Cisco Support Community
cancel
Showing results for 
Search instead for 
Did you mean: 
Announcements

Welcome to Cisco Support Community. We would love to have your feedback.

For an introduction to the new site, click here. And see here for current known issues.

New Member

Sharing experience on failover

Hi all

I have carried out an installation of Unity failover system. During the testing stage, we found two strange behaviours. (1) the first incoming calls can be reached to Primary(active) Unity and the subsequent calls will automatically route to Secondary(inactive) Unity. Then, It will force the Unity failover. (2) Sometimes, when the incoming call reach Primary(active), a "Temp Fail" displays on the phone's screen. All key entry are ignored.

I suspect this symptom is abnormal. Does any person has experience on Unity failover and clarify for me whether this symptom is normal or not.

I also found that the ports are connected/disconnected (Event ID: 101,102) to CCM repeatly when the working status of Unity is "primary(active), secondary(inactive)" but no problem when "primary(inactive), secondary(active). I have upgraded sp C for CCM and set the Call Waiting to False. It seems no improvement at all.

Thx for sharing experience.

Regards

config: CCM 3.1(4b) sp C, Unity 3.13, TSP 6.02(b)

5 REPLIES
New Member

Re: Sharing experience on failover

I had some similar problems when install a Unity Failover for the first time. I was able to clear most of it up by taking the time to set the ports in different Partitions on the callmanagers.

Unity 1 (primary) had 72 ports all in a Voicmail Partition. Unity 2 (failover) had 72 ports in a Voicemail FO partition. All ports would hunt to the next port (either CFB or CFNA) except the last port in each group. This port would be CFNA to the other partition. Phones on the system are not allow to dial the Voicemail FO ports. by doing this I stopped users from hitting the failover Unity and causing it to become active. Once I got this running most of my problem went away.

Let me know if this makes any sense, I will post a more detailed response if you need it.

adam

New Member

Re: Sharing experience on failover

Hi adam

I am very interested in your case. Did u set the same port no on Unity 1 & 2 and What is your Advance Call Forward Hop Flag setting in CCM. Could u mind telling me more detail. Thx.

By the way, the failover backup issue makes me headache. Do u have any backup strategy on failover system. I mean the VM system including Exchange server, DC and 2 Unity boxes because I know Veritas Backup Exec doesn't support Unity failover.

Thx for your valuable experience.

New Member

Re: Sharing experience on failover

Sorry this took so long.

Unity 1 has 72 ports, 29000 - 29071

Unity 2 has 72 ports, 29100 - 29171

I have edited my total number of hops to 144. this will allow the system to cycle through all the ports if it has to. I was having a huge problem when I could not figure out why my failover Unity was always active. I discovered that a user was miss dialing and hitting my Failover causing it to become active. that is when I put all the 29100 -29171 ports in a partition that users could not call, but the Primary Unity could forward to.

As for backup, I am not sure I have figured that all out. I am using the DiRT program. I have scheduled it to run every night, and then back up that directory. I would be interested in knowing what other people have done in this situation.

New Member

Re: Sharing experience on failover

We used soft extensions on all but the pilot VM ports. We had an issue where mis-dialed numbers to our failover ports kept failing the server to the failover.

Use the DIRT to back up the Unity data. Then back up your Exchange server with Backup Exec, etc. You don't need to back up both servers as it is quite easy to re-install Unity Failover and let it re-configure.

When restoring from complete loss, restore your Exchange server Information Store first, then use DIRT to restore Unity configuration, without restoring mailboxes. Then, re-run failover config.

New Member

Re: Sharing experience on failover

Hi Adam,

These days, I tried to solve the problems by reinstalling the Unity and changing some parameters on CCM and Unity. No connect/disconnect problem but I got some problems when testing the failover/failback (manually pressing failover/failback button for serveral times respectively, loosing the LAN...etc). I found the routing rule is changed wrongly. The default value of Attempt forward / Attempt sign-in are changed to false but can not be changed it back to normal. I thought it is the fault in the replication stage of Unity. Any other idea?

For the port setting, I follow the white paper from Cisco. I agree it is not good for failover when mis-dialing or too many calls unless the active Unity is out of service/order. Can u solve the problem by disabling the "failover if call arrive on the inactive secondary" in failover monitor instead of setting different partition because it will have less administrative effort. Any comment?

DiRT makes me headache becuase it needs to install all the things first (win 2K server.....etc). It is very time-consuming. No any other backup tools support failover besides DiRT?

Thx very much.

91
Views
0
Helpful
5
Replies
CreatePlease login to create content