Unable to view the Log file.

ArndW · Post by **ArndW** » Mon Aug 09, 2010 7:45 am

This error means that the log file for this job has become corrupted. The most common cause is that the disk (at one time) filled up.
For LOG files the solution is easy, just clear the log file.

chulett · Post by **chulett** » Mon Aug 09, 2010 8:24 am

Yup, typically corrupted from a full disk or you've blown through the 2GB Barrier. Issue a CLEAR.FILE RT_LOG43 from the Administrator or a TCL prompt.

ArndW · Post by **ArndW** » Mon Aug 09, 2010 10:07 am

Just doing a full log clear from the director (Job -> Clear Log -> immediate purge of all entries) should do the trick as well, if it doesn't then you can do a CLEAR.FILE (but remember that the purge settings will be gone).

prasanna_anbu · Post by **prasanna_anbu** » Mon Aug 09, 2010 10:33 am

chulett wrote:Yup, typically corrupted from a full disk or you've blown through the 2GB Barrier. Issue a CLEAR.FILE RT_LOG43 from the Administrator or a TCL prompt. ...

chulett , Thanks a lot. it is working fine now.

priyadarshikunal · Post by **priyadarshikunal** » Mon Aug 09, 2010 10:39 am

Then please mark the topic as resolved.

prasanna_anbu · Post by **prasanna_anbu** » Tue Aug 10, 2010 4:16 am

priyadarshikunal wrote:Then please mark the topic as resolved.

Sorry, after rerun again I faced the same error. any help please

priyadarshikunal · Post by **priyadarshikunal** » Tue Aug 10, 2010 4:21 am

the log file was corrupted and you fixed it by clearing the logfile. However you forgot to eliminate the root cause i.e. why they got corrupted. You are running out of disk again it seems.

prasanna_anbu · Post by **prasanna_anbu** » Tue Aug 10, 2010 4:31 am

priyadarshikunal wrote:the log file was corrupted and you fixed it by clearing the logfile. However you forgot to eliminate the root cause i.e. why they got corrupted. You are running out of disk again it seems.

But I can view other sequencer's logs in the same job.

ray.wurlod · Post by **ray.wurlod** » Tue Aug 10, 2010 5:05 am

Please run the following command from the Administrator client:

Code: Select all

UVFIXFILE RT_LOG43

Report your findings.

chulett · Post by **chulett** » Tue Aug 10, 2010 5:55 am

Then I assume you are once again generating a bazillion warning messages and the log has 'overflowed', pushing it past the 2GB threshold. You might want to monitor it real time after it gets 'fixed' again, see what is being logged.

prasanna_anbu · Post by **prasanna_anbu** » Tue Aug 10, 2010 6:58 am

chulett wrote:Then I assume you are once again generating a bazillion warning messages and the log has 'overflowed', pushing it past the 2GB threshold. You might want to monitor it real time after it gets 'fixed' a ...

again I used the CLEAR.FILE RT_LOG43 and reran the job, there is no warnings but there are 119650 entries as of now. is it cause the problem?

chulett · Post by **chulett** » Tue Aug 10, 2010 7:03 am

Too many log messages can cause the corruption issue, yes. All of those are from one run? What kind of messages are being logged?

prasanna_anbu · Post by **prasanna_anbu** » Tue Aug 10, 2010 7:18 am

chulett wrote:Too many log messages can cause the corruption issue, yes. All of those are from one run? What kind of messages are being logged? ...

For each batch id there is a insert statement, there are more than 10000 batch ids. all the insert statements presented in the log file. now the log file entries increased to 138285.

ArndW · Post by **ArndW** » Tue Aug 10, 2010 7:30 am

Then you have reached an impasse. Either reduce the number and size of messages or, better yet, don't log them in the job log but elsewhere. This heavy I/O to the log file is most certainly slowing your job down as well.

chulett · Post by **chulett** » Tue Aug 10, 2010 8:37 am

prasanna_anbu wrote:
chulett wrote:Too many log messages can cause the corruption issue, yes. All of those are from one run? What kind of messages are being logged? ...
For each batch id there is a insert statement, there are more than 10000 batch ids. all the insert statements presented in the log file. now the log file entries increased to 138285.

Why? Under what circumstances are 'all the insert statements' being logged? Perhaps you could post a small sample so we can see just what it is you are talking about...