AnsweredAssumed Answered

Recover ...

Question asked by greaterthandata on Nov 3, 2017
Latest reply on Nov 4, 2017 by gdurniak

I have a database that died mid-stream yesterday.  Users were in the database and one user was running a large import and the file closed on the server.  Tried multiple times, but failed to reopen the file on the server (verification would run and fail).  moved the file to a local computer and tried to open.  Then tried a compact only (in the recover -> advanced options), which also failed.


Finally ran the default (nothing checked in advanced options) recover and it completed with success.  The results in the dialog box show ...


file blocks: scanned ... and rebuilt xxx blocks , dropped 0 invalid data blocks

schema: scanned fields and tables, 0 items modified

structure: scanned; 0 items modified

field indexes: rebuilt


I then looked at the recover.log and of the roughly 27,000 rows, only 4 lines came back with a result that wasn't 0 ... four 8476 'errors' related to recovering four themes.  We don't use themes in this solution, so no biggie here.


My question is ... it looks like the recover worked beautifully and the problem was only related to indexes (which were rebuilt).  SO ... can I use the recovered file since no damage was present / nothing was lost???


In other words, normally I would use a backup copy (from 1 day ago) to import data from this recovered file, but it seems like since nothing was lost I can bypass this last step and use the newly created (recovered) file.  In fact, it might be better to use the recovered file since the indexes are now clean whereas in the backup they likely would have the same problem.


any thoughts would be extremely helpful!!


Thank you.