Replies
    4
    Views
    2,289

    How about a "q datamover f=d" Looks like the...

    How about a "q datamover f=d"

    Looks like the HLA of the datamover is pointing at localhost?
  27. Replies
    6
    Views
    2,603

    Ahah interesting - the query inclexcl didn't show...

    Ahah interesting - the query inclexcl didn't show it, presumably as it isn't parseable.

    cjhood has the right of it - you'll only see it at the server.
  28. Replies
    6
    Views
    2,603

    Try running a "dsmc query inclexcl" - might be an...

    Try running a "dsmc query inclexcl" - might be an issue with the server-side cloptset, that command will show you the full set
  29. Replies
    12
    Views
    4,192

    I've got a funny feeling that it isn't :( Mind...

    I've got a funny feeling that it isn't :(

    Mind you - a single 10 Gib link can carry quite a high data rate...over 1 GiB/s - what kind of throughput do you actually need? Might be better off with...
  30. Replies
    12
    Views
    4,192

    Aye - but won't TSM store the two mountpoints in...

    Aye - but won't TSM store the two mountpoints in the volume path, so if one goes away 50% of the volumes go boom?
  31. Replies
    12
    Views
    4,192

    Yep I can groove on that - but surely using 2x...

    Yep I can groove on that - but surely using 2x paths to the FILE class will create a SPOF on 50% of the volumes anyway?
  32. Replies
    12
    Views
    4,192

    Ahh I see - why not configurre an...

    Ahh I see - why not configurre an etherchannel/bonded pair?
  33. Replies
    12
    Views
    4,192

    Hi, Quite possibly it will - my main question...

    Hi,

    Quite possibly it will - my main question is...why do you need to do it this way?
  34. Replies
    10
    Views
    1,951

    I suspect that you're not going to get clean data...

    I suspect that you're not going to get clean data directly from TSM on this. There are a few problems. One you're already on top of - that any summary type records present averages. Additionally,...
  35. Replies
    7
    Views
    4,363

    Perhaps its an environment problem - are you sure...

    Perhaps its an environment problem - are you sure that Get-DatabaseAvailabilityGroup is a viable function when executing via the TSM service?

    Maybe a...


    Set-PSDebug -strict

    ...or if you're...
  36. Replies
    2
    Views
    843

    Typically in this scenario I run 2x schedulers...

    Typically in this scenario I run 2x schedulers (and 2x CADs if you want to go down that route)... one for each TSM client node. That'd collapse your problem down to a client side change.
  37. Replies
    7
    Views
    990

    The "code" tags inside the OP prevent...

    The "code" tags inside the OP prevent formatting...so the page is forced to render wide enough for the really long statement in there...
  38. Replies
    8
    Views
    1,536

    Did you do an import node * (or an import node...

    Did you do an import node * (or an import node dromedary) from that volume? Looks empty from your output...
  39. Replies
    8
    Views
    1,536

    np... I use the same command on instances here...

    np... I use the same command on instances here (although I don't use forced with exported data as I generally delete all exports in one go rather than specific tapes) - is the library client a...
  40. Replies
    8
    Views
    1,536

    Hehe - depends on whether you need the exported...

    Hehe - depends on whether you need the exported data...up to you mate.
  41. Replies
    8
    Views
    1,536

    Could be a volume defined to a stgpool which...

    Could be a volume defined to a stgpool which hasn't been used yet...

    What did the volhistory show on the instance that owned the volume (or a q vol for that matter)?
  42. Thread: Server-free

    by TonyB
    Replies
    4
    Views
    1,070

    It doesn't really do server-free. You're thinking...

    It doesn't really do server-free. You're thinking about something like using disk subsystem based snapshots (e.g. NetApp) perhaps?

    The only "server-free" that TSM does is that it permits NDMP...
  43. Replies
    8
    Views
    1,013

    Aye, kind of... You'll still have an rmtX and...

    Aye, kind of...

    You'll still have an rmtX and an rmtY - but if you look at the location code data in the lsdev output one will be designated primary and the other alternate.
  44. Replies
    8
    Views
    1,013

    The alternate pathing happens within the device...

    The alternate pathing happens within the device driver layer rather than the TSM layer - each rmt device entry in /dev is capable of talking over both paths to the physical device... assuming you...
  45. Replies
    8
    Views
    1,013

    You need to enable alternate pathing at the tape...

    You need to enable alternate pathing at the tape devices (lsdev -Cc tape | awk '{ print $1}' | xargs -n1 chdev -a alt_pathing=yes -l)

    Look for the devices marked primary in the lsdev -Cctape...
Results 1 to 45 of 384
Page 1 of 9 1 2 3 4