On Wed, 28 Nov 2012, Gerwin Klein wrote:
This may be entirely unrelated, but I've just had to re-clone the afp hg
repository in my home directory at TUM because it made mercurial crash
on pull, and failed integrity checking.
The only other time I've ever had to do that in the past few years of
using mercurial was because of file corruption due to a broken hard disk
(two cases). If this happens frequently to us, something may be very
wrong with storage on the macbroy/lxbroy machines.
The reliability of NFS at TUM has indeed degraded a bit, maybe already
more than 5 years ago, but it is hard to pin down.
From the many possibilities to host Mercurial repositories for multi-user
access, we have the worst possible solution. We knew this already in 2008
for the initial setup, and now there is this comprehensive overview:
http://mercurial.selenic.com/wiki/PublishingRepositories?action=show&redirect=ServerInstall
We have the "shared disk" configuration, with "Advantages: can use
existing setup", "Disadvantages: generally restricted to intranets, not
generally recommended due to general issues with network filesystem
reliability".
So we still carry a lot of CVS baggage stemming from 1993, not just in the
low-level technological sense.
Nothing new, all known already. I usually ignore it to avoid the trouble
of thinking about more fundamental changes.
Makarius
_______________________________________________
isabelle-dev mailing list
isabelle-...@in.tum.de
https://mailmanbroy.informatik.tu-muenchen.de/mailman/listinfo/isabelle-dev
