[isalist] Re: TMG Reboot Connectivity Issue
- From: Jim Harrison <Jim@xxxxxxxxxxxx>
- To: "isalist@xxxxxxxxxxxxx" <isalist@xxxxxxxxxxxxx>
- Date: Fri, 27 May 2011 14:17:15 +0000
I'm starting to suspect a VMWare / NLB compatibility
issue<http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1556>.
Does ESX support cross-host NLB?
What NLB mode (unicast/multicast) are you using?
Have you tried switching between modes to see how things shake out?
From: isalist-bounce@xxxxxxxxxxxxx [mailto:isalist-bounce@xxxxxxxxxxxxx] On
Behalf Of Jerry Young
Sent: Tuesday, May 24, 2011 6:46 AM
To: isalist@xxxxxxxxxxxxx
Subject: [isalist] Re: TMG Reboot Connectivity Issue
Jim,
Yup, I have 4 interfaces on the array servers and NLB is active on all of them.
We're not using DNS RR.
The problem was that TMG2 and EMS1 were on different ESX hosts from where they
couldn't communicate with one another. I moved TMG2 to the same ESX host that
EMS1 and TMG1 are on and the viola, they can talk again. For some reason, I'm
still in the process of convincing the VMWare gent that this is an issue with
his host's networking, either in the VMWare fabric or the switch fabric to
which it's connected. :/
While TMG2 couldn't talk to EMS1, rebooting TMG1 resulted in the same exact
behavior: TMG2 would go "offline".
On Tue, May 24, 2011 at 9:28 AM, Jim Harrison
<Jim@xxxxxxxxxxxx<mailto:Jim@xxxxxxxxxxxx>> wrote:
Is NLB involved at all?
DNS RR?
What's the problem being solved between TMG2 & EMS?
Is it possible that it's exhibiting in a different form for TMG1?
From: isalist-bounce@xxxxxxxxxxxxx<mailto:isalist-bounce@xxxxxxxxxxxxx>
[mailto:isalist-bounce@xxxxxxxxxxxxx<mailto:isalist-bounce@xxxxxxxxxxxxx>] On
Behalf Of Jerry Young
Sent: Saturday, May 21, 2011 7:14 AM
To: isalist@xxxxxxxxxxxxx<mailto:isalist@xxxxxxxxxxxxx>
Subject: [isalist] TMG Reboot Connectivity Issue
All,
I am experiencing this now and it's peculiar. Can anyone explain this
behavior? I'm thinking it's by design but want to understand the reasoning.
I have two array servers, TMG1 and TMG2. I have a separate Enterprise
Management Server (EMS1) acting as the Configuration Storage Server for both
TMG boxes.
Currently, TMG2 cannot communicate with EMS1. I know why and should be able to
resolve that soon. What I don't understand is that in this scenario where TMG1
and TMG2 can communicate but only TMG1 can communicate with EMS1, when one of
those TMG boxes are rebooted, the other TMG box drops off the network like it's
gone into some kind of blocking mode while it is rebooting. When the TMG box
that was rebooted comes back up, everything goes back to being good.
I'd love to understand why. :)
--
Cordially yours,
Jerry G. Young II, CISSP
Microsoft Certified Systems Engineer
Young Consulting & Staffing Services Company - Owner
www.youngcss.com<http://www.youngcss.com/>
--
Cordially yours,
Jerry G. Young II, CISSP
Microsoft Certified Systems Engineer
Young Consulting & Staffing Services Company - Owner
www.youngcss.com<http://www.youngcss.com/>
Other related posts:
