Conference iosg::all-in-1_v30

    
    Nicholas,
    
    > However, it seems that my RSF has not touched these files at all.
    > Why?
    
    Well, there are a number of errors and SMREORG.COM checks for the
    errors. I always thought that 'On error then Continue' allowed the code
    to continue after the first hit, but wouldn't after a second. After the
    second error, SMREORG.COM just stops, which happens to be before the
    reorganization of SDAF, so it is quite understandable.
    
    > Also, what can be done to prevent these failures?
    
    Appareantly someone was logged in. I don't think it was the POSTMASTER
    causing the problems as the Sender and Fetcher are locked. Someone
    who is allowed to do $ALLIN1/OVERWRITE=SHUTDOWN. The MANAGER?
    
    Ciao,
    
    	Simon
     

    Simon,
    
    I don't think there was anyone with the option /over=shutdown logged
    into the system. The only one who uses and controls the VMS account
    "ALLIN1" and the ALL-IN-1 account "Manager" is me.
    And that specific noght there were no other H-K procedures scheduled!
    What else could it be?
    
    rgrds
    
    Nicholas

    I would agree with Simon's analysis that some process is holding onto
    pending and partition files, and then smreorg.com bombs out. The 
    immediate thought goes to the Sender and Fetcher, but as far as I can 
    see these have been correctly closed down, and your job has moved on
    into start_batch.scp. Here we get the actual shutdown. Now it's just
    possible, that a) the shutdown did not kill all processes (some spawned
    ones still hanging around?) or b) the sm_fc_server_stop.scp did not do
    the business, leaving some locks open on the two files mentioned (have
    you other FC servers on the system?). 
    
    To proceed, you might want to try a show dev/files on those data files,
    or possibly set trace and verify in some of the chain of procedures
    involved e.g. smreorg.com, smjacket.com, smjacket.scp, start_batch.com,
    start_batch.scp and sm_fc_server_stop.scp (all OA$LIB).
    
    Cheers,
    DaveT                          

    ... oh and nearly forgot, check that your manager account has the 
    identifier OAFC$SYSMAN in order to manage servers (long shot!).
    
    Cheers,
    DaveT

    Dave,
    
    the identifier OAFC$SYSMAN does exist in the ALLIN1 VMS user profile.
    However, last Saturday night to Sunday morning at 02:30 am, I had
    scheduled the "test and repair mail areas" H-K procedure. Again the
    procedure exited because the file: OA$DATA_SHARE:PARTITION.DAT was
    locked again! 
    
    Here a few lines from the log
    file:OA$SM_FCVR_MAIL_AREA.LOG1992110802300000;1
    
    
    OA-I-LASTLINE, 02:36am
    %OA-I-LASTLINE, 02:36am
    <CR><LF><CR><LF><CR><LF><CR><LF><CR><LF>        MANAGER finished using
    ALL-IN-1
    ALL-IN-1 File Cabinet Verification and Repair Program
    =====================================================
    
    Version 3.0-13
    
    Could not open partition file OA$DATA_SHARE:PARTITION.DAT
            ?File is locked
    Most likely cause is that the File Cabinet Server has not stopped.
    This error can also occur if accounts are still logged in to ALL-IN-1.
    Terminating program
    %SMJACKET-I Invoking common END_BATCH procedure
    %END_BATCH-I Performing End batch job procedures for "TRM"
    
    
    1. In the log file I can see that the FC Server has been stopped.
    2. When ALL-IN-1 is shutting down it kicks out all users.
    
    Then, what is wrong?
    
    rgrds
    
    Nicholas
    

    Hello Nicholas,
    
    The log file does imply, as I thought, that the FCS is not being 
    closed down. Please check with $sho device/files oa$data:  to see 
    which processes have the partition file open. Perhaps you might want
    to kill them off and try the housekeeping again. 
                                      
    Also, check SM MFC MS to see the status of FC servers on your system.
    Lastly, please set trace to A1TRACE.LOG in
    oa$lib:sm_fc_server_stop.scp, in order to see what is going wrong with
    the FC server shutdown.
    
