[cisco-voip] Heads up on CSCsx32588 - impacting all Unity Connection 7.0(1) installs

[Pat Hayes]

For anyone running Unity Connection or CUCMBE 7.0(1), please be sure
to take a look at defect CSCsx32588, it is currently causing a fair
bit of trouble, see bug notes pasted below. Also documented here:

https://supportforums.cisco.com/docs/DOC-9120

The long and short of it is, you should get to 7.0(1)ES34 or later,
otherwise you will eventually run into this condition. All releases
from 7.0(2) and up also contain the fix. Once you are impacted (see
symptoms), the only way out of it is to call TAC and have them get in
there with root access to fix it. If you can upgrade prior to that,
though, you can avoid the whole mess. See the notes in the condition
section below for how to check your current status if you are running
an impacted version.

Symptom:
Some users hear failsafe.

Not able to delete messages. They get deleted but still exsist.

MWI is not working. If you manually try to reset it, you get the
following error: ISAM error: no free disk space
Could not reset Message Waiting Indicators

Trying to set traces you get the following error:
An error occured while saving: java.sql.SQLException

Applog shows the following errors:
Jan 28 10:52:09 BPTBNVMA070A local7 3 : 971: Jan 28 03:52:09 PM.344
UTC : %CUC_CSMGR-UCEVNT-3-EvtMiuDBWriterSQLExecError: Media component
(Miu) DBWriter encountered the following error executing SQL:
[0x8DBEFEF1] Failed executing stored procedure [0x8DBEFEF1] Error
0x8DBEFEF1 [0x8DBEFEF1] DbError: FAILED (hr=0x8dbefef1 IX000 : .
SprocRequest=[proc='UnityDirDb:csp_VMSServerModify'
params=[pObjectId='bba07c41-84a1-4a7c-a502-0d02a8b845db'
pAllowNoNewCalls='0'] reqId='f715f] App ID:CuCsMgr Cluster ID: Node
ID:BPTBNVMA070A

Saw error in the CuCsMgr log like this:
01/28/2009 10:56:28.663
|10651,AMER-US-CUCM-2-015,1FE25EF2F12D454D89E6C29C123653B9,CsMalUmss,18,
src=0 desc=IX000 : Could not insert new row into the table. [Class
IX:IX000] err=-1|

Looks like you can't add anything else to the DB.

CuCsMgr can also show the following errors:

01/27/2009 18:07:15.984 |13169,,,MiuGeneral,25,Miu DBWriter thread
error executing SQL: [0x8DBEFEF1] Failed executing stored procedure
[0x8DBEFEF1] Error 0x8DBEFEF1 [0x8DBEFEF1] DbError: FAILED
(hr=0x8dbefef1 IX000 : Could not insert new row into the table. [Class
IX:IX000]). Stored procedure request =
[proc='UnityDirDb:csp_VMSServerModify'
params=[pObjectId='bba07c41-84a1-4a7c-a502-0d02a8b845db'
pAllowNoNewCalls='0'] reqId='f715f8fa-4353-48b5-baef-8d40c7a04769']..
SQL=[proc='UnityDirDb:csp_VMSServerModify'
params=[pObjectId='bba07c41-84a1-4a7c-a502-0d02a8b845db'
pAllowNoNewCalls='0'] reqId='f715f8fa-4353-48b5-baef-8d40c7a04769']|

01/27/2009 18:07:16.023 |13169,,,-1,-1,Media component (Miu) DBWriter
encountered the following error executing SQL: [0x8DBEFEF1] Failed
executing stored procedure [0x8DBEFEF1] Error 0x8DBEFEF1 [0x8DBEFEF1]
DbError: FAILED (hr=0x8dbefef1 IX000 : Could not insert new row into
the table. [Class IX:IX000]). Stored procedure request =
[proc='UnityDirDb:csp_VMSServerModify'
params=[pObjectId='bba07c41-84a1-4a7c-a502-0d02a8b845db'
pAllowNoNewCalls='0'] reqId='f715f8fa-4353-48b5-baef-8d40c7a04769']..
SprocRequest=[proc='UnityDirDb:csp_VMSServerModify'
params=[pObjectId='bba07c41-84a1-4a7c-a502-0d02a8b845db'
pAllowNoNewCalls='0'] reqId='f715f8fa-4353-48b5-baef-8d40c7a04769']|


Conditions:
Unity Connection 7.0(1).
This is fixed in Connection 7.0(1)ES34 and 7.0(2).

To determine if you are hitting this bug refer to the following:
---------------------------------------------------------------------------------------------------------------------
There have been situations where a particular database table has
outgrown its early limits. In situations like this, we can view the
utilization though a command:

run cuc dbquery unitydyndb select * from vw_tableinformation

This will return a table of all the tables in unitydyndb. To analyze
the disk utilization, look for the column titled dbspacepagesfree.
This value should always be greater than zero. If it is close to zero
then we know a problem is brewing. Another command to issue is:

show cuc dbserver disk

This will return a list of the system partitions. Look for the column
titled Free MB. This should also always be greater than zero. If it is
close to zero then we know a problem is brewing. Typically the only
tables which grow into a state where you would want to check it is the
dynamic database. In issuing the above two commands, if you see the
dbspacepagesfree is zero, and the Free MB for dyn and
/var/opt/cisco/connection/db/dyn_dbs is zero then you are running into
this defect.
---------------------------------------------------------------------------------------------------------------------

Workaround:

Contact TAC, requires root access