> I suppose the network issue might help present the issue, but it's
> certainly not the cause of it. Malcome Rowe managed to determine the
> algorithm behind the corruption, but unfortunately, neither of us have
> come up with a successful reproduction recipe:
Sounds like a fun bug...
> > It seems to me that this particular user has something 'special' about
> > either her setup, configuration, or methods that makes her more likely to
> > have an issue. I'm trying to get her to reproduce it in an 'unused'
> > repository. The only thing that leaps to mind is her network issues.
> It's worth a shot. I've tried everything I can think of to trigger it,
> but I've never tried disrupting the network traffic. Perhaps it's the
> key we've been missing in the reproduction recipe.
I've been playing around with my 'random binary data commit' script while
simulating networking issues a bit, and I've not *quite* been able to
generate the "Can't read length line in file" error, but I *have* managed
to generate another issue:
"svnadmin: Malformed representation header"
I can't commit or update anything involving my file. I had a number of
transactions in 'test\db', and had trouble removing two of them with
'svnadmin rmtxns test' (svnadmin: Can't open file '8184-4\format': The
system cannot find the path specified).
Is this issue similar to the Can't read length line one? Perhaps just
corruption in another part of the data?
// Andrew MacKenzie | http://www.edespot.com
// GPG public key: http://www.edespot.com/~amackenz/public.key
// You have acquired a scroll entitled 'irk gleknow mizk'(n).--More--
// This is an IBM Manual scroll.--More--
// You are permanently confused.
// -- Dave Decot
Received on Fri Jun 2 00:42:48 2006
- application/pgp-signature attachment: stored