it's just a uri and some headers now. those can be function arguments
with no loss of clarity. *actual* additional arguments, for example a
TLS context with additional certificates, could be added on a new and
improved FileTransfer class that carries not just a backend reference
but some real, visible context for its transfers. curl not being very
multi-threading-friendly when using multi handles will make sharing a
bit hard anyway once we drop the single global download worker thread
Change-Id: Id2112c95cbd118c6d920488f38d272d7da926460
we don't even need this outside of tests. maybe we should not do
automatic retries at this level at all and use retrying wrappers
instead? at some point we may have to do this, but not just yet.
Change-Id: If0088aa55215be81f1770c25b3bb1b5268c65cf8
this is fragile, slow as fuck, breaks constantly under high concurrency,
and completely unnecessary since ssh bypasses the stdio file descriptors
*anyway*. we do still check that we see ssh messages to ensure that none
of our subprocess handling messes with ssh's /dev/tty, but that's it now
Change-Id: Ib8e31e1999f813d07a27efc63a9d3454a9e4fcdd
* changes:
treewide: make more settings conditionally available
libstore/build: only send overridden settings to the build hook
treewide: consistently mark overridden settings as such
The old behavior results in lots of concatenations happening for no good
reason and is an artifact of the technical limitations of the old parser
(combined with some lack of care for such details).
Change-Id: I0d78d6220ca6aeaa10bc437e48e08bf7922e0bb3
This is only a minor semantical distinction, but we should be able to
properly test it, and the parser tests rely on show for that.
Change-Id: I25e868cf9544e30cdff17deb5fd50a434e0f367e
Only overridden settings are sent to the daemon, and we're going to do the same
for the build hook to. It needs to be ensured that overridden settings are in
fact consistently marked as such, so that they actually get sent.
Change-Id: I7cd58d925702f86cf2c35ad121eb191ceb62a355
This was always a terrible idea independently of whether it crashes.
Stop doing it!
This commit was verified by running nix-shell on a trivial derivation
with --debug --verbose to get the vomit-level output of the shell rc
file and then diffing it before/after this change. I have reasonable
confidence it did not regress anything, though this code is genuinely
really hard to follow (which is a second reason that I split it into two
fmt calls).
Fixes: #533
Change-Id: I8e11ddbece2b12749fda13efe0b587a71b00bfe5
This change feels kind of gross and reveals a fair bit about the
disorganization of our tests, but I think it makes parts of it a bit
better.
Change-Id: Idb8d9a00cbd75d5c156678c6b408b42b59d5e4d7
A better fix than in 104448e75d, hence a
revert + the fix.
It turns out that this commit has the side-effect that when having e.g.
`StrictHostKeyChecking=accept-new` for a remote builder, the warnings à la
Warning: Permanently added 'builder' (ED25519) to the list of known hosts.
actually end up in the derivation's log whereas hostkey verification
errors don't, but only in the stderr of the `nix-build` invocation
(which was the motivation for the patch).
This change writes the stderr from the build-hook to
* the daemon's stderr, so that the SSH errors appear in the journal
(which was the case before 104448e75d)
* the client's stderr, as a log message
* NOT to the drv log (this is handled via `handleJSONLogMessage`)
I tried to fix the issue for legacy-ssh as well, but failed and
ultimately decided to not bother.
I know that we'll sooner or later replace the entire component, however
this is the part of the patch I have working for a while, so I figured I
might still submit it for the time being.
Change-Id: I21ca1aa0d8ae281d2eacddf26e0aa825272707e5
When the git default branch is not set to master the installcheck
test suite fails. This patch adjusts the test setup scripts to
ignore the system and user git config files.
GIT_CONFIG_SYSTEM is set to /dev/null to ignore /etc/gitconfig
GIT_CONFIG_GLOBAL is not set because the global config files
are loaded from $HOME or $XDG_CONFIG_HOME which we already
reset.
git documentation: https://git-scm.com/docs/git#Documentation/git.txt-codeGITCONFIGGLOBALcode
Change-Id: Ie73bbed1db9419c9885b9d57e4edb7a4047d5cce
There is absolutely no good reason these should show up in NARs besides
misconfigured systems and as long as the case hack exists, unpacking
such a NAR will cause its repacking to be wrong on systems with case
hack enabled.
