[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference iosg::all-in-1_v30

Title:*OLD* ALL-IN-1 (tm) Support Conference
Notice:Closed - See Note 4331.l to move to IOSG::ALL-IN-1
Moderator:IOSG::PYE
Created:Thu Jan 30 1992
Last Modified:Tue Jan 23 1996
Last Successful Update:Fri Jun 06 1997
Number of topics:4343
Total number of notes:18308

3816.0. "WPL$LIST_PROC queue on other node, startup fails" by GVPROD::ZAGURY (Alain Zagury@GEO - 821/4401) Wed Jan 26 1994 16:53

    Has anybody had this problem too : on a cluster, if the batch queue 
    WPL$LIST_PROC is defined on NODEA, and NODEB is rebooted first, the 
    A1V30START.COM procedure exits with the error :  
    
    Starting the WPS-PLUS batch queue for background LIST and SORT jobs   
    %JBC-E-NOSUCHNODE, remote node is unknown or not reachable
    
    Is this a known/reported problem fixed in a FPR or should I SPR it ?
    
    (ALL-IN-1 V3.0A, VMS 5.5-2)
    
    Regards,
    Alain Zagury
T.RTitleUserPersonal
Name		DateLines
3816.1never did get an answer to this oneAIMTEC::WICKS_AAtlanta's Most (In)famous WelshmanWed Jan 26 1994 16:597
    Alain
    
    It used to be THR-15813 if that one is still open.
    
    regards,
    
    Andrew.D.Wicks
3816.2It's closed, but I've opened another (THR-18680)...IOSG::PYEGraham - ALL-IN-1 Sorcerer's ApprenticeWed Jan 26 1994 17:5814
    Andy,
    
    You should have done, since I answered it on 24-mar-92 and it was
    closed on 17-aug-92.
    
    In my answer to this bug, which referred to this problem causing an
    installation to fail, I said that I didn't think we should suppress the
    error since it was important to report it.
    
    The circumstance .0 reports is one I didn't think of at the time, so
    I'll open another bug accordingly. What I intend we should do is only
    start the queue if it's on the current node.
    
    Graham
3816.3Do we have a workaround ?GIDDAY::SETHIOh what a feeling .....Tue Feb 01 1994 00:1313
    Hi Graham,

    Do we have a workaround for this problem ?  The customer has stated
    that their machines may reboot automagically and if the node that has
    the queue define does not come up before say node A they cannot
    guarantee that ALL-IN-1 will be up and running on node A.

    Can we have generic queues ?  If so what do we have to do to tailor the
    startup procedure ?

    Regards,

    Sunil
3816.4Field Test site sought :-)IOSG::PYEGraham - ALL-IN-1 Sorcerer's ApprenticeWed Feb 02 1994 09:5228
    Sunil,
    
    We haven't written the code yet, but it's just going to be something
    like this:
    
$ say " Starting the WPS-PLUS batch queue for background List and Sort jobs"
$ if f$getqui( "DISPLAY_QUEUE","QUEUE_NAME","WPL$LIST_PROC" ) .eqs. ""
$    then
$       initialize/ queue/ batch/ start/ wsdef=256/ wsextent=2048 -
          /wsquota=512/ job_limit=4 WPL$LIST_PROC
$    else
$       if (f$getqui("DISPLAY_QUEUE","QUEUE_STOPPED","WPL$LIST_PROC") .or. -
            f$getqui("DISPLAY_QUEUE","QUEUE_PAUSED","WPL$LIST_PROC") .or. -
            f$getqui("DISPLAY_QUEUE","QUEUE_STOPPING","WPL$LIST_PROC") .or. -
            f$getqui("DISPLAY_QUEUE","QUEUE_PAUSING","WPL$LIST_PROC") .or. -
            f$getqui("DISPLAY_QUEUE","QUEUE_STALLED","WPL$LIST_PROC")
           ) .and. -
           f$getsyi( "cluster_member", -                    
                     f$getqui("DISPLAY_QUEUE","SCSNODE_NAME","WPL$LIST_PROC") -
                   ) then -
           start/queue WPL$LIST_PROC
$ endif

    Of course we haven't tried this yet, and we're not sure if it's
    completely the right solution. So, if your customer would like to test
    this, we'd be most grateful!
    
    Graham