[Glass] cannot recover from tranlogs

[Johan Brichau via Glass]

Hi there,

Following an unclean stone shutdown (filesystem went in read-only mode), I had errors restarting the stone due to file system corruption. 
This is what the stone log said:

>     Extent 0 was not cleanly shutdown.
>     In extent 0, file size is inconsistent.
> 
>     Repository was not shutdown cleanly, recovery needed.
> RDbfRead failed for extentId 0 recordId 784944
> RDbfRead failed for extentId 0 recordId 784947
> 
>     An error was found while loading page allocation.
> 
>     Stone startup has failed.



So OK, this is a dev stone (GS 3.1.0.5) … on to some restore training then..
I copied in a clean extent and restored from the last full backup.
However, restoring from the tranlogs fails. The stone log says:

> --- 12/02/14 00:43:33 CET ---
>     Opened a transaction log file for read_nolocks. 
>        filename = /data/athenadevtranlogs/tranlog963.dbf
>     Restoring from current log directory to end of logs
> 
>     Opened a transaction log file for read_nolocks. 
>        filename = /data/athenadevtranlogs/tranlog964.dbf
> 
>     ***** FATAL ERROR *****
>     Fork in time detected at log location: 964.73.1, expected commitSeq 0.963491, found 0.963481, difference 10 commits
>        Refer to System Administration Guide
>     ForkInTime
> 
>     SHUTDOWN command was generated by  fork in time while reading tranlogs.
> 
>     GemStone is stopping the Admin Gem with session 2, processId 2649
> 
>     Stopping Reclaim Gem for extent 0 , session 3 processId 2647
>     Waiting for all tranlog writes to complete
>     Waiting for Timer thread to stop
>     Waiting for NetRead thread to stop
> 
>     Terminating stone.

What can I do next?

thx
Johan