The following matters will be written down my own experiences have suffered.
Thus, the right way for me even though others may not fit magnificence.
The following points must be tested when the real test DB with DB, do not you recommend that.
A few days before the DB server has encountered a problem (problem was physical) backup to restore the DB who went to work.
Received during normal backup restore not take it seriously because a lot of work, and I believe I set the server receives the data and then restore the backup began.
DB is to get out, but is not working properly. (The query does not return phenomenon occurred - sysindex error has occurred on the issue had been raised.)
The results confirm that this is a problem with the original DB and backup to the backup, even while there is a problem has become.
Without a backup because several awkward about was the.
And eventually managed to restore one way or another.
Backup and recovery is not possible in the original DB, the new equipment on a per-table basis, one hundred twelve, one was restored by passing glad.
Seriously this time to inform you that one thing should have been
DBA course, if you're familiar with it, but I was wondering if you will tell.
* DBCC CHECKDB (syntax see MSDN).
It is not simple to check whether the problem is in the DB is.
I knew about this before, but important enough that I do not think is very important at this time I saw you being to Bed.
Then, when it is going to be used?
It should be used periodically (once we / one month - too long is ..?!)
Cycle is not a fixed thing. It's only responsibility of the DBA. If you'd every possible diligent DBA.
Recovery should be used in exclusive mode when used.
Anyway, using it regularly have to check for abnormalities when the recovery is to ensure stability.
Also check your periodic Upload stability of the DB.
Thank you.
No comments:
Post a Comment