Commit graph

1093 commits

Author SHA1 Message Date
Eelco Dolstra
caaaff3954 Fix --repair failure on multiple-output derivations
If repair found a corrupted/missing path that depended on a
multiple-output derivation, and some of the outputs of the latter were
not present, it failed with a message like

  error: path ‘/nix/store/cnfn9d5fjys1y93cz9shld2xwaibd7nn-bash-4.3-p42-doc’ is not valid
2016-01-06 21:49:32 +01:00
Eelco Dolstra
743e310046 Fix non-Darwin build 2016-01-05 13:31:15 +01:00
Tuomas Tynkkynen
f770b9e6c5 libstore: mmap() returns MAP_FAILED, not NULL on failure 2016-01-05 13:26:35 +01:00
Eelco Dolstra
71a93a5f0e Don't allow sandbox profile except in relaxed mode
This makes Darwin consistent with Linux: Nix expressions can't break
out of the sandbox unless relaxed sandbox mode is enabled.

For the normal sandbox mode this will require fixing #759 however.
2016-01-04 20:01:13 +01:00
Eelco Dolstra
77ad443bd1 ~PathLocks(): Handle exceptions
Otherwise, since the call to write a "d" character to the lock file
can fail with ENOSPC, we can get an unhandled exception resulting in a
call to terminate().
2016-01-04 11:34:36 +01:00
Eelco Dolstra
b8258a4475 Fix regression in passAsFile
Caused by 8063fc497a. If tmpDir !=
tmpDirInSandbox (typically when there are multiple concurrent builds
with the same name), the *Path attribute would not point to an
existing file. This caused Nixpkgs' writeTextFile to write an empty
file. In particular this showed up as hanging VM builds (because it
would run an empty run-nixos-vm script and then wait for it to finish
booting).
2015-12-29 15:28:20 +01:00
Eelco Dolstra
52120123a5 Handle /tmp being a symlink
Hopefully fixes Darwin sandbox regression introduced in
8063fc497a.
2015-12-22 17:16:17 +01:00
Eelco Dolstra
f696af0fab Fix bad error message in Darwin chroots 2015-12-22 17:05:29 +01:00
Eelco Dolstra
8f67325a7c Build sandbox support etc. unconditionally on Linux
Also, use "#if __APPLE__" instead of "#if SANDBOX_ENABLED" to prevent
ambiguity.
2015-12-10 11:47:17 +01:00
Bjørn Forsman
65bd82d42a Clarify error message for hash mismatches (again)
This is arguably nitpicky, but I think this new formulation is even
clearer. My thinking is that it's easier to comprehend when the
calculated hash value is displayed close to the output path. (I think it
is somewhat similar to eliminating double negatives in logic
statements.)

The formulation is inspired / copied from the OpenEmbedded build tool,
bitbake.
2015-12-08 19:50:25 +01:00
Ludovic Courtès
d1e3bf01bc daemon: Add 'buildMode' parameter to 'buildPaths' RPC 2015-12-02 18:14:49 +01:00
Eelco Dolstra
8063fc497a Use deterministic $TMPDIR in sandbox
Rather than using $<host-TMPDIR>/nix-build-<drvname>-<number>, the
temporary directory is now always /tmp/nix-build-<drvname>-0. This
improves bitwise-exact reproducibility for builds that store $TMPDIR
in their build output. (Of course, those should still be fixed...)
2015-12-02 15:04:00 +01:00
Eelco Dolstra
0ab4d905e7 Merge branch 'p/sandbox-rename-minimal' of https://github.com/vcunat/nix 2015-11-25 14:53:42 +01:00
Eelco Dolstra
a0f0733413 Fix build failure introduced by #704
Also, make the FreeBSD checks conditional on FreeBSD.
2015-11-25 14:41:19 +01:00
Eelco Dolstra
cad40adce5 Merge pull request #704 from ysangkok/freebsd-support
FreeBSD support with knowledge about Linux emulation
2015-11-24 19:24:21 +01:00
Jude Taylor
279fa8f618 reintroduce host deps in tandem with sandbox profiles 2015-11-21 15:57:06 -08:00
Shea Levy
e0bd114e09 Revert "remove sandbox-defaults.sb"
As discussed in NixOS/nixpkgs#11001, we still need some of the old
sandbox mechanism.