    Cheers,
    DaveT

    i'm seeing a similar problem on 1 of my systems also.
    
    ALl-IN-1 3.0-1, VMS 5.5
    
    When RSF ran (1st time since upgrade), the following error appeared in
    the log file...
    
    Converting "OA$DATA_SHARE:PENDING.DAT" using FDL file
    "OA$LIB:PENDING.FDL"
    %CONV-F-OPENIN, error opening
    DATA$DISK:[ALLIN1.DATA_SHARE]PENDING.DAT;18 as in
    -RMS-E-FLK, file currently locked by another user
    Convert failed for "OA$DATA_SHARE:PENDING.DAT"
    .
    .
    Shared area reorganization starting
    Converting "DATA$DISK:[ALLIN1.SHARED_E]OA$DAF_E.DAT" using FDL file
    "OA$LIB:SDA
    %CONV-F-OPENIN, error opening
    DATA$DISK:[ALLIN1.SHARED_E]OA$DAF_E.DAT;18 as inp
    -RMS-E-FLK, file currently locked by another user
    Convert failed for "DATA$DISK:[ALLIN1.SHARED_E]OA$DAF_E.DAT"
    .
    .
    
    Looks like all the ohter .DAT files converted.  Since everything else
    looks ok, I'm assuming ( knw i shouldn't do that) that there must have
    been a sub-process going somewhere that didn't get shut down.  what can
    be done to ensure that subprocesses get logged out when ALL-IN-1 shuts
    down?
    
    	ann
    
    

    Ann,
    
    are you sure it wasn't the FCS that had those files locked.
    
    P.S thanks for the "secret" package (:==:)
    
    Regards,
    
    Andrew.D.Wicks

    well... in the top part of the log file, the server was stopped.
    in the bottom of the file, the server wasstarted.
    i looked in the log file for the server startup and it looked ok (no
    errors).
    
    	ann

    ok, anohter RSF question:
    
    this is what the log showed from the run last night:
    
    Converting "DATA$DISK:[OA$SHARA]OA$DAF_A.DAT" using FDL file
    "OA$LIB:SDAF.FDL"
    %SYSTEM-W-NOSUCHFILE, no such file
     \DATA$DISK:[OA$SHARA]OA$DAF_A.DAT;-1\
    %SYSTEM-W-NOSUCHFILE, no such file
     \DATA$DISK:[OA$SHARA]OA$DAF_A.DAT;-1\
    %RMS-E-FNF, file not found
    Convert failed for "DATA$DISK:[OA$SHARA]OA$DAF_A.DAT"
    Converting "DATA$DISK:[A1DATA.DATA_SHARE]OA$DAF_E.DAT" using FDL file
    "OA$LIB:S
    "DATA$DISK:[A1DATA.DATA_SHARE]OA$DAF_E.DAT" converted - was 272541
    blocks, now 
    Converting "DATA$DISK:[OA$SHARD]OA$DAF_D.DAT" using FDL file
    "OA$LIB:SDAF.FDL"
    %SYSTEM-W-NOSUCHFILE, no such file
     \DATA$DISK:[OA$SHARD]OA$DAF_D.DAT;-1\
    %SYSTEM-W-NOSUCHFILE, no such file
     \DATA$DISK:[OA$SHARD]OA$DAF_D.DAT;-1\
    %RMS-E-FNF, file not found
    Convert failed for "DATA$DISK:[OA$SHARD]OA$DAF_D.DAT"
    
    
    i have customized SMREORG on this system to do the daf's in a certain
    order (cause of size) and this did not even try to convert DAF_C. DAF_C
    is on a totally different disk.  SMREORG has also been customized to do
    PURGE/KEEP=1, we just don't have the space for 2 version of all these
    files.
    
    	ann