This should not have any security impact on Lix to fix, but it was one
of the vectors for CVE-2024-45593:
https://github.com/NixOS/nix/security/advisories/GHSA-h4vv-h3jq-v493
Change-Id: I85b6075aacc069ee7039240b0f525804a2d8edcb
This is capped at 12 because 3.7 seconds of startup is painful enough
and 5.5 seconds with 24 was more annoying.
Change-Id: I327db40fd98deaa5330cd9cf6de99fb07b2c1cb0
This also rewrites a lot of the command handling in the fixtures
library, since we want to more precisely control which way that the nix
store is set up in the tests, rather than the previous method of
renaming /nix/store to some temp dir (which allows builds but does not
allow any /nix/store paths or stability across runs, which is a
significant issue for snapshot testing).
It uses a builder to reduce the amount of state carelessly thrown
around.
The evil NARs are inspired by CVE-2024-45593
(https://github.com/NixOS/nix/security/advisories/GHSA-h4vv-h3jq-v493).
No bugs were found in this endeavor.
Change-Id: Iee41b055fa96529c5a3c761f680ed1d0667ba5da
I am tired of bad shell scripts, let me write bad python quickly
instead. It's definitely, $100%, better.
This is not planned as an immediate replacement of the old test suite,
but we::jade would not oppose tests getting ported.
What is here is a mere starting point and there is a lot more
functionality that we need.
Fixes: #488
Change-Id: If762efce69030bb667491b263b874c36024bf7b6
There was a bug report about a potential call to `memcpy` with a null
pointer which is not reproducible:
#492
This occurred in `src/libstore/filetransfer.cc` in `InnerSource::read`.
To ensure that this doesn't happen, an early return is added before
calling `memcpy` if the length of the data to be copied is 0.
This change also adds a test that ensures that when `InnerSource::read`
is called with an empty file, it throws an `EndOfFile` exception.
Change-Id: Ia18149bee9a3488576c864f28475a3a0c9eadfbb
with async runtime scheduling we can no longer guarantee exact error
counts for builds that do not set keepGoing. the old behavior can be
recovered with a number of hacks that affect scheduling, but none of
those are very easy to follow now advisable. exact error counts will
like not be needed for almost all uses except tests, and *those* had
better check the actual messages rather than how many they got. more
messages can even help to avoid unnecessary rebuilds for most users.
Change-Id: I1c9aa7a401227dcaf2e19975b8cb83c5d4f85d64
The test suite can load the global configuration files under certain
circumstances, and, though we would really rather it didn't ever do that
at all, we should at least break the mechanism.
Fixes: #474
Change-Id: Ib27cb43dd5dfaa70ac491c395b5ba308fd7bd289
This moves the "legacy"/"nix2" commands under a new `src/legacy/`
directory, instead of being scattered around in a bunch of different
directories.
A new `liblegacy` build target is defined, and the `nix` binary is
linked against it.
Then, `RegisterLegacyCommand` is replaced with `LegacyCommand::add`
calls in functions like `registerNixCollectGarbage()`. These
registration functions are called explicitly in `src/nix/main.cc`.
See: #359
Change-Id: Id450ffc3f793374907599cfcc121863b792aac1a
this was a debugging aid from day one that should not have any impact on
build semantics, and if it *does* have an impact on build semantics then
build semantics are seriously broken. keeping the order imposed by these
keys will be impossible once we let a real event loop schedule our jobs.
Change-Id: I5c313324e1f213ab6453d82f41ae5e59de809a5b
the new event loop could very occasionally notice that a dependency of
some goal has failed, process the failure, cause the depending goal to
fail accordingly, and in the doing of the latter two steps let further
dependencies that previously have not been reported as failed do their
reporting anyway. in such cases a goal could fail with "1 dependencies
failed", but more than one dependency failure message was shown. we'll
now report the correct number of failed dependency goals in all cases.
