Login | Register For Free | Help
Search for: (Advanced)

Mailing List Archive: nsp: juniper

Strange Latency on EX-4248t virtual chassis

 

 

nsp juniper RSS feed   Index | Next | Previous | View Threaded


sronan at fattoc

Nov 16, 2009, 11:26 AM

Post #1 of 4 (591 views)
Permalink
Strange Latency on EX-4248t virtual chassis

Hoping someone here can help me with the issue I describe below. I know 4u seems like a VERY small amount of latency, but it's an issue for our system.

I have a set of five ex-4248t switches, connected via virtual chassis such that no two switches are more then two hops apart, as show below.

1 to 2
2 to 4
4 to 5
1 to 3
3 to 5

I have host A connected to switch 1, and host B connected to switch 2, my latency is 27u, but when I move host B to switch 3, the latency goes up to 31u, these numbers are 100% repeatable, so it's not just some simple variation in a one time test.

This seems odd, since switch 2 and 3 are both a directly connected to switch 1.

What is even more odd, is the farther away from switch 1 you get numerically, the higher the latency gets (1 to 4 is 34u, and 1 to 5 is 37u)

This is a huge issue for us, as we cannot connect hosts to anything other then switch 1 and 2 currently due to the increase in latency.

Any input from the group would be appreciate.
_______________________________________________
juniper-nsp mailing list juniper-nsp [at] puck
https://puck.nether.net/mailman/listinfo/juniper-nsp


techtalm at gmail

Nov 16, 2009, 10:05 PM

Post #2 of 4 (547 views)
Permalink
Re: Strange Latency on EX-4248t virtual chassis [In reply to]

Who is the RE and who is the Backup RE?
Although SW 1 is connected directly to SW2 and SW3 it goes to one of them
through all the other switches, I think it is SW3 which SW1 goes all the
round.

mtc

-----Original Message-----
From: juniper-nsp-bounces [at] puck
[mailto:juniper-nsp-bounces [at] puck] On Behalf Of Shane Ronan
Sent: Monday, November 16, 2009 9:27 PM
To: juniper-nsp
Subject: [j-nsp] Strange Latency on EX-4248t virtual chassis

Hoping someone here can help me with the issue I describe below. I know 4u
seems like a VERY small amount of latency, but it's an issue for our system.

I have a set of five ex-4248t switches, connected via virtual chassis such
that no two switches are more then two hops apart, as show below.

1 to 2
2 to 4
4 to 5
1 to 3
3 to 5

I have host A connected to switch 1, and host B connected to switch 2, my
latency is 27u, but when I move host B to switch 3, the latency goes up to
31u, these numbers are 100% repeatable, so it's not just some simple
variation in a one time test.

This seems odd, since switch 2 and 3 are both a directly connected to switch
1.

What is even more odd, is the farther away from switch 1 you get
numerically, the higher the latency gets (1 to 4 is 34u, and 1 to 5 is 37u)

This is a huge issue for us, as we cannot connect hosts to anything other
then switch 1 and 2 currently due to the increase in latency.

Any input from the group would be appreciate.
_______________________________________________
juniper-nsp mailing list juniper-nsp [at] puck
https://puck.nether.net/mailman/listinfo/juniper-nsp
No virus found in this incoming message.
Checked by AVG - www.avg.com
Version: 8.5.425 / Virus Database: 270.14.66/2504 - Release Date: 11/16/09
07:43:00

_______________________________________________
juniper-nsp mailing list juniper-nsp [at] puck
https://puck.nether.net/mailman/listinfo/juniper-nsp


sronan at fattoc

Nov 17, 2009, 6:21 AM

Post #3 of 4 (546 views)
Permalink
Re: Strange Latency on EX-4248t virtual chassis [In reply to]

Interesting, so you are saying it doesn't always take the shortest path?


