New ggate broken?
Daniel Eriksson
daniel_k_eriksson at telia.com
Sat Jul 9 06:47:08 GMT 2005
I wrote:
> This breaks ggate for me pretty badly.
>
> Reading from a ggate device is hosed, it generates bogus data. On the
> client I get tons of these:
>
> bad block -349963294392358026, ino 28874759
> bad block -3509451653830704557, ino 28874759
> bad block -3897387729193514355, ino 28874759
> bad block 3626146008983070931, ino 28874759
> bad block 354768189201479432, ino 28874759
> bad block 6451810762019986516, ino 28874759
> bad block -3206562763347448980, ino 28874759
>
> Both server and client are up-to-date. Server is UP, client is SMP.
>
> I am unfortunately forced to back down to a working version
> because the
> servers need to be up. However, I should be able to help out with some
> testing, just not right away.
The broken version was 2005.07.08.22.00.00, the one I backed down to
(which I now use, so no hardware malfunction) is 2005.07.05.21.00.00.
Also, it looks like writing from the client to the server worked. After
I backed down the files I had written were ok.
One more interesting datapoint is that the exported device was GBDE
encoded (I exported the .bde device on the server).
/Daniel Eriksson
More information about the freebsd-current
mailing list