For the first time in as long as I can remember (possibly ever, I'm not sure), I had a job failure last night due to a deadlock.
14-Oct 04:30 minbar-dir JobId 27066: Using Device "FileStorage" to write. 14-Oct 04:30 minbar-dir JobId 27066: Error: bdb.h:140 bdb.h:140 update UPDATE Media SET +VolJobs=4,VolFiles=0,VolBlocks=0,VolBytes=232,VolABytes=0,VolHoleBytes=0,VolHoles=0,VolMounts=1,VolErrors=0,VolWrites=1,MaxV +olBytes=0,VolStatus='Append',Slot=0,InChanger=0,VolReadTime=0,VolWriteTime=55,VolType=1,VolParts=0,VolCloudParts=0,LastPartB +ytes=0,LabelType=0,StorageId=16,PoolId=5,VolRetention=2592000,VolUseDuration=82800,MaxVolJobs=0,MaxVolFiles=0,Enabled=1,Loca +tionId=0,ScratchPoolId=0,RecyclePoolId=1,RecycleCount=0,Recycle=0,ActionOnPurge=0,CacheRetention=0,EndBlock=0 WHERE +VolumeName='INCR-20201014-04:30' failed: Deadlock found when trying to get lock; try restarting transaction 14-Oct 04:30 minbar-dir JobId 27066: Fatal error: Catalog error updating Media record. bdb.h:140 update UPDATE Media SET +VolJobs=4,VolFiles=0,VolBlocks=0,VolBytes=232,VolABytes=0,VolHoleBytes=0,VolHoles=0,VolMounts=1,VolErrors=0,VolWrites=1,MaxV +olBytes=0,VolStatus='Append',Slot=0,InChanger=0,VolReadTime=0,VolWriteTime=55,VolType=1,VolParts=0,VolCloudParts=0,LastPartB +ytes=0,LabelType=0,StorageId=16,PoolId=5,VolRetention=2592000,VolUseDuration=82800,MaxVolJobs=0,MaxVolFiles=0,Enabled=1,Loca +tionId=0,ScratchPoolId=0,RecyclePoolId=1,RecycleCount=0,Recycle=0,ActionOnPurge=0,CacheRetention=0,EndBlock=0 WHERE +VolumeName='INCR-20201014-04:30' failed: Deadlock found when trying to get lock; try restarting transaction 14-Oct 04:30 asgard-sd JobId 27066: Fatal error: Error getting Volume info: 1993 Update Media error 14-Oct 04:30 asgard-sd JobId 27066: Warning: Warning cannot use Volume "INCR-20201014-04:30", update_volume_info failed. 14-Oct 04:30 narn-fd JobId 27066: Fatal error: job.c:3012 Bad response from SD to Append Data command. Wanted 3000 OK data , got len=76 msg="3903 Error append data: Error getting Volume info: 1993 Update Media error " The DB itself is even recommending the CORRECT course of action on the part of the application or DB driver here: "Deadlock found when trying to get lock; try restarting transaction" It appears that for once in a blue moon, two jobs tried to perform this operation *exactly* simultaneously. Really, the DB error message here is giving the correct advice: When you encounter a deadlock, the first thing an application should do is back off briefly and then retry the transaction, with updated data if appropriate. It is a fact of life when using SQL databases that occasionally, deadlocks happen. Applications that use databases should be prepared to handle them. This is NOT the database's responsibility to handle. The database's responsibility ends at notifying the application that a deadlock occurred. The database does not know, and cannot know, what the application developer intended or what time constraints exist upon the validity of the data in the transaction. -- Phil Stracchino Babylon Communications ph...@caerllewys.net p...@co.ordinate.org Landline: +1.603.293.8485 Mobile: +1.603.998.6958 _______________________________________________ Bacula-devel mailing list Bacula-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-devel