On Nov 17, 2009, at 1:05 AM, <techtalm [at] gmail> <techtalm [at] gmail> wrote:

> Who is the RE and who is the Backup RE?
> Although SW 1 is connected directly to SW2 and SW3 it goes to one of them
> through all the other switches, I think it is SW3 which SW1 goes all the
> round.
>
> mtc
>
> -----Original Message-----
> From: juniper-nsp-bounces [at] puck
> [mailto:juniper-nsp-bounces [at] puck] On Behalf Of Shane Ronan
> Sent: Monday, November 16, 2009 9:27 PM
> To: juniper-nsp
> Subject: [j-nsp] Strange Latency on EX-4248t virtual chassis
>
> Hoping someone here can help me with the issue I describe below. I know 4u
> seems like a VERY small amount of latency, but it's an issue for our system.
>
> I have a set of five ex-4248t switches, connected via virtual chassis such
> that no two switches are more then two hops apart, as show below.
>
> 1 to 2
> 2 to 4
> 4 to 5
> 1 to 3
> 3 to 5
>
> I have host A connected to switch 1, and host B connected to switch 2, my
> latency is 27u, but when I move host B to switch 3, the latency goes up to
> 31u, these numbers are 100% repeatable, so it's not just some simple
> variation in a one time test.
>
> This seems odd, since switch 2 and 3 are both a directly connected to switch
> 1.
>
> What is even more odd, is the farther away from switch 1 you get
> numerically, the higher the latency gets (1 to 4 is 34u, and 1 to 5 is 37u)
>
> This is a huge issue for us, as we cannot connect hosts to anything other
> then switch 1 and 2 currently due to the increase in latency.
>
> Any input from the group would be appreciate.
> _______________________________________________
> juniper-nsp mailing list juniper-nsp [at] puck
> https://puck.nether.net/mailman/listinfo/juniper-nsp
> No virus found in this incoming message.
> Checked by AVG - www.avg.com
> Version: 8.5.425 / Virus Database: 270.14.66/2504 - Release Date: 11/16/09
> 07:43:00
>

_______________________________________________
juniper-nsp mailing list juniper-nsp [at] puck
https://puck.nether.net/mailman/listinfo/juniper-nsp


sronan at fattoc

Nov 17, 2009, 6:48 AM

Post #4 of 4 (554 views)
Permalink
Re: Strange Latency on EX-4248t virtual chassis [In reply to]

SW1 is the RE
SW5 is the Backup
2, 3 & 4 are Linecards

On Nov 17, 2009, at 1:05 AM, <techtalm [at] gmail> <techtalm [at] gmail> wrote:

> Who is the RE and who is the Backup RE?
> Although SW 1 is connected directly to SW2 and SW3 it goes to one of them
> through all the other switches, I think it is SW3 which SW1 goes all the
> round.
>
> mtc
>
> -----Original Message-----
> From: juniper-nsp-bounces [at] puck
> [mailto:juniper-nsp-bounces [at] puck] On Behalf Of Shane Ronan
> Sent: Monday, November 16, 2009 9:27 PM
> To: juniper-nsp
> Subject: [j-nsp] Strange Latency on EX-4248t virtual chassis
>
> Hoping someone here can help me with the issue I describe below. I know 4u
> seems like a VERY small amount of latency, but it's an issue for our system.
>
> I have a set of five ex-4248t switches, connected via virtual chassis such
> that no two switches are more then two hops apart, as show below.
>
> 1 to 2
> 2 to 4
> 4 to 5
> 1 to 3
> 3 to 5
>
> I have host A connected to switch 1, and host B connected to switch 2, my
> latency is 27u, but when I move host B to switch 3, the latency goes up to
> 31u, these numbers are 100% repeatable, so it's not just some simple
> variation in a one time test.
>
> This seems odd, since switch 2 and 3 are both a directly connected to switch
> 1.
>
> What is even more odd, is the farther away from switch 1 you get
> numerically, the higher the latency gets (1 to 4 is 34u, and 1 to 5 is 37u)
>
> This is a huge issue for us, as we cannot connect hosts to anything other
> then switch 1 and 2 currently due to the increase in latency.
>
> Any input from the group would be appreciate.
> _______________________________________________
> juniper-nsp mailing list juniper-nsp [at] puck
> https://puck.nether.net/mailman/listinfo/juniper-nsp
> No virus found in this incoming message.
> Checked by AVG - www.avg.com
> Version: 8.5.425 / Virus Database: 270.14.66/2504 - Release Date: 11/16/09
> 07:43:00
>

_______________________________________________
juniper-nsp mailing list juniper-nsp [at] puck
https://puck.nether.net/mailman/listinfo/juniper-nsp

nsp juniper RSS feed   Index | Next | Previous | View Threaded
 
 


Interested in having your list archived? Contact Gossamer Threads
 
  Web Applications & Managed Hosting Powered by Gossamer Threads Inc.