| Subject | RE: [firebird-support] Can't run GFix |
|---|---|
| Author | Helen Borrie |
| Post date | 2007-12-12T09:07:45Z |
At 05:59 PM 12/12/2007, you wrote:
- allowing logins to a database during a -Replace_database restore
- using a file-copy utility to copy a database whilst it is online
- copying a file-copied database file over an existing database
- and others...
A dying hard disk could also cause a database to get progressively corrupted.
gfix can't fix it. It probably can't even analyse it.
In my opinion, you should not try to keep using this database. It will require professional recovery. Wait until someone returns who has the know-how to track what is going on at your site. I hope someone else can give you some less gloomy opinions.
./heLen
>I am not the DB-admin (at all) so everything is new to me. But I'll try toThis can happen where the database you are using was a file-copy from a different operating system; or where the database is restored on a different OS but the backup file was created as non-transportable. Do you know the history of this database? If these actions are in its history then you do have corruption and it could be extensive, with damaged system tables.
>fiure out that...
>
>OK... Found something... Hope this helps aven a little. Now we actually know
>how to run GFix, it was about we did not know that it must be taken offline
>first (with GFix)...
>
>internal gds software consistency check (decompression overran buffer (179))
>DBSERVER (Client) Wed Nov 28 22:35:53 2007Server crashed.
>C:\Program Files\Firebird\Firebird_1_5\bin\fbserver.exe: terminated
>abnormally (4294967295)
>and BackupBackup encounters an internal error whose cause derives from something beyond the knowledge or control of the Firebird engine. Probably damaged system tables. This corruption is serious, such as misaligned data (due to filesystem incompatibility) or physical damage that has happened because somebody did something stupid to a database while page writes were in progress. "Something stupid" could be:
>
>gbak: 900000 records written
>gbak: ERROR: internal gds software consistency check (decompression overran
>buffer (179))
>gbak: ERROR: gds_$receive failed
>gbak: Exiting before completion due to errors
>gbak: ERROR: internal gds software consistency check (can't continue after
>bugcheck)
- allowing logins to a database during a -Replace_database restore
- using a file-copy utility to copy a database whilst it is online
- copying a file-copied database file over an existing database
- and others...
A dying hard disk could also cause a database to get progressively corrupted.
gfix can't fix it. It probably can't even analyse it.
In my opinion, you should not try to keep using this database. It will require professional recovery. Wait until someone returns who has the know-how to track what is going on at your site. I hope someone else can give you some less gloomy opinions.
./heLen