[Nix-dev] Troubles with Hydra's failure cache

[Michael Raskin]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Eelco Dolstra wrote:
> Yes, I can do that manually by logging in to the Hydra server and doing "rm
> /nix/var/nix/db/failed/*", then restarting the build (that can actually be done
> through the web interface).  That could be made a bit nicer: the restart button
> in the web interface should get rid of the cached failures.

It would be nice to have, maybe, a tree in Nix repo where committers
could commit set of jobs (+some metainformation for accountability?) and
clean false failures.. Reliance on a small set of people with ssh access
has some scalability drawbacks..
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.12 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJKdvLbAAoJEE6tnN0aWvw3MFgH+gNIEGZkZKrKt0SeUwHFfa16
G9m2P+uwJMtWlOtHMmwUBpe+NjfiDeF1TDLq4FvOE41TRwb54P6309Q1VSnXzOE0
Fghg9+T1eEhwqtmq1P4rEaWxpzfPmOrpoduofAQQZsnJSnY+Di1zRhbDq73MeGPj
FsLgMF89Fs1S9zSX8L04WK/8N9QFVmbBZk+XgxbaqupECS5S68hQeBEGqSnp7LRV
FkyGVQZNE6D5d/u1l8wfZn80X2ZoKzQ0oHWPf/hmI6tiwps/6nGGZLWTRPwmIIx0
YMnM4N53sG7f1rl7J+OKSSz0KEKDzP3j/62Ra/nz5Q3V6r7aHVKQPyyGHNnbXnI=
=vLDp
-----END PGP SIGNATURE-----