cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
3233
Views
5
Helpful
9
Replies

FCRasSVR database full

Jeremy Janusz
Level 1
Level 1

The FCRasSVR database filled up (2048 MB) a couple of days ago in this UCCE deployment, which is causing issues such as agent statistics not populating in Supervisor Desktop.  I am not able to find anything about this anywhere in any CAD documentation.  Is it ok to just allocate more space to the DB, or does it need to stay 2048 MB?  If it needs to stay, what can be done to free up space?

1 Accepted Solution

Accepted Solutions

shirpati
Cisco Employee
Cisco Employee

I think you are hitting the 2GB size limit of MSDE which is used by the FCRasSVR db

You can shrink the db from SQL Enterprise Manager, right click on the ‘fcrassvr’ (RASCAL) database, and select ‘Shrink Database…’ under ‘All Tasks’.

After clicking the ‘Files…’ button, select the appropriate file that needs to be shrunk (either the data file or the transaction log) and click ‘OK’

You can see the respective size of both the database and the transaction log by right-clicking the database and selecting ‘Properties’ (Data Files and Transaction Log tabs, respectively).

Truncate the below table to freeup all the data

truncate table FcRasStateLogToday

truncate table FcRasRecordLog

truncate table FcRasCallLogWeek

truncate table FcRasCallLogToday

MSmerge_tombstone (this one will clear up most of the data)

You should check that  that all CAD  jobs specially the cleanup one are running correctly(unders sql server agent) these jobs help to purge the db automatically.  Check agent jobs on both servers to make sure they’re running.  If jobs do not run successfully, change their credentials to run under “”. 

Check that the RASCAL replication is working

use select count * from fcrascalllogweek" on both systems databases.  This should now  return the same number of rows if replication is set up correctly if the system is live this might not exactly matchup as it takes some time for the two db's to replicate.

Hope this helps.

Shirish.

View solution in original post

9 Replies 9

shirpati
Cisco Employee
Cisco Employee

I think you are hitting the 2GB size limit of MSDE which is used by the FCRasSVR db

You can shrink the db from SQL Enterprise Manager, right click on the ‘fcrassvr’ (RASCAL) database, and select ‘Shrink Database…’ under ‘All Tasks’.

After clicking the ‘Files…’ button, select the appropriate file that needs to be shrunk (either the data file or the transaction log) and click ‘OK’

You can see the respective size of both the database and the transaction log by right-clicking the database and selecting ‘Properties’ (Data Files and Transaction Log tabs, respectively).

Truncate the below table to freeup all the data

truncate table FcRasStateLogToday

truncate table FcRasRecordLog

truncate table FcRasCallLogWeek

truncate table FcRasCallLogToday

MSmerge_tombstone (this one will clear up most of the data)

You should check that  that all CAD  jobs specially the cleanup one are running correctly(unders sql server agent) these jobs help to purge the db automatically.  Check agent jobs on both servers to make sure they’re running.  If jobs do not run successfully, change their credentials to run under “”. 

Check that the RASCAL replication is working

use select count * from fcrascalllogweek" on both systems databases.  This should now  return the same number of rows if replication is set up correctly if the system is live this might not exactly matchup as it takes some time for the two db's to replicate.

Hope this helps.

Shirish.

A well known problem and Shirish provided excellent advice. Just connect to the MSDE CADSQL database from another box where you have SQL Enterprise running. Go to the SQL server registration wizard and add \CADSQL. Then  follow the help he provided.

Under certain conditions the FACRasSvr can become "overfull" and FACRasSvr DB will show as "(Suspect)". This is a much harder problem to solve, but you can fix it. Post back if this is the case and I'll help you.

Regards,

Geoff

Well, FACRasSvr became overfull, but it does not show as "Suspect" anywhere that I can see.  If I try to connect from SQL Enterprise, it says that it cannot connect to the database due to the file being full, or it is out of memory.  If you wouldn't mind, I would appreciate hearing how to solve this, because I have a TAC case open but they haven't been able to solve it yet.  They had me copy the database over to a full SQL instance, truncate it, then copy it back to the PG (shutting down MSDE SQL at the appropriate times, of course).  I did this, as instructed, but still can't connect to the database. 

Send me an email to geoff AT hp DOT com and I'll send you the document I wrote regarding this. Perhaps it will help you. I am very surprised that the method TAC provided did not work. When you copied the DB to a full SQL instance you were certainly able to connect to it - since you truncated it.

Regards,

Geoff

I have a similar problem.

FCRasSvr_Data.mdf size of about 7GB

Biggest table MSmerge_tombstone 41 million records

What should I do?

Follow the excellent advice above from Shirish and truncate that table. It's the culprit.

Regards,

Geoff

Thank you. So did

1. Stop replication between CAD sides

2. Delete replication on Mssql

3. Truncate

4. Shrink

5. Start replication between CAD sides

FCRasSvr_Data.mdf size 460Mb

Yep, that'll fix it.

Regards,

Geoff

Hi,

I have this error in my IPCC Express not Enterprise

CRS could not allocate space for object FCRascalllogToday in database FCRasSvr
Error: 1105 severity 17 state 2

Also for FCRasStatelogToday

So there is no free space in FCRasSvr

I need to confirm the method I should follow from these three:

1. Right click the FCRasSvr DB > shrink > then choose the transaction log files

or

2. shrink the log files by running the batch file runTruncateHistDBLogs.bat in wfavid

or

3. or use this method:

osql -Usa -Ppassword -ddb_cra

USE database_name

GO

DBCC SHRINKFILE (database_name_log.mdf), where database_name is FCRasSvr

GO

Also Can I TRUNCATE these tables as advised above without affecting the call center health

truncate table FcRasStateLogToday

truncate table FcRasRecordLog

truncate table FcRasCallLogWeek

truncate table FcRasCallLogToday

MSmerge_tombstone (this one will clear up most of the data)

Regards