Repeated login attempt to the source DB

[Konec]

Dear Community,

I have CDC replication set up for the Oracle source. It worked for two days without any problems and ingested data. But now I have the following messages in the log:

[SOURCE_CAPTURE ]I: Trying to reconnect to the source database
[SOURCE_CAPTURE ]I: The connection to the source database reestablished

It's been there for 4 hours and the latency is increasing.

I was trying to ping the server and it´s up and running. I also tried directly using the account that QLIK uses to log into the DB and it works.

Do you have any tips on what it could be and how to check/fix it?

Regards,
Martin

[Heinvandenheuvel]

It seems that one step in the connection process is failing. Is there not VERBOSE logging output beyond what you provide? Anyway - switch task logging for SOURCE_CAPTURE to VERBOSE and let it try again. Hopefully the extended logging information will provide a clue as to what is going on. Perhaps some specific Redo/Arch log access fails.

>>  I also tried directly using the account that QLIK uses to log into the DB and it works.

Good!  Now try the TEST button on the source connection, Note: Details for any issue with that will be logged in REPSRV.LOG, not the task log (as there is no task active) for this testing.

Hein.

[Konec]

Hi Hein, thanks for reaction.

1.) 'Test Connection' was successful. 

2.) Verbose was probably helpful ( I just hide the path,...)

It looks that part of the log is corrupted? 

[SOURCE_CAPTURE ]I: Trying to reconnect to the source database
[SOURCE_CAPTURE ]T: Oracle Replicate Log Reader disconnect thread 1
[SOURCE_CAPTURE ]I: The connection to the source database reestablished
[SOURCE_CAPTURE ]V: Failed to fetch from thread '1'
[SOURCE_CAPTURE ]V: Get next log event from thread 1
[SOURCE_CAPTURE ]V: Going to retrieve archived REDO log with sequence *****
[SOURCE_CAPTURE ]T: Going to open Redo Log with original name ****
[SOURCE_CAPTURE ]T: Thread 1 connected to Oracle server ***
[SOURCE_CAPTURE ]T: Going to execute statement **
[SOURCE_CAPTURE ]T: Oracle v9 Redo log header **
[SOURCE_CAPTURE ]T: The header block from Redo log *** is corrupted. Read Redo sequence *** instead of ****. Retry after reconnect
[SOURCE_CAPTURE ]T: Close Redo log ***
[SOURCE_CAPTURE ]I: Trying to reconnect to the source database 

[Heinvandenheuvel]

Yes, that looks bad. The mention of V9 is also worrysome. 

If this is a recent development effort as perhaps suggest by it running for a few days, then I would just experiment with a fresh start 'by timestamp' to 'skip' the bad log. 

Use a select on V$ARCHIVED_LOGS to find out where to start based on time and/or sequence numbers.

If that works and it does not happen again, then be happy!

If it does happen again then you may need to huddle with your DBA to explain this issue. Perhaps the file is being compressed incorrectly?

Hein.

[Konec]

I used the timestamp resume and it works!

I also contact DBA to get more info what´s happened with the log.

Thank you Hein

[DesmondWOO]

Hi @Konec ,

Are the redo log files on the Linux system? "The header block from Redo log *** is corrupted" may be caused by disk cache issue. Please find this article for details.

To determine whether the redo log is corrupt, please refer to this article.

Regards,
Desmond

[deepaksahirwar]

Hi @Konec ,

Welcome! We’re delighted to have you in the Qlik Community portal. Thank you for bringing your question to us.

The issue you’re experiencing with Qlik Replicate seems to be related to the connection to your Oracle source database. The log messages indicate that the connection is being reestablished repeatedly, which could be due to a variety of factors.

Here are a few steps you can take to troubleshoot this issue:

• Check the Logs: Analyze the logs thoroughly to understand the root cause. Look for any error messages or warnings that might indicate what’s causing the connection issue.
• Verbose Logging: Switch task logging for SOURCE_CAPTURE to VERBOSE and let it try again. The extended logging information might provide a clue as to what is going on.
• Test Connection: Try the ‘Test Connection’ button on the source connection. Any issues with this will be logged in REPSRV.LOG, not the task log.
• Check for Corrupted Logs: The logs might indicate that part of the log is corrupted. If this is the case, you might need to skip the bad log.

If none of these steps resolve the issue, it might be helpful to reach out to Qlik Support or the Qlik Community for further assistance.

I hope this information helps! If you have any other questions, feel free to ask.

If our response has been helpful, please consider clicking “Accept as Solution”. 
This will assist other users in easily finding the answer. ‌‌

Best Regards,
Deepak

[SushilKumar]

Hello Team,

The log messages indicate that the connection is being reestablished repeatedly,

Require analysis and it pertain to many scenarios.

1. network issue 

2. using Oracle full Client to connect or using connection strings 

however, it requires analysis of Qlik logs to identify the potential issue along with the error.

regards,

Sushil Kumar