GTM adding LTM as server - Reason no reply from big3d : timed out

Could be multiple reasons, but usually indicates an inability of the two devices to communicate (access control or routing).

First, are you adding the LTM pair as a BIG-IP System (Redundant)? It should be done this way rather than as individual BIG-IP systems.

In order to see what is happening, try running iqdump to see if your GTM can talk to your LTM. Run this from the GTM CLI (bash prompt):

iqdump LTMipaddress

Yes I am adding the LTM pair as a redundant pair. Actually if I remove all the manually created VS under the GTM group then the server is shown as green (this is probably default behavior if there is no VS to check I presume). So it does look like it's working but it is odd that no VS is being detected under that LTM group (unless it only detects a very specific type/class of VS).

I also tried the IQDUMP as you suggested but it is not that clear what I should be looking for, I do see that under "server", there are traffic and it shows a status of 1 as active.

The iqdump suggestion was meant to ensure that your GTM and LTM were communicating properly with each other. If there's a big3d communication failure between the two units (which it seems there is according to your log message), iqdump should indicate that.

So in the GTM under Global Traffic -> Servers, does the LTM HA pair server object show as green? And is it just the virtual server discovery is not working?

Yes the LTM HA pair does show up as green but no VS is being discovered even though I have active VS on the LTM.

Did you perform a big3d install from the GTM to your LTMs?

http://support.f5.com/kb/en-us/solutions/public/13000/300/sol13312.html

If so, what versions of big3d are running on your GTM and LTM?

big3d -v

I read that article before and I know I used "bigip_add" but I am not sure if I also used "big3d_install", anyway I used the latter now but still no VS.

The LTMs/GTM are all running the same version 11.4.1.608.0 (they are all running VE versions if that makes any difference).

Are you still seeing the bigip monitor timeout messages in the GTM log? If so it may indicate a communications problem between your GTM and LTMs. I see no other reason why this shouldn't be working. Can you verify that all of the supporting layer 2 and 3 networking is correct and that no layer 4-7 network devices are blocking communications?

Nope, no more error/timeout anymore when I do "show gtm server virtual-servers" as the VS are now mark as active.

I can telnet to both LTM self-IPs (off the same vlan as the GTM) off 4353, and I can also telnet to the VS on the LTM off all its service ports (80/443/135). There is nothing else in between as they are all VMs (this is a lab environment which I am just doing some testing).

May be best to open a case with F5 support at this point. It sounds like everything is configured the way it should be.

Thanks for your help. :)