This reverts commit d760c2638c.
2015-11-21 16:40:24 -05:00
Jude Taylor
b9b7bb1806 re-fix permissions for GHC 2015-11-19 16:06:21 -08:00
Shea Levy
5deb7fbdfb Merge branch 'sandbox-profiles' of git://github.com/pikajude/nix
Temporarily allow derivations to describe their full sandbox profile.
This will be eventually scaled back to a more secure setup, see the
discussion at #695
2015-11-19 17:44:11 -05:00
Peter Simons
6ad10591ce src/libstore/build.cc: clarify error message for hash mismatches
Nix reports a hash mismatch saying:

  output path ‘foo’ should have sha256 hash ‘abc’, instead has ‘xyz’

That message is slightly ambiguous and some people read that statement
to mean the exact opposite of what it is supposed to mean. After this
patch, the message will be:

  Nix expects output path ‘foo’ to have sha256 hash ‘abc’, instead it has ‘xyz’
2015-11-19 12:42:37 +01:00
janus
8a74a125bc FreeBSD can build Linux 32-bit binaries 2015-11-17 14:16:08 +00:00
Shea Levy
58d2fac91d AutoDelete: Add default constructor with deletion disabled 2015-11-16 05:53:10 -05:00
Shea Levy
4390142315 Use AutoDelete for sandbox profile file 2015-11-15 06:08:50 -05:00
Jude Taylor
bd09a4c967 simplify build.cc using modern C++ features 2015-11-14 14:11:03 -08:00
Jude Taylor
4876bb012e simplify build permissions 2015-11-14 14:11:03 -08:00
Jude Taylor
d760c2638c remove sandbox-defaults.sb 2015-11-14 14:11:03 -08:00
Jude Taylor
8a7f0dfd68 use per-derivation sandbox profiles 2015-11-14 14:10:43 -08:00
Vladimír Čunát
b39622a487 rename chroot to sandbox (fixes #656, close #682)
- rename options but leav old names as lower-priority aliases,
  also "-dirs" -> "-paths" to get closer to the meaning
- update docs to reflect the new names (old aliases are not documented),
  including a new file with release notes
- tests need an update after corresponding changes to nixpkgs
- __noChroot is left as it is (after discussion on the PR)
2015-11-10 22:32:51 +01:00
Eelco Dolstra
8fdd156a65 Add option to verify build determinism
Passing "--option build-repeat <N>" will cause every build to be
repeated N times. If the build output differs between any round, the
build is rejected, and the output paths are not registered as
valid. This is primarily useful to verify build determinism. (We
already had a --check option to repeat a previously succeeded
build. However, with --check, non-deterministic builds are registered
in the DB. Preventing that is useful for Hydra to ensure that
non-deterministic builds don't end up getting published at all.)
2015-11-09 23:16:24 +01:00
Eelco Dolstra
96c2ebf004 Revert "Allow using /bin and /usr/bin as impure prefixes on non-darwin by default"
This reverts commit 79ca503332. Ouch,
never noticed this. We definitely don't want to allow builds to have
arbitrary access to /bin and /usr/bin, because then they can (for
instance) bring in a bunch of setuid programs. Also, we shouldn't be
encouraging the use of impurities in the default configuration.
2015-11-09 21:28:38 +01:00
Eelco Dolstra
4384bbd2e1 optimizePath(): Detect some .links corruption
If automatic store optimisation is enabled, and a hard-linked file in
the store gets corrupted, then the corresponding .links entry will
also be corrupted. In that case, trying to repair with --repair or
