Good Day,
Our indexing process goes through control files. Each control file should be processed successfully. The log file lists all the DBs that the control file was matched against. Only one DB should match.
So a typical segment, that refers to one control file, from this log file is -
10 Oct 2010 18:04:29,508 main INFO mwpodcast rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO mwvideo rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO fulton rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO atdvideo rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO foxvideo rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO rtvideo rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO newshubvideo rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO smartmoneyvideo rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO smartmoneyaudio rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO apfeed accepted /pmclient/txtsrv/19/76/89/29.ctl
Today, however, I realized that two issues took place –
One control file was processed multiple times successfully as we can see via -
grep "/pmclient/txtsrv/19/76/89/29.ctl" cm.log | grep accepted
10 Oct 2010 18:04:29,508 main INFO apfeed accepted /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,784 main INFO sjprint accepted /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:30,050 main INFO smprint accepted /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:30,320 main INFO smcom accepted /pmclient/txtsrv/19/76/89/29.ctl
And another control file was processed, without finding a match.
Both issues stem from incorrect definitions of either the control file or the DBs.
Therefore, I would like to have an audit process, that will check every control file listed in the log file and ensures that the control file was processed successfully exactly once.
Any ideas how to implement such a thing?
Regards,
Dan
Our indexing process goes through control files. Each control file should be processed successfully. The log file lists all the DBs that the control file was matched against. Only one DB should match.
So a typical segment, that refers to one control file, from this log file is -
10 Oct 2010 18:04:29,508 main INFO mwpodcast rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO mwvideo rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO fulton rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO atdvideo rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO foxvideo rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO rtvideo rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO newshubvideo rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO smartmoneyvideo rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO smartmoneyaudio rejected /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,508 main INFO apfeed accepted /pmclient/txtsrv/19/76/89/29.ctl
Today, however, I realized that two issues took place –
One control file was processed multiple times successfully as we can see via -
grep "/pmclient/txtsrv/19/76/89/29.ctl" cm.log | grep accepted
10 Oct 2010 18:04:29,508 main INFO apfeed accepted /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:29,784 main INFO sjprint accepted /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:30,050 main INFO smprint accepted /pmclient/txtsrv/19/76/89/29.ctl
10 Oct 2010 18:04:30,320 main INFO smcom accepted /pmclient/txtsrv/19/76/89/29.ctl
And another control file was processed, without finding a match.
Both issues stem from incorrect definitions of either the control file or the DBs.
Therefore, I would like to have an audit process, that will check every control file listed in the log file and ensures that the control file was processed successfully exactly once.
Any ideas how to implement such a thing?
Regards,
Dan