On Sun, Jun 20, 2010 at 08:19:37PM -0400, Chris Ball wrote: > Hi Trevor, > > > If it's a race condition in Bazaar, that might explain why I > > didn't have problems when I ran the BE tests earlier. > > I agree, definitely looks like a bzr bug. Interesting that we weren't > hitting this before my last merge with your branch, though -- i.e. my > revno 330 always works, and my revno 331 (post-merge) always fails. > I think it might just be that we have better test coverage now..
I imagine so, but I can't think up a relationship between the problem (lock file release race) and our particular failed test (recursive remove). > I'm surprised there still isn't an upstream bzr fix. We might have to > think up a workaround My favorite workaround is: 1) disable recursive remove test 2) hope nothing else triggers bzr race Not very robust, but the robust fix would be to patch bzr, and I'm not going to learn enough of the bzr internals to be able to catch races :p. -- This email may be signed or encrypted with GPG (http://www.gnupg.org). The GPG signature (if present) will be attached as 'signature.asc'. For more information, see http://en.wikipedia.org/wiki/Pretty_Good_Privacy My public key is at http://www.physics.drexel.edu/~wking/pubkey.txt
pgp0NQb5yzDlU.pgp
Description: PGP signature
_______________________________________________ Be-devel mailing list [email protected] http://void.printf.net/cgi-bin/mailman/listinfo/be-devel
