[c-nsp] ace loadbalance prolem

[Matthew Huff]

Care to post a sanitized configuration? Is it a bridged vlan, or LB on a stick (Client NAT)?



-----Original Message-----
From: cisco-nsp-bounces at puck.nether.net [mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Arne Larsen / Region Nordjylland
Sent: Friday, November 02, 2012 3:07 PM
To: 'cisco-nsp at puck.nether.net'
Subject: [c-nsp] ace loadbalance prolem

Hi all
Is there someone that has seen something similar.
I've got four ace load-balancer, two ace blade and two 4710.
Both setup's are connected to a 6500-vss environment, the two 4710 with an etherchannel, and the others located in the chassi.
We have an application that is used for medical journal purpose.
Some client tasks makes a lot off GET functions when browsing a patients journal.
The load balancing inject a delay of 2-3 sec, and that is on both setup's. Direct calls to the servers are much faster
I thought the is was a bandwidth problem, but no packet are being drop, or at least I can see.
If I check the resources of the sbl's they are not overloaded at all.
I've trying to figure out the past couple of days, what this is.
I'm pretty sure that it has to do with the amount of concurrent connections, but I just can't see any counters, counting any errors.
I've had a Cisco consultant to look at it, also with out any luck.
Is there someone that could guide me in the right direction.

/Arne


_______________________________________________
cisco-nsp mailing list  cisco-nsp at puck.nether.net
https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/