[cisco-voip] Expressway Cluster failover for MRA

[Croft, Keith]

>>> 1 - Make sure the primary site handles all MRA registrations and the DR site is only used when the primary is down.
The cluster shares a common DB and distribute the calls across the cluster. If you’re going to use  SRV records to try this,  only use different priorities if you have different clusters allowing failover from one primary cluster to another (secondary) cluster.

Here is what we just deployed to deal with the added load of remote workers. I have a couple of lessons learned.  We recently rebuilt our entire Expressway C and E cluster dedicated to MRA.
We weighted them the same but have 2C’s and 2E’s in DC1 and in DC2 we put their peers, 2 additional  2C’s and 2E’s.  doesn’t matter which DC is impacted we are covered with 1 cluster.

We created a cluster of 4 MRA C and E “Large VM’s”  for a total of 10,000 registrations.  I would avoid Small VMs  (because there's no capacity gain from clustering 2 or more Small VMs this is why we had to  rebuild with Large VM).
If MRA endpoints lose connection to their initial peer, they automatically re-register to another one in the cluster. MRA call preservation is out the window and you will have to place or take another call. Failover to the next server was seamless during testing  (Maintenance mode or rebooting the expressway) by the time the call drops and I place another call my system has already registered to a different peer on the cluster.

Respectfully,
Keith Croft | Collaboration Engineer – IT Infrastructure & Security

From: cisco-voip <cisco-voip-bounces at puck.nether.net> on behalf of "Mark H. Turpin" <mturpin at covene.com>
Date: Tuesday, July 7, 2020 at 2:41 PM
To: Gerence Guan <cisco.guan at gmail.com>, Anthony Holloway <avholloway+cisco-voip at gmail.com>
Cc: cisco-voip voyp list <cisco-voip at puck.nether.net>
Subject: Re: [cisco-voip] Expressway Cluster failover for MRA

I don't believe running unclustered is supported though.

The way I interpreted this section in Unsupported Deployments: https://www.cisco.com/c/en/us/td/docs/voice_ip_comm/expressway/config_guide/X12-6/exwy_b_mra-expressway-deployment-guide/exwy_b_mra-expressway-deployment-guide_chapter_011.html<https://nam05.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.cisco.com%2Fc%2Fen%2Fus%2Ftd%2Fdocs%2Fvoice_ip_comm%2Fexpressway%2Fconfig_guide%2FX12-6%2Fexwy_b_mra-expressway-deployment-guide%2Fexwy_b_mra-expressway-deployment-guide_chapter_011.html&data=02%7C01%7CKeith.Croft%40wwt.com%7C9a9e4931a0d24fe82f0408d822adbe98%7Ca2d8e6b4e26e44218f3dec288c827c7d%7C1%7C0%7C637297476905880278&sdata=hOI%2F7zXPUoOhGUecWCtVs3n6IE92Q%2B1BH6tpaL84Mmg%3D&reserved=0> read to me like you needed to have your C's and E's clustered for your UC zones.

That's just my interpretation, though, I might be wrong.



________________________________
From: cisco-voip <cisco-voip-bounces at puck.nether.net> on behalf of Gerence Guan <cisco.guan at gmail.com>
Sent: Monday, July 6, 2020 8:04 PM
To: Anthony Holloway <avholloway+cisco-voip at gmail.com>
Cc: cisco-voip voyp list <cisco-voip at puck.nether.net>
Subject: Re: [cisco-voip] Expressway Cluster failover for MRA

*** EXTERNAL EMAIL - DO NOT CLICK LINKS ***
@Brian
Clustering is not that critical. As long as the Jabber can register back via the DR without any manual system level changes. It is acceptable even if users need to logout and login jaber again.

@Anthony
it would be good if someone has that table. It will help a lot.










On Tue, Jul 7, 2020 at 1:57 AM Anthony Holloway <avholloway+cisco-voip at gmail.com<mailto:avholloway%2Bcisco-voip at gmail.com>> wrote:
Brian,

This wouldn't support failover in all scenarios though, correct?  E.g., CUCM sub to sub failover.

