[prev in list] [next in list] [prev in thread] [next in thread] 

List:       net-snmp-coders
Subject:    Re: Net-snmp time out issue for snmp walk request
From:       Magnus Fromreide <magfr () lysator ! liu ! se>
Date:       2012-04-16 19:30:40
Message-ID: 1334604640.2124.10.camel () sara
[Download RAW message or body]

On Mon, 2012-04-16 at 15:26 +0530, Suresh kumar wrote:
> Hello All
> 
> We are  using net-snmp 5.7.1 and extended the snmp agent with our
> enterprise specific mibs. We have used " mib2c -c mib2c.iterate.conf"
> option to generate skeleton code for Table objects. Now the tables we
> are implementing contains dynamic data, so as per our implementation,
> the snmpd fetches current data on every time it gets a walk
> request(after catche time out time ). However it takes a long time to
> fetch the data from third party service (CORBA service in our case)
> and loading it in snmp table. During this it responds back timeout for
> the snmp walk request (Timeout: No Response from xxx.xxx.xxx.xx)
> 
> We also tried out using increase cache time out "(define
> XX_PORTGROUPTABLE_TIMEOUT  60 " to 600) and using option –t in snmp
> utility side.
> 
Did this help?
> 
> 1. We would like to know what is the best approach to implement
> dynamic table object in net-snmp? We are re-populating the complete
> table for snmp walk request if the time lag between two successive
> requests exceeds the cache timeout. Is it the right approach ?
> 
> 2. Does net-snmp support retry mechanism? i.e. Since in our case it
> takes around 5-10 minutes to populate complete table, is there any way
> that will keep the client session alive instead of responding timeout.


The client timeout is handled completely in the client so there is no
way for the server to increase it.

I have to admit that 5-10 minutes sounds rather excessive, what is the
structure of your data? Could you split the data set into smaller parts
that you could fetch quicker from the CORBA service? Could you look into
why the fetch operation is so sluggish? Is it because you are fetching a
quadrillion items or is there a bad (quadratic or worse) algorithm
somewhere in there?

/MF



------------------------------------------------------------------------------
For Developers, A Lot Can Happen In A Second.
Boundary is the first to Know...and Tell You.
Monitor Your Applications in Ultra-Fine Resolution. Try it FREE!
http://p.sf.net/sfu/Boundary-d2dvs2
_______________________________________________
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders

[prev in list] [next in list] [prev in thread] [next in thread]