Conference csc32::consolemanager

Jo,

shjould be able to if he makes it a literal expression and not a regular
expression

Shaun

Hmmm,
ok I've spoken to this customer now, (someone else took the
troublestatement before) and it seems what he wants is for 
* to be treated as a wildcard, expanded out and mean when he 
checks for it it matches anything so he has basically a test
that can take action A (happens to be to ring a bell) every 
time any event is received from a console being managed. 

After a further chat it turns out he's doing this cause he 
doesn't have a list of all messages that come from certain 
machines, so basically aalthough he is still looking to be 
able to use * as a wildcard alone it seems his problem could 
also be fixed by a scan profile for his clients (MLS+).

So 2 questions - can you get any event to register with 
* without either using another character too (which means
you may miss some) or having it treat every character as 
an event rather than every line?   and does anyone out there 
have an MLS+ scan profile? (hopeful aren't I :-)


Jo. 

    Jo,
     First of all I do *not* recommend doing what he's trying to do. if he
    has a lot of one-shot actions you can kill a system in short order
    with all the process creations. Having said that, I read your note to
    mean that you want every *line* of event text to cause an event to
    occur. To do that, you need to define an event to be a literal
    expression as follows:
    
    *^
    
    which means match on anything followed by a CR, LF or FF  - CR, LF
    being the typical record terminators.
    
    Please talk him out of doing this. 
    
    Regards,
       Dan

Jo,

PCM works in alphabetical order of events defined to a system.

There fore you need to make the wildcard event the last event. Call it
"ZZZZZZZZZZ" or something like that. Then the text string should be "*^".

This will give them what they want - Allthough I'd reccomend against it

Shaun

Hiya chaps, 


well I've been out of the office for a while,
during which time someone else took over this one
and recommended specifying ^ as a regular event. 

The customer still wasn't too happy with this as 
he still gets a consistant stream of beeps (a single 
beep being the action he's trying to get for each 
line that comes up), in fact he says he gets so many 
beeps he has to reboot the machine immediately. 

Now if I understand rightly the recommendation from 
Dan in .3 actually suggests defining *^ (note the 
additional asterix) as a literal expression (instead
of a regular one. 

Will this really make the difference of meaning 
one beep for every line of code rather than one 
beep for every character? 

Not questioning your help just by now I suspect 
I might get an amazing exploding customer (and 
the fall-out will land on me no doubt) if he 
gets another suggestion that doesn't do what we 
say it will :-}


Oh and I'll remember to suggest he names the 
wildcard event ZZZzzzzz so it's last once I 
do talk to him... 


Jo_picking_up_bits_again. 

    >Will this really make the difference of meaning
    >one beep for every line of code rather than one
    >beep for every character?
    
    Yep! Set it up and try it out.
    
    Regs,
      Dan

>  Yep! Set it up and try it out.
 

Preceeded by get h/w, get kit, install product,
configure product etc etc :-) all of which I've 
only just been given the go-ahead/time to do

:-) 

Thanks lots ... this may well save my skin this
time!



Jo.