Cache JDBC [SQLCODE: <-400>:<Fatal error occurred>][Cache Error: <<UNDEFINED>srvPU+3^%SYS.DBSRV *%qsc(486)>][Location: <ServerLoop>]

1,254 views
Skip to first unread message

PavM

unread,
Oct 15, 2015, 5:08:29 AM10/15/15
to Caché, Ensemble, DeepSee
Hi

We are getting the following error when an update statement is executed.

Does anyone know what the root cause of this error could be?

The same update statement and with the same data works on one server but fails on another server. 
This error has been occurring randomly on our servers.


java.sql.SQLException: [SQLCODE: <-400>:<Fatal error occurred>]
[Cache Error: <<UNDEFINED>srvPU+3^%SYS.DBSRV *%qsc(486)>]
[Location: <ServerLoop>]
    at com.intersys.jdbc.CacheConnection.getServerError(CacheConnection.java:1310)
    at com.intersys.jdbc.CacheConnection.processError(CacheConnection.java:1532)
    at com.intersys.jdbc.InStream.readHeader(InStream.java:164)
    at com.intersys.jdbc.CachePreparedStatement.Update(CachePreparedStatement.java:279)
    at com.intersys.jdbc.CachePreparedStatement.executeUpdate(CachePreparedStatement.java:227)
    at weblogic.jdbc.wrapper.PreparedStatement.executeUpdate(PreparedStatement.java:174)


Any help is greatly appreciated.

Pav


Brendan Bannon

unread,
Oct 15, 2015, 5:46:54 AM10/15/15
to intersystems...@googlegroups.com

Pav

 

There is something wrong with the connection info.  The %qsc array is used internally by InterSystems to store information about different cursors you have open for the connection.

 

If you disconnect the process and reconnect do you still get the error?

 

Are you using Connection Pooling?  If Yes you might need to terminate the Cache Process to force us to make a new process with a clean %qsc array.

 

If the problem persists you should report this to InterSystems Support.

 

Brendan

--
--
Caché, Ensemble, DeepSee

---
You received this message because you are subscribed to the Google Groups "Caché, Ensemble, DeepSee" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intersystems-publi...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

PavM

unread,
Oct 15, 2015, 11:05:07 AM10/15/15
to Caché, Ensemble, DeepSee, Brendan...@intersystems.com
Hi Brendan,

Thanks for your response.

It always works when we try re-running the same process. But as it's production, these failures are not acceptable.

We use WebLogic and our connections & connection pools are managed by WebLogic. The parameters are correct as the process is successful 90% of the times.

The 10% failures are what we are trying to diagnose. What could be the root cause? And how can the error be prevented? 

Do you think there could be something in the code?

Is there a restriction of a single JDBC connection for Cache? Does Cache throws locking errors with more than one connection?

I would greatly appreciate any help on this.

Thanks,
Pav

To unsubscribe from this group and stop receiving emails from it, send an email to intersystems-public-cache+unsub...@googlegroups.com.

Brendan Bannon

unread,
Oct 15, 2015, 11:35:41 AM10/15/15
to PavM, Caché, Ensemble, DeepSee

Pav

 

I don’t think this is going to be a simple answer.

 

It would be best if you could open a WRC issue and work with someone on figuring out what is going wrong.

 

As you say the basics must be right if it works 90% of the time.

 

 

You can send an email to sup...@intersystems.com asking for help or you can call 617 621 0700

 

 

Brendan

To unsubscribe from this group and stop receiving emails from it, send an email to intersystems-publi...@googlegroups.com.

PavM

unread,
Oct 15, 2015, 1:26:37 PM10/15/15
to Caché, Ensemble, DeepSee, pavani.m...@gmail.com, Brendan...@intersystems.com
Thanks Brendan! Already opened a ticket last Friday and waiting for them to give me more information.

Thanks,
Pav

To unsubscribe from this group and stop receiving emails from it, send an email to intersystems-public-cache+unsub...@googlegroups.com.

Reply all
Reply to author
Forward
0 new messages