Commit graph

8722 commits

Author SHA1 Message Date
Eelco Dolstra
bcfe98acff * Prefer building on a remote machine over a local machine. This
makes more sense for the build farm, otherwise every nix-store
  invocation will lead to at least one local build.  Will come up with
  a better solution later...
2008-12-04 16:35:47 +00:00
Eelco Dolstra
9850262a72 * Build RPMs, Debs, coverage analysis. 2008-12-04 15:25:28 +00:00
Eelco Dolstra
909fbb9de1 * When using build hooks, for any nix-store -r build operation, it is
necessary that at least one build hook doesn't return "postpone",
  otherwise nix-store will barf ("waiting for a build slot, yet there
  are no running children").  So inform the build hook when this is
  the case, so that it can start a build even when that would exceed
  the maximum load on a machine.
2008-12-04 14:29:41 +00:00
Eelco Dolstra
5dfba0b4db * Force allocation of a pseudo-terminal to clean up the remote
nix-store process when the connection is interrupted.
2008-12-04 13:36:52 +00:00
Eelco Dolstra
9ccdb80de3 * Don't ignore errors from writing to stderr. That way, when
nix-store -r (or some other operation) is started via ssh, it will
  at least have a chance of terminating quickly when the connection is
  killed.  Right now it just runs to completion, because it never
  notices that stderr is no longer connected to anything.  Of course
  it would be better if sshd would just send a SIGHUP, but it doesn't
  (https://bugzilla.mindrot.org/show_bug.cgi?id=396).
2008-12-04 13:13:31 +00:00
Eelco Dolstra
9fd9c4c635 * Support multiple system types per remote machine, e.g. a machine
list like

    root@example.org x86_64-linux /root/.ssh/id_buildfarm 1
    root@example.org i686-darwin  /root/.ssh/id_buildfarm 1

  This is possible when the Nix installation on example.org itself has
  remote builds enabled.
2008-12-04 12:20:06 +00:00
Eelco Dolstra
63b8f09d8d 2008-12-04 10:45:47 +00:00
Eelco Dolstra
f8713e1287 * Dirty hack to make nix-push work properly on derivations: the
derivation should be a source rather than a derivation dependency of
  the call to the NAR derivation.  Otherwise the derivation (and all
  its dependencies) will be built as a side-effect, which may not even
  succeed.
2008-12-04 10:40:41 +00:00
Eelco Dolstra
82ae85de27 * addToStore() in nix-worker: don't write the NAR dump received from
the client to a temporary directory, as that is highly inefficient.
2008-12-03 18:05:14 +00:00
Eelco Dolstra
5eaf644c99 * A simple API for parsing NAR archives. 2008-12-03 17:30:32 +00:00
Eelco Dolstra
cdee317419 * Backwards compatibility. 2008-12-03 17:02:29 +00:00
Eelco Dolstra
d95b68fde3 2008-12-03 16:15:38 +00:00
Eelco Dolstra
ff762fb499 * Pass HashType values instead of strings. 2008-12-03 16:10:17 +00:00
Eelco Dolstra
1307b22223 * Made addToStore() a lot more efficient: it no longer reads the path
being copied 3 times in the worst case.  It doesn't run in constant space,
  but it didn't do that anyway.
2008-12-03 15:51:17 +00:00
Eelco Dolstra
64519cfd65 * Unify the treatment of sources copied to the store, and recursive
SHA-256 outputs of fixed-output derivations.  I.e. they now produce
  the same store path:

  $ nix-store --add x
  /nix/store/j2fq9qxvvxgqymvpszhs773ncci45xsj-x

  $ nix-store --add-fixed --recursive sha256 x
  /nix/store/j2fq9qxvvxgqymvpszhs773ncci45xsj-x

  the latter being the same as the path that a derivation

    derivation {
      name = "x";
      outputHashAlgo = "sha256";
      outputHashMode = "recursive";
      outputHash = "...";
      ...
    };

  produces.

  This does change the output path for such fixed-output derivations.
  Fortunately they are quite rare.  The most common use is fetchsvn
  calls with SHA-256 hashes.  (There are a handful of those is
  Nixpkgs, mostly unstable development packages.)
  
