--- Dave Shield <[EMAIL PROTECTED]> wrote:
> On Wed, 2005-06-15 at 19:12, Steve S. Law wrote:
>
> > > How do the iterator 'get_{first,next}' hook
> routines
> > > obtain the list of entries to work with?
> > > Do you have an internal list, or something
> similar?
> > > How is this generated?
>
> > Iterator routines are generated by mib2c tool.
> There
> > is a global pointer to the list -
> > struct applTable_entry *applTable_head , which is
> a
> > linked list, also generated by mib2c
>
> Yes - mib2c generated the framework of the module
> implementation, including the code to look through
> a linked list.
>
> But how is this list populated?
> Are you monitoring some underlying system, or is
> this
> module purely internal to the SNMP agent?
sub-agent is monitoring some external application
processes. For each process being monitored, there
will be a row in the table. Whenever snmpget command
is made, sub-agent will send the request in a format
that is understood by monitored process. The monitored
process then returns the value of the requested
columnar object to the sub-agent. The data only exists
in external processes. Sub-agent simply gathers
information and populates the MIB table. When external
processes start up, they register with sub-agent by
sending a message containing a row ID and PID which
together form an index of the table. Sub-agent then
creates the rows for those processes by calling
applTable_createEntry function in a loop. If there are
10 external processes started up, then the MIB will
have 10 rows, but their columns will have no value yet
(either zero or NULL) because no SNMP get request has
been issued yet.
>
> What does an individual row of the table represent?
>
>
>
> > For my situation, I just need to write/implement a
> new
> > function applTable_removeEntry which will look for
> the
> > entry and delete it from the linked list. Correct?
>
> Not necessarily, no.
> The MIB data doesn't exist in isolation - it's the
> SNMP
> representation of some external subsystem. When one
> of
> these external "things" disappears, then the
> corresponding
> row in the table should be removed. Conversely,
> when
> a row is deleted from the table, it should have a
> knock-on
> effect on the relevant external "thing".
I want to delete the row when that external process
dies. One of the variables I monitor is
applOperStatus. If a process is dead and someone tries
to query the process, there will be time out error and
sub-agent will return applOperStatus as "down" for
that process. Now I want to delete the row. I should ,
shouldn't I? and my question is the way to implement
is just find the row corresponding to the process in
the list and if found, remove it from the list. All
this needs to be coded by myself of course?
>
> That's why I'm asking about this link. How you
> build
> the linked list will affect how to handle deletions.
>
> Dave
>
>
-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
_______________________________________________
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
