Prev: move database from sparc to linux
Next: dfs lock handle
From: The Magnet on 21 Jun 2010 09:15 Hi, We are running RMAN on 10gR2 without a recovery catalog. Everything will be running fine, maybe for days or weeks. Then RMAN comes up with errors like this: RMAN-06207: WARNING: 33 objects could not be deleted for DISK channel(s) due RMAN-06208: to mismatched status. Use CROSSCHECK command to fix status Why randomly does RMAN decide to throw errors? We use the following script of select / parse to make a report of results: SELECT REPLACE(session_recid ||':'|| operation ||':'|| object_type ||':'|| status,' ','-') || ':|' FROM v$rman_status WHERE TRUNC(start_time) = TRUNC(SYSDATE); Thoughts?
From: Arne Ortlinghaus on 21 Jun 2010 09:53 Hi "Magnet", in our RMAN scripts we have added the following commands before making a new backup: delete obsolete; crosscheck backup; change archivelog all crosscheck; delete expired backup; Arne Ortlinghaus ACS Data Systems ----- Original Message ----- From: "The Magnet" <art(a)unsu.com> Newsgroups: comp.databases.oracle.server Sent: Monday, June 21, 2010 3:15 PM Subject: RMAN > > Hi, > > We are running RMAN on 10gR2 without a recovery catalog. Everything > will be running fine, maybe for days or weeks. Then RMAN comes up > with errors like this: > > RMAN-06207: WARNING: 33 objects could not be deleted for DISK > channel(s) due > RMAN-06208: to mismatched status. Use CROSSCHECK command to > fix status > > Why randomly does RMAN decide to throw errors? We use the following > script of select / parse to make a report of results: > > SELECT REPLACE(session_recid ||':'|| operation ||':'|| object_type > ||':'|| status,' ','-') || ':|' > FROM v$rman_status > WHERE TRUNC(start_time) = TRUNC(SYSDATE); > > Thoughts? >
From: joel garry on 21 Jun 2010 13:46 On Jun 21, 6:15 am, The Magnet <a...(a)unsu.com> wrote: > Hi, > > We are running RMAN on 10gR2 without a recovery catalog. Everything > will be running fine, maybe for days or weeks. Then RMAN comes up > with errors like this: > > RMAN-06207: WARNING: 33 objects could not be deleted for DISK > channel(s) due > RMAN-06208: to mismatched status. Use CROSSCHECK command to > fix status > > Why randomly does RMAN decide to throw errors? We use the following > script of select / parse to make a report of results: > > SELECT REPLACE(session_recid ||':'|| operation ||':'|| object_type > ||':'|| status,' ','-') || ':|' > FROM v$rman_status > WHERE TRUNC(start_time) = TRUNC(SYSDATE); > > Thoughts? Possible problems you haven't told us enough to discount are: You are deleting things with the OS and there is an error in your logic, or a conflict with retention time. Standby db or you've enabled and disabled an archive destination. jg -- @home.com is bogus. http://www.signonsandiego.com/news/2010/jun/20/app-helps-drivers-avoid-trouble-traffic-and-cops/
From: John Hurley on 22 Jun 2010 07:45 Magnet: > We are running RMAN on 10gR2 without a recovery catalog. Everything > will be running fine, maybe for days or weeks. Then RMAN comes up > with errors like this: > > RMAN-06207: WARNING: 33 objects could not be deleted for DISK > channel(s) due > RMAN-06208: to mismatched status. Use CROSSCHECK command to > fix status > > Why randomly does RMAN decide to throw errors? We use the following > script of select / parse to make a report of results: > > SELECT REPLACE(session_recid ||':'|| operation ||':'|| object_type > ||':'|| status,' ','-') || ':|' > FROM v$rman_status > WHERE TRUNC(start_time) = TRUNC(SYSDATE); > > Thoughts? If RMAN is telling you that you are missing CROSSCHECKs somewhere ... you are probably missing CROSSCHECKs somewhere. Maybe some combination of scheduled stuff and some people doing some things manually? Possibly some jobs in the database scheduler failing from time to time?
|
Pages: 1 Prev: move database from sparc to linux Next: dfs lock handle |