--repair-path won't work, because the new "good" file will be replaced
by a hard link to the corrupted file. We can catch most of these cases
by doing a sanity-check on the file sizes.
2015-11-09 21:28:25 +01:00
Jude Taylor
c10ef2d134 fix syntax error 2015-11-03 09:03:32 -08:00
Jude Taylor
3b52e54e31 darwin: allow reading system locale and zoneinfo 2015-11-03 08:43:09 -08:00
Jude Taylor
d6bd51ffbe allow reading ICU data 2015-10-31 02:52:42 -07:00
Jude Taylor
95c66cc950 add special devices to sandbox-defaults 2015-10-30 20:32:11 -07:00
Eelco Dolstra
1f735a3440 <nix/fetchurl.nix>: Support xz-compressed NARs 2015-10-30 12:34:30 +01:00
Eelco Dolstra
dae5dc7ade <nix/fetchurl.nix>: Support downloading and unpacking NARs
This removes the need to have multiple downloads in the stdenv
bootstrap process (like a separate busybox binary for Linux, or
curl/mkdir/sh/bzip2 for Darwin). Now all those files can be combined
into a single NAR.
2015-10-30 11:27:47 +01:00
Eelco Dolstra
5c28943e8f int2String() -> std::to_string() 2015-10-29 13:26:55 +01:00
Jude Taylor
9155fbd386 use nixDataDir instead of appending /share to PREFIX 2015-10-21 14:44:44 -07:00
Jude Taylor
992cda1b11 clarifying comment 2015-10-21 14:39:16 -07:00
Jude Taylor
76f3ba42fd move preBuildHook defaulting to globals.cc 2015-10-21 14:39:08 -07:00
Jude Taylor
2dc8e19a17 restore old DEFAULT_ALLOWED_IMPURE_PREFIXES 2015-10-21 14:38:49 -07:00
Jude Taylor
ff6953cb03 Add resolve-system-dependencies.pl 2015-10-21 12:38:52 -07:00
Jude Taylor
15c7a79ed3 remove usr paths from allowed inputs 2015-10-21 12:38:52 -07:00
Jude Taylor
5bf4085000 allow access to SystemVersion for python builders 2015-10-21 12:38:52 -07:00
Jude Taylor
6a8cee19b3 fix line reading in preBuildHook 2015-10-21 12:38:52 -07:00
Jude Taylor
1fc0fe1baf remove sandbox defaults into a new file 2015-10-21 12:38:52 -07:00
Jude Taylor
d4cac051f7 restore allowed impure prefixes 2015-10-21 12:38:52 -07:00
Jude Taylor
4b8c71b4ef remove an unneeded default impure-dep 2015-10-21 12:38:52 -07:00
Jude Taylor
e770f941d6 make sandbox builds more permissive 2015-10-21 12:38:52 -07:00
Jude Taylor
1ea2aa9e99 add a few more permissions 2015-10-21 12:38:52 -07:00
Eelco Dolstra
f5b32ff47d Allow builtin fetchurl regardless of the derivation's system attribute 2015-10-21 15:14:42 +02:00
Eelco Dolstra
01615b5f63 Show progress indicator for builtin fetchurl 2015-10-21 15:14:42 +02:00
Eelco Dolstra
5db358d4d7 Disable TLS verification for builtin fetchurl
This makes it consistent with the Nixpkgs fetchurl and makes it work
in chroots. We don't need verification because the hash of the result
is checked anyway.
2015-10-21 15:14:42 +02:00
Eelco Dolstra
357d31b339 Fix segfault in builtin fetchurl
The stack allocated for the builder was way too small (32 KB). This is
sufficient for normal derivations, because they just do some setup and
then exec() the actual builder. But for the fetchurl builtin
derivation it's not enough. Also, allocating the stack on the caller's
stack was fishy business.
2015-10-21 15:14:42 +02:00
Tuomas Tynkkynen
2075ec83e3 Allow building ARMv6 stuff on ARMv7
This allows building a Raspberry Pi image on modern, faster boards.
2015-10-08 13:32:28 +02:00
Eelco Dolstra
14f7e1fc42 Revert to CURLOPT_PROGRESSFUNCTION
CURLOPT_XFERINFOFUNCTION isn't widely supported yet.

