[cisco-voip] Strange Unity Connection 8.5.1ES16.11900- 16 DB error - EvtDblsDbeventConnectError

[Pat Hayes]

The "utils dbreplication..." commands that you are running only tell
part of the story. Unity Connection shares a common platform with
CUCM. When replication is set up, there are actually two sets of DBs
that replicate, one is essentially the CUCM db (needed for some
platform functionality) and the others are Unity Connection specific.
The commands that you ran are for the CUCM db. Both need to be healthy
for the cluster to function properly, but the error you cite appears
to be related to the Connection replication. To check the CUC specific
replication, you can use "show cuc cluster status" and to force the
sub to pull down the pub's db, you can run "utils cuc cluster
overwritedb".


On Tue, Jul 19, 2011 at 8:47 AM, Gregory Wenzel <gwenzit at gmail.com> wrote:
> Hello everyone,
>
> I have a strange issue with a Unity Connection HA cluster I just build for a
> client .
> Cisco Unity Connection version: 8.5.1ES16.11900-16
>
> On Cisco MCS7845i4 servers.
>
> Yesterday I had the error message when I tried to access Reporting in Unity
> Serviceability, stated sub could not be reached, I also could not get to the
> web admin page stated "pools exhausted" message thats not the whole message
> but you get the point.
>
> Last night I ran the utils dbreplication forcedatasyncsub command. This
> destroyed the sub database and forces it to take the backup from the pub cuc
> server. After the reboot and waited about an hour, the error messages went
> away on the cuc admin page, i could access reports.
>
>
>
> My customer says these messages are still coming into Syslog.
>
> Jul 19 07:53:27 192.168.42.8 32526: : : 606: pr1unitypub1.client.org: Jul 19
> 2011 11:53:27 AM.254 UTC :  %UC_UCEVNT-3-EvtDblsDbeventConnectError:
> %[AppID=CuCsMgr][ClusterID=][NodeID=pr1unitypub1]: Database synchronization
> thread in component Miu (pid=18712) was unable to connect to the database
> notification publisher.  Error 0x80fe006e. 0x80FE006E
>
>
>
> When I look into the pub and run a Utils dbreplications status command I get
>
> admin:file view activelog
> cm/trace/dbl/sdi/ReplicationStatus.2011_07_19_08_41_40.out
>
> SERVER                 ID STATE    STATUS     QUEUE  CONNECTION CHANGED
> -----------------------------------------------------------------------
> g_aa1unitysub1_ccm8_5_1_11900_21    3 Active   Connected       0 Jul 18
> 23:04:05
> g_pr1unitypub1_ccm8_5_1_11900_21    2 Active   Local
> 0
>
> end of the file reached
>
>
>
> I run this on the Publisher CUC Server.
>
>
>
> I do have to say one other thing, The client states there are no dropped
> packets between these two servers when they look on their firewalls.
>
> Ping time between each server from each server to eachother is 24ms avg.
>
>
>
> The network is as follows.
>
>
>
> CUCPub=====6509====ASA-------<100MB VPN>------ASA======6509====CUC8Sub
>
>
>
> And they upgraded their IPS ASA from v8.0.5 to V8.4.1 three weeks ago on the
> CUCsub side and last week on the CUCpub side.
>
> Im about to get on a TAC call but thought some of you experts may have some
> input. My FW coworker strongly thinks its a fw issue, and I agree, I just
> need to get the unity TAC engineer stuff out of the way before I run down
> that hill...
>
>
>
> Thanks in Advance
>
> --
> Greg Wenzel, CCVP
>
>
> _______________________________________________
> cisco-voip mailing list
> cisco-voip at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-voip
>
>