[cisco-voip] UCCX 10.6.1(SU2) Upgrade

[Jose Colon II]

Just a heads up to anyone upgrading to 10.6.1 SU2 here is what I just
experienced. I have not found any documentation that would indicate this as
normal behavior.

I started my update and everything was fine. Went through and did the
upgrade on node 1 then node 2. Started the switch version on node 1 then
node 2. Everything comes back up and I do a "utils dbreplications
runtimestate" and it shows that it is syncing. I keep an eye on it by
continusly running the command I get some of the following output.

*admin:utils dbreplication runtimestate*

*Server Time: Sun Apr  9 02:41:09 CDT 2017*

*Cluster Replication State: BROADCAST SYNC Started on 1 server(s) at:
2017-04-09-02-35*
*     Info: 28 tables on 1 servers are out of sync. Attempting to correct.*
*     Progress: The system will attempt five possible iterations to
correct. Currently engaged in Attempt #1*
*     Tables Status: Repairing table #8 (cuicreportdefinitionparameter)
against server: uccxcolo02*
*     Use CLI to see detail: 'file view activelog
cm/trace/dbl/20170409_023527_dbl_repl_output_Broadcast.log'*

*DB Version: ccm10_6_1_11002_15*
*Repltimeout set to: 300s*
*PROCESS option set to: 1*

*Cluster Detailed View from uccx01 (2 Servers):*

*                                      PING      DB/RPC/   Replication
 REPLICATION SETUP*
*SERVER-NAME         IP ADDRESS        (msec)    DbMon?    Group ID
(RTMT) & Details*
*-----------         ----------        ------    -------   -----------
 ------------------*
*uccxcolo02          192.168.99.33     2.286     Y/Y/Y     (g_5)
 (0) Syncing...*
*uccx01              192.168.221.23    0.019     Y/Y/Y     (g_2)
 (2) Setup Completed*

*I have multiple of these table telling me that they are trying to be
repaired. *

*admin:utils dbreplication runtimestate*

*Server Time: Sun Apr  9 02:42:46 CDT 2017*

*Cluster Replication State: BROADCAST SYNC Started on 1 server(s) at:
2017-04-09-02-35*
*     Info: 28 tables on 1 servers are out of sync. Attempting to correct.*
*     Progress: The system will attempt five possible iterations to
correct. Currently engaged in Attempt #1*
*     Tables Status: Repairing table #23 (cuicchartseriesmap) against
server: uccxcolo02*
*     Use CLI to see detail: 'file view activelog
cm/trace/dbl/20170409_023527_dbl_repl_output_Broadcast.log'*

*DB Version: ccm10_6_1_11002_15*
*Repltimeout set to: 300s*
*PROCESS option set to: 1*

*Cluster Detailed View from uccx01 (2 Servers):*

*                                      PING      DB/RPC/   Replication
 REPLICATION SETUP*
*SERVER-NAME         IP ADDRESS        (msec)    DbMon?    Group ID
(RTMT) & Details*
*-----------         ----------        ------    -------   -----------
 ------------------*
*uccxcolo02          192.168.99.33     2.318     Y/Y/Y     (g_5)
 (0) Syncing...*
*uccx01              192.168.221.23    0.016     Y/Y/Y     (g_2)
 (2) Setup Completed*

Eventually the repairs completed and the system now shows (2) setup
complete under replication setup for both nodes.

However when going in to test the system I can now see that things are not
right. My live data reports are not showing up its just a blank screen at
the tab. We also use some 3rd party finesse software and those gadgets are
not showing up as well. After a little bit of investigation I found that
after the upgrade the Finesse XML layout had defaulted. We use custom
layouts for all of our teams and they were now all defaulted.

Now I do have a separate DEV/Test system that I use to test upgrades with,
this system is completely live and working along with call manager and
unity all setup as a production DEV/Test system. I did upgrade my DEV/Test
system before upgrading our live production system and this did not happen.
Everything came up as expected and was working without changing the finesse
layout to the default XML settings for finesse.

Not sure why this happened or if it was something with the tables being
repaired for CUIC data during sync. One thing that is different is that my
DEV/Test is not a HA pair. My live production system is a HA pair.

Previous version - 10.6.1.11001-31
Active version - 10.6.1.11002-15


Glad I had my XML layouts saved!

Have a good day!
Jose
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://puck.nether.net/pipermail/cisco-voip/attachments/20170410/9afbb9ba/attachment.html>