http://hydra.nixos.org/build/26679495
2015-10-08 13:15:21 +02:00
Eelco Dolstra
21ecd106ed Show progress during downloads 2015-10-07 17:31:50 +02:00
Manuel Jacob
cbc177ca01 Use pkg-config-provided LDFLAGS for libsqlite3 and libcurl.
Previously, pkg-config was already queried for libsqlite3's and
libcurl's link flags.  However they were not used, but hardcoded
instead.  This commit replaces the hardcoded LDFLAGS by the ones
provided by pkg-config in a similar pattern as already used for
libsodium.
2015-10-06 22:28:30 +02:00
Eelco Dolstra
1abda8e173 nix-store --serve: Implement log size limit 2015-10-06 17:33:30 +02:00
Eelco Dolstra
4ba6bc184c Shut up clang warnings 2015-09-18 01:22:35 +02:00
Eelco Dolstra
71a5161365 Implement buildDerivation() via the daemon 2015-09-03 12:56:59 +02:00
Eelco Dolstra
73332fd342 Filter build-chroot-dirs entries that conflict with derivation outputs
Fixes https://github.com/NixOS/nixpkgs/issues/9504.

Note that this means we may have a non-functional /bin/sh in the
chroot while rebuilding Bash or one of its dependencies. Ideally those
packages don't rely on /bin/sh though.
2015-09-02 14:54:12 +02:00
Eelco Dolstra
d711fe8f5e Remove unused variable 2015-09-02 14:46:04 +02:00
Eelco Dolstra
e12cf82782 Prevent .chroot from being GC'ed when using LocalStore::buildDerivation()
Fixes #616.
2015-08-24 11:13:31 +02:00
Eelco Dolstra
eadb86f447 nix-collect-garbage: Revive --max-freed
Fixes #609.
2015-08-21 13:57:53 +02:00
Manolis Ragkousis
26221e44eb Remove unneeded HAVE_UNSHARE.
* src/libstore/build.cc (CHROOT_ENABLED): Remove HAVE_UNSHARE.
2015-08-19 14:11:15 +03:00
Eelco Dolstra
34dfbd9394 Don't include <iostream> before config.h
This breaks the build on 32-bit systems.

http://hydra.nixos.org/build/24373658
2015-08-04 11:12:31 +02:00
Eelco Dolstra
30d19a2bdc Handle debug messages from runChild()
Turns out that "nix-build -vvv" with chroots enabled has been broken
for some time, because some debug message got interpreted as an error.
2015-08-03 18:04:32 +02:00
Eelco Dolstra
f3dda728a4 Remove unnecessary parentheses 2015-07-26 12:07:14 +02:00
Eelco Dolstra
16c9935fa9 Fix fetchurl of executable file
Pointed out by @cstrahan, thanks!
2015-07-23 22:25:04 +02:00
Eelco Dolstra
1993b10d11 Fix Darwin build
Turns out getgrouplist() is not POSIX.

http://hydra.nixos.org/build/23881243
2015-07-21 14:45:24 +02:00
Eelco Dolstra
0a2bee307b Make <nix/fetchurl.nix> a builtin builder
This ensures that 1) the derivation doesn't change when Nix changes;
2) the derivation closure doesn't contain Nix and its dependencies; 3)
we don't have to rely on ugly chroot hacks.
2015-07-20 04:38:46 +02:00
Eelco Dolstra
eda2f36c2a Provide more detailed info about build status to hydra-queue-runner
In particular, hydra-queue-runner can now distinguish between remote
build / substitution / already-valid. For instance, if a path already
existed on the remote side, we don't want to store a log file.
2015-07-20 03:20:03 +02:00
Eelco Dolstra
ccf31dbc25 nix-copy-closure: Add -v flag
And make exportPath() less spammy by default.
2015-07-20 01:52:07 +02:00
Eelco Dolstra
b3491c781c More cleanup 2015-07-20 01:16:16 +02:00
Eelco Dolstra
6bd2c7bb38 OCD: foreach -> C++11 ranged for 2015-07-17 20:13:56 +02:00
Eelco Dolstra
1511aa9f48 Allow remote builds without sending the derivation closure
Previously, to build a derivation remotely, we had to copy the entire
closure of the .drv file to the remote machine, even though we only
need the top-level derivation. This is very wasteful: the closure can
contain thousands of store paths, and in some Hydra use cases, include
source paths that are very large (e.g. Git/Mercurial checkouts).

