Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations SkipVought on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Avaya MG and LSP Mode

Status
Not open for further replies.

FXLL

Technical User
May 10, 2016
24
0
0
US
Hi Comunity,

I have a question about the Avaya MGs and the LSPs.

Do you know what is the time that an Avaya MG waits before to go to LSP mode without a Recovery Rule set on it?

I mean, we have several G450 in different locations, and we are having issues with the network side (ISP).
they have network flaps from less than 3 mins, all of my G450 have the Recovery Rule set in 3 mins.

The Customer is requesting to remove the Recovery Rule to have the MG goes to LSP and we will manually register into the CM once the Network is stable.

Now I being asked of what is the time that the MG is going to wait until it goes to LSP without a Recovery Rule.

display system-parameters mg-recovery-rule 1
SYSTEM PARAMETERS MEDIA GATEWAY AUTOMATIC RECOVERY RULE
Recovery Rule Number: 1
Rule Name:
Migrate H.248 MG to primary: immediately
Minimum time of network stability: 3


Thanks in Advance--
 
Recovery rules only affect failback to core, not failover to LSP.
Recovery rule with 3 mins means the gateway will only come back from the LSP to the core after 3 minutes of network stability and you can't set it lower than that in case your WAN is flapping.

Without a recovery rule, the gateway in LSP mode will never migrate back.

The recovery rules can be immediate after X minutes, or with 0 calls, or time of day based, or either after hours or with 0 calls.


As far as how long it waits to go to LSP mode, do a "show rec" in the gateway.
005(super)# show rec

RECOVERY TIMES
--------------------------------
Primary Search : 3
Total Search : 10
Transition Point: 1

So, if you show mgc list, and you have entries 1.1.1.1,2.2.2.2,3.3.3.3, then the gateway will try 1.1.1.1 for 3 minutes, then the 2nd entry is past the transition point of 1 (to say, only the 1st of 3 entries is considered 'primary'), at which point it will try entry 2,3,1,2,3,1,2 for 1 minute each for a total of 10 mins before it reboots itself and starts over. I think that's how it works. I can't remember if it will try the 1st entry ever again after 3 minutes after it passed it's transition point until it reboots, but I think that's how it works.

If you want to control how long the gateway waits before trying LSP mode, then have a primary search of that length and ensure the LSP's entry in the mgc list is after the transition point.

 
*you also have your link loss delay timer in the core CM. see pg 27 - It's the time that the core CM keeps the calls up on the failed gateway in the event it re-registers.

In theory, if you have a gateway with stations and trunks, and a call comes in from one of those trunks to a station and they are connected, and the WAN flaps - despite not affecting the media path between the station and trunk, if the gateway recovers within the time of the link loss delay timer, then call features can be re-established. That could be a valid concern for your environment as well if the WAN is unreliable.
 
Thank You Kyle.

That helped me to understand what is the issue here,
they have the Recovery Rule Set to Migrate H.248 MG to primary: immediately

So they are dropping calls each time that there is a Flapp and then the Network is stable for more than 3 mins.

Thanks Again.
 
so then you don't want to migrate back as soon as the network has been stable for 3 minutes. I'd say maybe a recovery rule of 0 active calls is better.
 
You could opt for turning off the automatic recovery and then you are in control , when you are confident the WAN is stable you issue the "enable mg-return" command , this will gracefully return the gateways to core , this command stays active for 60 min , in a PN this is similar to the "get forced-takeover ipserver-inerface" , its really down to personal preference but I genraly opt for the manual recovery or a recovery rule that specifies a minimum time of WAN stability and 0 active calls to ensure minimal impact , the only time this is a problem is on a 24hr operation.

ACSS (UC/SBCE/SM/SME)

Not that they mean a thing anymore , get a brain dump pass the test crash the system.
 
Thank You Both,

That clarify all my concerns about this topic.

Kind Regards-
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top