• October 30, 2020, 10:17:53 AM
Welcome, Guest. Please login or register. Registration is free.
Did you miss your activation email?

Author Topic: VSP7000 Stack Issue  (Read 3052 times)

0 Members and 1 Guest are viewing this topic.

Offline CptnBlues63

  • Sr. Member
  • ****
  • Posts: 100
VSP7000 Stack Issue
« on: October 28, 2013, 04:45:48 PM »
Ok, so I'm trying to create two redundant stacks of VSP7000's interconnected with IST's but I'm having an issue.

We started with two VSP7000's IST'd together.  IST ports were 1/1, Uplink ports were 1/24  We temporarily assigned 1/17 as an IST and 1/18 as an Uplink.  We then joined two VSP7000's (1 per side) to the original pair in a stack down configuration so we have two, two switch stacks with IST's between them.

After creating the stacks and in preparation to move the temp IST from 1/17 to 2/1 and the temp Uplink from 1/18 to 2/24 (their permanent homes) we connected to 2/2 in order to maintain access to the EDM.  However, it wouldn't work.  We double checked all settings (PVID=1, Allowed VLAN's=1, Access) and they were correct but we couldn't even ping the switch's IP much less anything else.  We've tried several ports on switch two and several different VLAN's with no joy.  Yet when we moved to 1/16 (configured the same as 2/2) we did have connectivity.

I've checked through everything I can think of and it all seems to be setup and configured correctly so I'm hoping the problem is I need to reboot both switches in the stack in order for all changes to take effect and switch two to start working as it should.

Is this correct or have I missed something important in the process?


Offline CptnBlues63

  • Sr. Member
  • ****
  • Posts: 100
Re: VSP7000 Stack Issue
« Reply #1 on: October 30, 2013, 10:19:35 AM »
No worries, I figured it out myself this morning when making the necessary changes on the other stack in this cluster.

After making the changes we checked and had no connectivity in switch 2 just like the first side.  So we rebooted the stack.  After doing so, the second switch began communicating properly so we rebooted the first side as well and checked and it began working too.


Offline CptnBlues63

  • Sr. Member
  • ****
  • Posts: 100
Re: VSP7000 Stack Issue
« Reply #2 on: October 30, 2013, 01:17:16 PM »
I spoke too soon.

Once you've unplugged the server from the port on unit 2 it stops working.  After rebooting the second stack, we moved the connection back to 2/2 in the first one and it wasn't working.....again.

Offline Dominik

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 1564
    • Networkautobahn
Re: VSP7000 Stack Issue
« Reply #3 on: October 31, 2013, 12:31:54 PM »
Strange problem, thanks for sharing that information.
I only did VSP7k switchclster setups with single switches so far. Do you have only problems with SMLT ports on
unit 2 or do you have problems with all ports on unit 2 regardless if they are in a SMLT or not ?
Itīs always the networks fault!
networkautobahn.com

Offline TankII

  • Hero Member
  • *****
  • Posts: 556
Re: VSP7000 Stack Issue
« Reply #4 on: October 31, 2013, 03:32:24 PM »
We used to see issued like that on 470's if VLACP was enabled.

TankII

Offline CptnBlues63

  • Sr. Member
  • ****
  • Posts: 100
Re: VSP7000 Stack Issue
« Reply #5 on: November 09, 2013, 09:50:17 AM »
Sorry to take so long replying, I've been away on my annual hunting trip and only returned yesterday.  I will still be away from work until the 18th as I always take the first two weeks in November off for my hunting trip, then we need to butcher the meat and of course, I like a couple extra days to hunt locally too (gotta love living in the Boreal Forest where game is plentiful).

Dominik
The SMLT and IST ports on unit two seem to be working just fine.  But, we haven't unplugged either and then plugged it back in to see if that changes things.

TankII
I don't believe we enabled VLACP on the cluster but I will certainly check that out too.

Offline CptnBlues63

  • Sr. Member
  • ****
  • Posts: 100
Re: VSP7000 Stack Issue
« Reply #6 on: November 20, 2013, 11:12:52 AM »
Ok, so while I was away, my coworker got the ports on unit 2 working.  Before I explain, here's some background.

We started with two production switches IST'd together in a cluster configured as follows:
1/1 - IST
1/24 - Uplink to core switches

1/2-15 - SLT's to servers

1/17 - temp IST (intended for 2/1)
1/18 - temp Uplink (intended for 2/24)

After ensuring the new switches were running the same FW/SW and the config was "factory default" we connected them and brought them up.  Then we connected to 2/2 with a server so we could access the http interface while me moved the temp IST/Uplink's to their permanent locations on unit 2.  The port didn't work (2/2).  We then plugged into 1/16 and it did work so we proceeded to move the IST and Uplink and tested and still 2/2 would not work even though the IST and Uplink ports on unit 2 were.  We rebooted one side of the cluster with a server plugged into 2/2 and it began working.  However, once we removed the connection and then put it back a minute or two later, it stopped working again.

This is where we were at when I left on my hunting trip.

While I was away my coworker plugged the server into 2/18 and it worked.  He discovered that only the ports that mirrored the server SLT connections on unit 1 did not work (ie: 2/2-15) while all the rest did (2/16-23).  So, he made a new SLT, added port 2/2 to the SLT, then removed it and deleted the SLT and voila, the port worked even after unplugging and plugging back in.  He did the same thing to all the affected ports (2/2-15) and now they're all working properly on both stacks one either side of the cluster.

Has anybody else here ever built a cluster of stacked VSP7000's the same way (ie: starting with two single switches configured and in production with an IST between them)?  If so, did you encounter a similar problem?

This makes me think there may be an issue with the SW/FW of the switches that caused this issue that Avaya should look into.