So now there is a new operation, StoreAPI::buildDerivation(), that
performs a build from an in-memory representation of a derivation
(BasicDerivation) rather than from a on-disk .drv file. The only files
that need to be in the Nix store are the sources of the derivation
(drv.inputSrcs), and the needed output paths of the dependencies (as
described by drv.inputDrvs). "nix-store --serve" exposes this
interface.

Note that this is a privileged operation, because you can construct a
derivation that builds any store path whatsoever. Fixing this will
require changing the hashing scheme (i.e., the output paths should be
computed from the other fields in BasicDerivation, allowing them to be
verified without access to other derivations). However, this would be
quite nice because it would allow .drv-free building (e.g. "nix-env
-i" wouldn't have to write any .drv files to disk).

Fixes #173.
2015-07-17 17:57:40 +02:00
Eelco Dolstra
dd48c06bb6 Typo 2015-07-02 00:30:16 +02:00
Ludovic Courtès
9aed117395 Preserve supplementary groups of build users
The following patch is an attempt to address this bug (see
<http://bugs.gnu.org/18994>) by preserving the supplementary groups of
build users in the build environment.

In practice, I would expect that supplementary groups would contain only
one or two groups: the build users group, and possibly the “kvm” group.

[Changed &at(0) to data() and removed tabs - Eelco]
2015-07-01 14:57:48 +02:00
Eelco Dolstra
ff4de4cb27 GC: Handle ENOSPC creating/moving to the trash directory
Issue #564.
2015-06-30 21:41:26 +02:00
Eelco Dolstra
2bc9c84327 Use posix_fallocate to create /nix/var/nix/db/reserved 2015-06-22 15:54:55 +02:00
Eelco Dolstra
ba63ec6f39 Make /nix/var/nix/db/reserved bigger
Issue #564.
2015-06-22 15:47:40 +02:00
Eelco Dolstra
b65875f859 Export outputPaths function
This is useful for the new hydra-queue-runner.
2015-06-10 16:17:06 +02:00
Eelco Dolstra
7c4501886d Use std::vector::data() 2015-06-09 10:54:46 +02:00
Eelco Dolstra
b64988bb35 Allow substitutes for builds that have preferLocalBuild set
Not substituting builds with "preferLocalBuild = true" was a bad idea,
because it didn't take the cost of dependencies into account. For
instance, if we can't substitute a fetchgit call, then we have to
download/build git and all its dependencies.

Partially reverts 5558652709 and adds a
new derivation attribute "allowSubstitutes" to specify whether a
derivation may be substituted.
2015-06-04 16:30:22 +02:00
Eelco Dolstra
07d7e7df84 Chown files created for passAsFile
Nixpkgs' writeTextAsFile does this:

  mv "$textPath" "$n"

Since $textPath was owned by root, if $textPath is on the same
filesystem as $n, $n will be owned as root. As a result, the build
result was rejected as having suspicious ownership.

http://hydra.nixos.org/build/22836807
2015-06-04 14:07:43 +02:00
Ludovic Courtès
b755752f76 Add a ‘verifyStore’ RPC
Hello!

The patch below adds a ‘verifyStore’ RPC with the same signature as the
current LocalStore::verifyStore method.

Thanks,
Ludo’.

