Using "procfix"...

2006-03-03 Thread Jean-Sebastien Morisset 
Hi everyone,

This morning I found automountd down on a few servers, so I'm looking at
the "procfix" feature in snmpd.conf. The details on it's use are a
little sketchy. For example, is a trap sent for the failure, or is the
"fix" applied first. Is a trap sent after the fix command? Here's my
related config sections:

procautomountd  1 1
procfix automountd  /usr/sbin/svcadm enable autofs
monitor -S -r 128 -o prNames -o prErrMessage "procTable" prErrorFlag != 0

Any info on what I _shoudl_ expect from the above would be appreciated.
:-)

Thanks,
js.
-- 
Jean-Sebastien Morisset, Sr. UNIX Administrator <[EMAIL PROTECTED]>
Personal Home Page 
Underwater and Travel Photographs 


---
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
___
Net-snmp-users mailing list
Net-snmp-users@lists.sourceforge.net
Please see the following page to unsubscribe or change other options:
https://lists.sourceforge.net/lists/listinfo/net-snmp-users

Re: Using "procfix"...

2006-03-03 Thread Dave Shield 
On Fri, 2006-03-03 at 08:47 -0500, Jean-Sebastien Morisset wrote:
> This morning I found automountd down on a few servers, so I'm looking at
> the "procfix" feature in snmpd.conf. The details on it's use are a
> little sketchy. For example, is a trap sent for the failure, or is the
> "fix" applied first. Is a trap sent after the fix command?

Neither.
"proc" and "procfix" don't actually trigger traps at all.
At least not on their own - that's the realm of the (newer)
"monitor" directive.

The basic use of "proc" is to set a flag when it detects an
error.  It's up to an external management application to poll
the agent to spot such error flags.

The basic use of "procfix" is to run a command when it is
explicitly told to do so (by some external management app).
This command will *not* be run automatically (at least not
on its own), or trigger a trap - either before or after
being run.



One possibility with the 5.3 release is to have the DisMan
Event MIB trigger a suitable SET request (instead of sending
a trap), that *will* execute the command automatically.

This would look something as follows (untested!):

setEvent jimllFixIt  prErrFix = 1
monitor -S -r 128 -e jimllFixIt "procTable" prErrorFlag != 0


This functionality is only available with the 5.3 release.
The previous Event MIB implementation did not support set
events, just notifications.

Dave


---
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
___
Net-snmp-users mailing list
Net-snmp-users@lists.sourceforge.net
Please see the following page to unsubscribe or change other options:
https://lists.sourceforge.net/lists/listinfo/net-snmp-users

Re: Using "procfix"...

2006-03-03 Thread Jean-Sebastien Morisset 
On Fri, Mar 03, 2006 at 02:32:11PM +, Dave Shield wrote:
> On Fri, 2006-03-03 at 08:47 -0500, Jean-Sebastien Morisset wrote:
>> This morning I found automountd down on a few servers, so I'm looking at
>> the "procfix" feature in snmpd.conf. The details on it's use are a
>> little sketchy. For example, is a trap sent for the failure, or is the
>> "fix" applied first. Is a trap sent after the fix command?
> 
> Neither.
> "proc" and "procfix" don't actually trigger traps at all.
> At least not on their own - that's the realm of the (newer)
> "monitor" directive.
> 
> The basic use of "proc" is to set a flag when it detects an
> error.  It's up to an external management application to poll
> the agent to spot such error flags.
> 
> The basic use of "procfix" is to run a command when it is
> explicitly told to do so (by some external management app).
> This command will *not* be run automatically (at least not
> on its own), or trigger a trap - either before or after
> being run.
> 
> 
> 
> One possibility with the 5.3 release is to have the DisMan
> Event MIB trigger a suitable SET request (instead of sending
> a trap), that *will* execute the command automatically.
> 
> This would look something as follows (untested!):
> 
> setEvent jimllFixIt  prErrFix = 1
> monitor -S -r 128 -e jimllFixIt "procTable" prErrorFlag != 0
> 
> 
> This functionality is only available with the 5.3 release.
> The previous Event MIB implementation did not support set
> events, just notifications.

Dave,

Excellent - thank you very much for your continued replies. I don't know
where you find the time, but it's very much appreciated.

In your example, would that mean a trap would never be sent, even if the
"procfix" command failed? Is there a way to run the procfix and then
generate a trap if the proc is still missing afterwards (checking the
procfix exit code may not be reliable -- procfix may run ok but not
restore the process)?

Thanks,
js.
-- 
Jean-Sebastien Morisset, Sr. UNIX Administrator <[EMAIL PROTECTED]>
Personal Home Page 
Underwater and Travel Photographs 


---
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
___
Net-snmp-users mailing list
Net-snmp-users@lists.sourceforge.net
Please see the following page to unsubscribe or change other options:
https://lists.sourceforge.net/lists/listinfo/net-snmp-users

Re: Using "procfix"...

2006-03-03 Thread Dave Shield 
On Fri, 2006-03-03 at 09:46 -0500, Jean-Sebastien Morisset wrote:
> Excellent - thank you very much for your continued replies.
> I don't know where you find the time,

No - neither do I.



> In your example, would that mean a trap would never be sent,
> even if the "procfix" command failed?

Correct.  "proc" and "procfix" have *nothing* to do with
traps whatsoever.   The only thing that does is "monitor",
and it's completely arbitrary as to what MIB objects this
reports on.

>   Is there a way to run the procfix and then
> generate a trap if the proc is still missing afterwards

No.

The only possibility would be to wait for the monitor
entry to probe the prError flag again (or to have some
external management application do this instead).

But proc/procfix have no involvement with traps at all.

Dave


---
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
___
Net-snmp-users mailing list
Net-snmp-users@lists.sourceforge.net
Please see the following page to unsubscribe or change other options:
https://lists.sourceforge.net/lists/listinfo/net-snmp-users