Does anyone have a nice table of failover scenarios covered and not covered by expressway clustering versus not?

On Mon, Jul 6, 2020 at 9:29 AM Brian Meade <bmeade90 at vt.edu<mailto:bmeade90 at vt.edu>> wrote:
I would not use Expressway clustering and just have 2 different C/E pairs with different SRV Weights/Priorities instead.

On Mon, Jul 6, 2020 at 3:17 AM Gerence Guan <cisco.guan at gmail.com<mailto:cisco.guan at gmail.com>> wrote:
Hi Everyone.

I was googling the answer for MRA failover and found this maillist.
Got a similar setup as Jonathan's environment.
Having a pair of expressway C&E in primary DC, and planning to setup another pair of expressway C&E in the DR site. All MRA should go via primary DC, only use DR site when primary is down.

Can I achieve this with different priorities in SRV? Anyone tested  or make it working?

Best Regards,
Guan


>>> On Jan 28, 2020, at 8:49 PM, Jonathan Charles <jonvoip at gmail.com<https://nam05.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpuck.nether.net%2Fmailman%2Flistinfo%2Fcisco-voip&data=02%7C01%7CKeith.Croft%40wwt.com%7C9a9e4931a0d24fe82f0408d822adbe98%7Ca2d8e6b4e26e44218f3dec288c827c7d%7C1%7C0%7C637297476905890273&sdata=%2BiCz8XEyBOA1pd0l%2BgRSxoNuvD5CIAmTq2oYxieMxbo%3D&reserved=0>> wrote:

>>>

>>> We have two pairs of Expressway clusters (C/E) at two different

>>> locations (primary and DR)...

>>>

>>> The cluster is up, however, we want to make sure that we are in

>>> Active/Standby.

>>>

>>> Currently, we have one of our SRV records for collab-edge set at 5 (the

>>> backup is at 10) with the same weight.

>>>

>>> The clustering guide says we should set the priority and weight on both

>>> SRV records the same, which will cause half of the registrations to go to

>>> the DR site. It is far away and has less capability.

>>>

>>> How do we:

>>>

>>> 1 - Make sure the primary site handles all MRA registrations and the DR

>>> site is only used when the primary is down.

>>> 2 = Make sure failover occurs automatically... currently Jabber users

>>> have to log out and back in to connect to the DR site.

>>>
_______________________________________________
cisco-voip mailing list
cisco-voip at puck.nether.net<mailto:cisco-voip at puck.nether.net>
https://puck.nether.net/mailman/listinfo/cisco-voip<https://nam05.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpuck.nether.net%2Fmailman%2Flistinfo%2Fcisco-voip&data=02%7C01%7CKeith.Croft%40wwt.com%7C9a9e4931a0d24fe82f0408d822adbe98%7Ca2d8e6b4e26e44218f3dec288c827c7d%7C1%7C0%7C637297476905890273&sdata=%2BiCz8XEyBOA1pd0l%2BgRSxoNuvD5CIAmTq2oYxieMxbo%3D&reserved=0>
_______________________________________________
cisco-voip mailing list
cisco-voip at puck.nether.net<mailto:cisco-voip at puck.nether.net>
https://puck.nether.net/mailman/listinfo/cisco-voip<https://nam05.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpuck.nether.net%2Fmailman%2Flistinfo%2Fcisco-voip&data=02%7C01%7CKeith.Croft%40wwt.com%7C9a9e4931a0d24fe82f0408d822adbe98%7Ca2d8e6b4e26e44218f3dec288c827c7d%7C1%7C0%7C637297476905900265&sdata=9GSS1uujt2hn%2Fks%2BxbBO12fZgGdHbJBdDBpcn5xj%2FOk%3D&reserved=0>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://puck.nether.net/pipermail/cisco-voip/attachments/20200707/7b0468f9/attachment.htm>