Just like in the MySQL service module it really makes sense to provide a
way to inject SQL on the first start of the database cluster.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This should integrate the logging more tightly into systemd, so for
example "systemctl status mysql" actually gives an overview about what's
actually going on.
This removes the logError option attribute, so in case you still want to
write into a logfile, I've introduced an option called extraOptions, so
you can use something like:
services.mysql*.extraOptions = ''
log-error = /var/log/mysql_err.log
'';
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Grub uses mdadm to find out the device it is on, especially when mdadm itself
resides in a separate boot partition. When bootstrapping from a NixOS
installation CD, it's not a big issue because usually the paths from the Nix
store of the installation CD are matching with the ones in the chrooted
environment.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This checks if nixpart is able to mount the filesystems from scratch again, just
with the information provided by the kickstart file.
Found an odd issue about findmnt here, because it seems to not show /mnt/boot,
even though it _is_ mounted and even shows up in /proc/self/mountinfo. I'm not
quite sure whether this is a bug or I'm doing something wrong here, but might
need some investigation.
Mountpoints are checked by adding empty canary files, remounting and checking if
the same canaries still exist. If they don't, the partitioner either has
formatted the filesystem or just not mounted the device. Either way, both
shouldn't happen, but that's why we're testing it, no? :-)
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
As the whole partitioning run is quite an invasive procedure, we want to
especially make sure that it doesn't unmount any filesystems that were mounted
before the partitioner was run.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This will ensure that we don't get errors because the kernel doesn't recognize
the new partitioning scheme on some conditions or architectures, such as i686.
See here for the Hydra build log on i686:
http://hydra.nixos.org/build/5432090/download/1/log.html
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
At the moment, we still use kickstart syntax, but this is going to change
soon[TM] to be more NixOS-integrated. The tests use emptyDiskImages option
introduced in the previous commit, so we don't have to create a whole bunch of
duplicate expressions.
Testing itself is done exclusively on /dev/vdb and /dev/vdc. And there is a
check (ensureSanity) to make sure none of the tests are actually mutating
/dev/vda.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This allows to add additional raw disk images to the VM, which therein are
available as /dev/vdb, /dev/vdc, /dev/vde and so on. Especially when testing
partitioning, this could be useful.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This works around a bug in infinality that causes broken rendering in
some cases. Issue NixOS/nixpkgs#663.
Upstream suggests that "slight" is a better/safer default in any case.
It also looks better, IMHO, YMMV.
