Bruce,

     Thanks for the information, I found the structure in the Admin guide,
     I'm sorry for calling it rubbish ;-). The reason I asked is this: We
     monitor our ADSM servers with T/EC and this works fine. Because I want
     to see errors on the Tivoli receiver and I obviously do not see them
     in my TEC console, I decided to enable those specific errors for the
     fileexit receiver, and then have a Sentry filemonitor watch that file
     for the occurrence of an error. As it turns out, the filemonitor does
     not understand the file, so I'll have to come up with another way of
     catching errors on the Tivoli receiver.

     Are there any plans for a filexit2 receiver that writes one line of
     plain text for every event? (Nothing wrong with some wishful thinking,
     is there?)

     Bye,

     Chris




______________________________ Reply Separator _________________________________
Subject: Re: Rubbish in FILEEXIT file
Author:  "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU> at inet-1
Date:    11/6/98 12:16


Chris,

The file exit record structure is listed in the Administrator's Guide.  The
file exit structure is the same as an user-exit event structure.  The structure
contains the event name, number, commethod, tcpip address, etc.  The main
reasons for writing the file exit event as a data structure are:

1:  Consistency with the user-exit
2:  Easier parse when the structure change in the future - you won't have
different line lengths, you simply read records.
     The fields themselves may change in size in the future, not just the
number of fields.

-Bruce Huang-
IBM ADSM Development
IBM ADSM Development



ADSM-L AT VM.MARIST DOT EDU on 06/11/98 06:54:06
Please respond to ADSM-L AT VM.MARIST DOT EDU
To: ADSM-L AT VM.MARIST DOT EDU
cc:
Subject: Rubbish in FILEEXIT file


     Hi all,

     I enabled some server events for the FILE receiver and now I see those
     events appear in the configured file. So far so good.

     When I try to grep on the file I get "Maximum line length exceeded".
     It looks like all events that occur between 'begin eventl file' and
     'end eventl file' are written on one line.
     I also note the appearance of a lot of null-characters in the file.
     Does anyone know why the server does not write one clean line per
     event?

     Server is 3.1.1.5 on AIX 4.2.1

     Thank you,

     Chris Roelofs