* Documented the computation of store paths (in store-api.cc).
2008-12-03 15:06:30 +00:00
Eelco Dolstra
09bc0c502c * Install the release notes. 2008-11-29 00:31:39 +00:00
Eelco Dolstra
5d4eb9dd07 * Moved the build farm job for building Nix to the Nix tree. 2008-11-26 01:13:29 +00:00
Eelco Dolstra
0c478d2f4d * Ignore carriage returns. 2008-11-25 02:30:35 +00:00
Eelco Dolstra
5024bde8f4 * Handle prematurely ended logfiles, i.e. make sure we emit enough
close tags.
2008-11-25 01:06:15 +00:00
Eelco Dolstra
2ab09a55cf * Bump the version number. 2008-11-20 21:51:58 +00:00
Eelco Dolstra
6bbff48079 2008-11-20 17:22:42 +00:00
Eelco Dolstra
b8eb32f4d2 * Urgh. 2008-11-20 16:42:52 +00:00
Eelco Dolstra
bba87589cc 2008-11-20 16:28:04 +00:00
Eelco Dolstra
f3e2e6076a 2008-11-20 16:10:55 +00:00
Eelco Dolstra
53cca4a445 2008-11-20 16:06:13 +00:00
Eelco Dolstra
4213b8d8ec * Urgh. 2008-11-20 15:44:59 +00:00
Eelco Dolstra
eb86b6f5a5 2008-11-20 15:08:39 +00:00
Eelco Dolstra
3d2035ea86 * Blindly doing a replacement of occurences of $bindir (when running
the tests) is a bad idea when $bindir = /usr and some programs (like
  perl) live there.  Fortunately it doesn't seem to be needed anymore.
2008-11-20 15:08:34 +00:00
Eelco Dolstra
285d26374a * Don't set the prefix to /nix by default, rather use the Autoconf
default of /usr/local.  However, localstatedir and storedir are set
  to /nix/var/nix and /nix/store respectively unless they're
  explicitly overriden.
2008-11-20 14:14:35 +00:00
Eelco Dolstra
a55113411f * Nix daemon: reload the configuration file after forking (NIX-100). 2008-11-20 12:25:11 +00:00
Eelco Dolstra
c202523e53 2008-11-20 12:01:05 +00:00
Eelco Dolstra
aab530e971 * Primop builtins.storePath for declaring a store path as a
dependency.  `storePath /nix/store/bla' gives exactly the same
  result as `toPath /nix/store/bla', except that the former includes
  /nix/store/bla in the dependency context of the string.

  Useful in some generated Nix expressions like nix-push, which now
  finally does the right thing wrt distributed builds.  (Previously
  the path to be packed wasn't an explicit dependency, so it wouldn't
  be copied to the remote machine.)
2008-11-19 23:26:19 +00:00
Eelco Dolstra
60564410ef * Patterns. 2008-11-19 17:50:25 +00:00
Eelco Dolstra
2668a43388 2008-11-19 17:27:52 +00:00
Eelco Dolstra
63ccd72496 * Updated the manual. 2008-11-19 17:00:32 +00:00
Eelco Dolstra
e13da525a7 * Files in the info directory starting with "." are temporary files
and don't indicate path validity.
2008-11-19 16:27:07 +00:00
Eelco Dolstra
5d250ad1ea * nix-store --dump-db: be more streamy. 2008-11-19 16:26:34 +00:00
Eelco Dolstra
7509d70f9d * Documented some of the sharing mechanisms. 2008-11-19 15:20:46 +00:00
Eelco Dolstra
2369b122d1 * Install documentation in $(docdir) (i.e. share/doc/nix). 2008-11-19 13:19:09 +00:00
Eelco Dolstra
6c2c771af7 * Removed obsolete option. 2008-11-19 13:18:46 +00:00
Eelco Dolstra
07d3a38726 * Remove references to Berkeley DB, including most of the
troubleshooting section.  W00t.
2008-11-19 11:58:33 +00:00
Eelco Dolstra
f5325d292d * Release notes. 2008-11-19 10:59:36 +00:00
Eelco Dolstra
fa791116a3 * Get rid of nix-pack-closure / nix-unpack-closure, they're redundant. 2008-11-18 14:43:40 +00:00
Eelco Dolstra
3f4ed681c2 * Prevent zombies. Previous the SIGCHLD handler only reaped one
zombie at a time, so if multiple children died before the handler
  got to run, some of them would not be cleaned up.
2008-11-14 16:50:01 +00:00
Eelco Dolstra
6fedb7aa0f * Restore SIGPIPE to SIG_DFL when running the builder. This prevents
subtle and often hard-to-reproduce bugs where programs in pipes
  either barf with a "Broken pipe" message or not, depending on the
  exact timing conditions.  This particularly happened in GNU M4 (and
  Bison, which uses M4).
2008-11-14 15:46:45 +00:00
Eelco Dolstra
a519bb0635 * Some somewhat ad hoc mechanism to allow the build farm to monitor
build progress.
2008-11-12 11:08:27 +00:00
Eelco Dolstra
96598e7b06 * Pass the --no-build-output flag to the daemon. 2008-11-11 15:11:10 +00:00
Eelco Dolstra
4166b11a53 * Add /dev/pts to the default nix.conf. 2008-11-11 14:59:20 +00:00
Eelco Dolstra
2b7c839b4e * Typo. 2008-11-11 14:58:37 +00:00
Eelco Dolstra
709b55ee02 * Put the chroots under /nix/var/nix/chroots to reduce the risk of
disasters involving `rm -rf' on bind mounts.  Will try the
  definitive fix (per-process mounts, apparently possible via the
  CLONE_NEWNS flag in clone()) some other time.
2008-10-29 15:34:48 +00:00