>From aef46c03ca77eb6344f4892672eb6d9d06432041 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ludovic=20Court=C3=A8s?= <ludo@gnu.org>
Date: Mon, 1 Jun 2015 23:17:10 +0200
Subject: [PATCH] Add a 'verifyStore' remote procedure call.
2015-06-02 02:14:24 +02:00
Eelco Dolstra
4ca5a9dcfd nix-collect-garbage: Don't call nix-env
Also, make sure --delete-older-than doesn't delete the current
generation.
2015-05-21 16:28:30 +02:00
Eelco Dolstra
22b1a8d43f Move profiles.{cc,hh} to libstore 2015-05-21 15:42:59 +02:00
Rok Garbas
000de699e9 cygwin: explicitly include required c headers 2015-05-13 09:37:12 +02:00
Shea Levy
71083f9e5e Don't try to map /bin/sh to a store path on non-Linux 2015-05-12 16:36:15 -04:00
Shea Levy
4d652875bd Add the pre-build hook.
This hook can be used to set system-specific per-derivation build
settings that don't fit into the derivation model and are too complex or
volatile to be hard-coded into nix. Currently, the pre-build hook can
only add chroot dirs/files through the interface, but it also has full
access to the chroot root.

The specific use case for this is systems where the operating system ABI
is more complex than just the kernel-support system calls. For example,
on OS X there is a set of system-provided frameworks that can reliably
be accessed by any program linked to them, no matter the version the
program is running on. Unfortunately, those frameworks do not
necessarily live in the same locations on each version of OS X, nor do
their dependencies, and thus nix needs to know the specific version of
OS X currently running in order to make those frameworks available. The
pre-build hook is a perfect mechanism for doing just that.
2015-04-18 16:56:02 -04:00
Shea Levy
fd6774e285 Revert "Add the pre-build hook."
Going to reimplement differently.

This reverts commit 1e4a4a2e9f.
2015-04-18 14:59:58 -04:00
Eelco Dolstra
035aeb9547 Fix using restricted mode with chroots 2015-04-16 18:46:17 +02:00
Shea Levy
1e4a4a2e9f Add the pre-build hook.
This hook can be used to set system specific per-derivation build
settings that don't fit into the derivation model and are too complex or
volatile to be hard-coded into nix. Currently, the pre-build hook can
only add chroot dirs/files.

The specific use case for this is systems where the operating system ABI
is more complex than just the kernel-supported system calls. For
example, on OS X there is a set of system-provided frameworks that can
reliably be accessed by any program linked to them, no matter the
version the program is running on. Unfortunately, those frameworks do
not necessarily live in the same locations on each version of OS X, nor
do their dependencies, and thus nix needs to know the specific version
of OS X currently running in order to make those frameworks available.
The pre-build hook is a perfect mechanism for doing just that.
2015-04-12 12:56:38 -04:00
Eelco Dolstra
c1f04fae35 Implement a TTL on cached fetchurl/fetchTarball results
This is because we don't want to do HTTP requests on every evaluation,
even though we can prevent a full redownload via the cached ETag. The
default is one hour.
2015-04-09 11:55:36 +02:00
Eelco Dolstra
1711679ea5 Revert /nix/store permission back to 01775
This broke NixOS VM tests.

Mostly reverts 27b7b94923,
5ce50cd99e,
afa433e58c.
2015-04-07 13:21:26 +02:00
Eelco Dolstra
afa433e58c Chroot builds: Provide world-readable /nix/store
This was causing NixOS VM tests to fail mysteriously since
5ce50cd99e. Nscd could (sometimes) no
longer read /etc/hosts:

open("/etc/hosts", O_RDONLY|O_CLOEXEC) = -1 EACCES (Permission denied)

Probably there was some wacky interaction between the guest kernel and
the 9pfs implementation in QEMU.
2015-04-02 16:59:40 +02:00
Eelco Dolstra
7ea6ecf855 addToStore(): Take explicit name argument 2015-03-25 17:06:12 +01:00