Re: cache problems with table data mib2c

Ah.  I thought there would only be one load per command (like snmptable),
not for every get that is happening.  I think I will avoid the -1 setting.

Thanks.

Repeat question from previous note (in case I go with the autounload option
for some tables) -


-- Re this excerpt from the cache_handler documentation page:

xxx - method to prevent cache from expiring while a request is being
processed (e.g. delegated request). proposal: set a flag, which would be
cleared when request finished (which could be acomplished by a dummy data
list item in agent req info & custom free function).

Has this been implemented, or will it be in the near future?  Thanks.


Kathy McLeod



                                                                           
             Dave Shield                                                   
             <D.T.Shield@liver                                             
             pool.ac.uk>                                                To 
             Sent by:                  Kathy McLeod/Rochester/IBM@IBMUS    
             dave.shield@googl                                          cc 
             email.com                 net-snmp users                      
                                       <[email protected] 
                                       et>                                 
             06/17/2010 03:03                                      Subject 
             AM                        Re: cache problems with table data  
                                       mib2c                               
                                                                           
                                                                           
                                                                           
                                                                           
                                                                           
                                                                           




On 16 June 2010 20:14, Kathy McLeod <[email protected]> wrote:
> Problem 2:  the -1 setting is not working for me for some of the
commands.

"for some of the commands" is a bit vague.
Can you be more specific?

A test such as:
    $ snmpgetnext LLDP-MIB::lldpPortConfigAdminStatus ; \
       snmpgetnext LLDP-MIB::lldpPortConfigAdminStatus ; \
       snmpgetnext LLDP-MIB::lldpPortConfigAdminStatus ; \
       snmpgetnext LLDP-MIB::lldpPortConfigAdminStatus

works as expected:

LLDP-MIB::lldpPortConfigAdminStatus.5 = INTEGER: rxOnly(2)
LLDP-MIB::lldpPortConfigAdminStatus.9 = INTEGER: rxOnly(2)
LLDP-MIB::lldpPortConfigAdminStatus.13 = INTEGER: rxOnly(2)
LLDP-MIB::lldpPortConfigAdminStatus.17 = INTEGER: rxOnly(2)

showing that the cache is being reloaded for every request,
with the index values increasing automatically.


Now that index behaviour might well confuse a command that ends up
sending multiple SNMP requests (such as snmptable or snmpbulkwalk),
since the list of valid rows is changing under its feet.
   And a command such as "snmpwalk" (or "snmptable -CB") relies
on reaching the end of the table, and moving on the next one.  Which
will never happen in this case, so the command will run indefinitely.

But that's a consequence of the re-indexing in your dummy load command.
If you replace this with a fixed allocation
     "for(i=0; i<4; i++)"  instead of  "for(i=j; i<j+4; i++)"
            then everything should work OK.

Dave



------------------------------------------------------------------------------
ThinkGeek and WIRED's GeekDad team up for the Ultimate 
GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the 
lucky parental unit.  See the prize list and enter to win: 
http://p.sf.net/sfu/thinkgeek-promo
_______________________________________________
Net-snmp-users mailing list
[email protected]
Please see the following page to unsubscribe or change other options:
https://lists.sourceforge.net/lists/listinfo/net-snmp-users