[cisco-voip] CDR flat files not being deleted

[Ruben Montes (EU)]

Hi,
 
I'm not sure how to look at it: could it be CCMCDR sp_unprepare;1 ?
 
Regards,
 
Ruben
________________________________

De: Ryan Ratliff [mailto:rratliff at cisco.com]
Enviado el: mié 02/08/2006 14:40
Para: Ruben Montes (EU)
CC: cisco-voip at puck.nether.net
Asunto: Re: [cisco-voip] CDR flat files not being deleted



Looks like you are right on.  What process is pid 80 for?  The only 
other things that should be accessing CDR are CDRInsert, Aupair, and 
SQL EM.  Anything else is most likely the CIPT job.

-Ryan

On Aug 2, 2006, at 4:14 AM, Ruben Montes ((EU)) wrote:

Hi Ryan,

you are right with your description of the problem: CIPT Optimization 
is still running since the day I am having the eventviewer error. To 
manually kill the pid for this job, this only affects the CDR, am I 
right? As the CDR is not yet being used, I assume I can do what you 
suggest without impacting the other jobs in the database, right?

I think the pid for the job that is causing problems is 80, that is 
blocking pid 62 and pid 62 is blocking some other jobs. Is pid 80 the 
CIPT optimization?

I have only used Enterprise Manager to validate that the replication 
between Pub and Sub is ok and running, so all this stuff is new for 
me...

By the way, we're running CCM 4.1(3)SR3A: should this problem be 
fixed in this version?

Thanks in advance,

Ruben

De: Ryan Ratliff [mailto:rratliff at cisco.com]
Enviado el: mar 01/08/2006 15:08
Para: Ruben Montes (EU)
CC: cisco-voip at puck.nether.net
Asunto: Re: [cisco-voip] CDR flat files not being deleted

Most likely you are running into an issue with sql deadlock.

There are a couple bugs out there, CSCsd95330 being one of them.

Basically Aupair and the CIPT Optimize SQL job get into a deadlock on
the CDR database.  Aupair is trying to remove old records to bring
the number of CDR records down to the max (1.5 million by default)
and the CIPT Optimize job is doing it's thing.  This deadlock keeps
CDRInsert from inserting new records.

If you look in SQL Enterprise Manager under Management->SQL Server
Agent->Jobs the CIPT Optimizations job will still be running.  A
quick check of the history shows it normally only runs for a few
minutes.  Manually killing the pid in SQL for the job will recover
everything and once Aupair finishes it's thing CDRInsert will run
normally.

If you are at all unsure about doing this please open a TAC SR and
unicast me the SR number.

-Ryan

On Aug 1, 2006, at 9:01 AM, Ruben Montes ((EU)) wrote:

Hi,

I've tried to restart the CDRInsert and I've had this "nice" message:

No response to restart the server

I supoose it's time to open a TAC SR??

Regards,

Ruben


-----Mensaje original-----
De: Ryan Ratliff [mailto:rratliff at cisco.com]
Enviado el: mar 01/08/2006 14:37
Para: Ruben Montes (EU)
CC: cisco-voip at puck.nether.net
Asunto: Re: [cisco-voip] CDR flat files not being deleted

Is the CDRInsert service running on your publisher?  If so try
restarting it.  If it is slow to restart (or fails to stop at all)
then it could be that some database activity has blocked CDRInsert.

You can use the command sp_who2 to find out which sql pid has locked
the CDR database.   If you get to that point and need assistance
still please open a TAC SR.

-Ryan

On Aug 1, 2006, at 6:11 AM, Ruben Montes ((EU)) wrote:

Hi,

I'm having some trouble with CDR service. I've noticed that the last
three weeks I have an Eventviewer error every night telling that the
CDR files are being backed up and not deleted.
Nothing has been changed and this has never happened before. What
could have happened?
I have enabled detailed traces but traces/DBL/DCRinsert files are
empty (0Kb)

Thanks in advance,

Ruben
_______________________________________________
cisco-voip mailing list
cisco-voip at puck.nether.net
https://puck.nether.net/mailman/listinfo/cisco-voip








-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://puck.nether.net/pipermail/cisco-voip/attachments/20060802/c3a96fe4/attachment.html