Jump to content


Photo

SNMP & Persistence Issues LB 340

SMNP PERSISTANCE 340 3.6.1.011

This topic has been archived. This means that you cannot reply to this topic.
1 reply to this topic

#1 James Lynes

James Lynes
  • Members
  • 2 posts

Posted 23 November 2011 - 11:12 AM

We are having issues with Adaptive Scheduling and Persistence on our 340 Load Balancers and I'm not sure if they are related.

I have Layer 4 -TCP service which has the persistence time as 0. Yet my clients always return to the same real host. What I want to see, is each request from a client to land on a different real server. Where am i going wrong?


Also I have set this service's Policy to be Weighted Round Robin, with the Adaptive Scheduling of SNMP. I have installed Wireshark to the real servers and i can see that the SNMP request was accepted and then returns the correct value to the Barracuda. However, when I am looking at the Server Health screen, the weight always remain at 100. I have 4 real servers and I was sticking to one real server (issue above), I then set the CPU of the real server to 100% for 10mins and I continued to land on this server each time.

Can someone please give me some guidance.


Thanks

Jim

#2 Neeraj

Neeraj
  • Product Managers
  • 72 posts

Posted 24 November 2011 - 03:26 AM

James,

Is this an ALL port service? Persistence is never off in that case.

The second issue about the weight not changing has been fixed in 4.1 firmware.

-Neeraj