[cisco-voip] UCCX 10.6 | CUCM 10.5.2 | CAD Extension out of service

[Justin Steinberg]

I always turn off LRO.  I've had weird issues that are next to impossible
to troubleshoot at the application layer and disabling LRO is the solution.
   I would also recommend that if you are using the 2500 user template that
you make sure your callmanagers are using 2 vCPUs when using CCX.  I have
found the system to be much more reliable and stable with 2 vCPU.

Justin

On Tue, Jun 9, 2015 at 8:40 AM, Ryan Huff <ryanhuff at outlook.com> wrote:

> Hi Justin, thank you for your reply.
>
> CUCM 10.5.2su1 is in the active partition on all nodes (10.5.2.11900-3).
>
> LRO is interesting; my elastic sky version is 5.5 and according to the
> DocWiki (http://docwiki.cisco.com/wiki/Disable_LRO) it seems to only be
> an issue with 4.1 and more specifically, with CUCM < 8.6 with ESXi 4.1.
> I've also seen several UC evangelical sites proclaim that LRO needs to be
> disabled where ever Virtual Cisco UC runs, versions be damned.
>
> I'm in pre-flight, so now is definitely the time to power down and knock
> LRO off.
>
> I guess I am more interested in the *why* at this point (plight of an
> engineer). I suppose this issue (if it is even an issue) will self-heal
> itself in v11 when CAD is no longer an option, but still plenty of need to
> support existing CAD.
>
> This is the first time I used this version mix (UCCX 10.6 | CUCM 10.5.2su1
> | CAD), and I haven't ran into this issue before.
>
> Thanks for any thoughts you put into this.
>
> -Ryan
>
>
> ------------------------------
> Date: Tue, 9 Jun 2015 07:41:46 -0400
> Subject: Re: [cisco-voip] UCCX 10.6 | CUCM 10.5.2 | CAD Extension out of
> service
> From: jsteinberg at gmail.com
> To: ryanhuff at outlook.com
> CC: cisco-voip at puck.nether.net
>
>
> Are you on UCM 10.5.2su1 ?   I haven't seen that issue at all.
>
> Is your VMware LRO disabled ?
>
> Do your UCMs have one or two vCPU ?
> On Jun 9, 2015 12:42 AM, "Ryan Huff" <ryanhuff at outlook.com> wrote:
>
> Has anyone been noticing that in this environment (UCCX 10.6 | CUCM 10.5.2
> | CAD); extensions will sometimes report as OutOfService (and subsequently
> cause CAD login issues) when in fact, the phone is registered, associated
> to the jTAPI/RMCMSubsys user, has CTI Control enabled (and is setup
> correctly in all other respects)?
>
> What seems to resolve this issue for me; is to super copy the phone to a
> fake mac address, delete the original and then re-mac the newly created
> device. Doing a cluster reboot seems to address this as well.
>
> This is a new 2 cluster build. Everything about the phone seems to
> function fine within call manager when CAD is having a login issue. I've
> changed the addressing mode on the phone (and CTI ports) to ipv4 only,
> turned JAL off on the phone. Just really odd and not very easy to define.
>
> I know CAD is near death and Finesse is the future, but I have to support
> CAD in this scenario.
>
> Thanks,
>
> Ryan
>
> _______________________________________________
> cisco-voip mailing list
> cisco-voip at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-voip
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://puck.nether.net/pipermail/cisco-voip/attachments/20150609/2add62fb/attachment.html>