Change-Id: I5aa95dcb2db4de4fd5fee8acbf5db833531d81a8
Since fb38459d6e, each `ref` is appended
with `refs/heads` unless it starts with `refs/` already. This regressed
two use-cases that worked fine before:
* Specifying a commit hash as `ref`: now, if `ref` looks like a commit
hash it will be directly passed to `git fetch`.
* Specifying a tag without `refs/tags` as prefix: now, the fetcher prepends
`refs/*` to a ref that doesn't start with `refs/` and doesn't look
like a commit hash. That way, both a branch and a tag specified in
`ref` can be fetched.
The order of preference in git is
* file in `refs/` (e.g. `HEAD`)
* file in `refs/tags/`
* file in `refs/heads` (i.e. a branch)
After fetching `refs/*`, ref is resolved the same way as git does.
Change-Id: Idd49b97cbdc8c6fdc8faa5a48bef3dec25e4ccc3
When `nix fmt` is called without an argument, Nix appends the "." argument before calling the formatter. The comment in the code is:
> Format the current flake out of the box
This also happens when formatting sub-folders.
This means that the formatter is now unable to distinguish, as an interface, whether the "." argument is coming from the flake or the user's intent to format the current folder. This decision should be up to the formatter.
Treefmt, for example, will automatically look up the project's root and format all the files. This is the desired behaviour. But because the "." argument is passed, it cannot function as expected.
Upstream-PR: https://github.com/nixos/nix/pull/11438
Change-Id: I60fb6b3ed4ec1b24f81b5f0d76c0be98470817ce
like kj::joinPromisesFailFast this allows waiting for the results of
multiple promises at once, but unlike it not all input promises must
be complete (or any of them failed) for results to become available.
Change-Id: I0e4a37e7bd90651d56b33d0bc5afbadc56cde70c
like a normal semaphore, but with awaitable acquire actions. this is
primarily intended as an intermediate concurrency limiting device in
the Worker code, but it may find other uses over time. we do not use
std::counting_semaphore as a base because the counter of that is not
inspectable as will be needed for Worker. we also do not need atomic
operations for cross-thread consistency since we don't have multiple
threads (thanks to kj event loops being confined to a single thread)
Change-Id: Ie2bcb107f3a2c0185138330f7cbba4cec6cbdd95
Without this, verifying TLS certificates would fail on macOS, as well
as any system that doesn't have a certificate file at /etc/ssl/certs/ca-certificates.crt,
which includes e.g. Fedora.
Change-Id: Iaa2e0e9db3747645b5482c82e3e0e4e8f229f5f9
This is better for privacy and to avoid leaking netrc credentials in a
MITM attack, but also the assumption that we check the hash no longer
holds in some cases (in particular for impure derivations).
Partially reverts 5db358d4d7.
(cherry picked from commit c04bc17a5a0fdcb725a11ef6541f94730112e7b6)
(cherry picked from commit f2f47fa725fc87bfb536de171a2ea81f2789c9fb)
(cherry picked from commit 7b39cd631e0d3c3d238015c6f450c59bbc9cbc5b)
Upstream-PR: https://github.com/NixOS/nix/pull/11585
Change-Id: Ia973420f6098113da05a594d48394ce1fe41fbb9
These stack traces kind of suck for the reasons mentioned on the
CppTrace page here (no symbols for inline functions is a major one):
https://github.com/jeremy-rifkin/cpptrace
I would consider using CppTrace if it were packaged, but to be honest, I
think that the more reasonable option is actually to move entirely to
out-of-process crash handling and symbolization.
The reason for this is that if you want to generate anything of
substance on SIGSEGV or really any deadly signal, you are stuck in
async-signal-safe land, which is not a place to be trying to run a
symbolizer. LLVM does it anyway, probably carefully, and chromium *can*
do it on debug builds but in general uses crashpad:
https://source.chromium.org/chromium/chromium/src/+/main:base/debug/stack_trace_posix.cc;l=974;drc=82dff63dbf9db05e9274e11d9128af7b9f51ceaa;bpv=1;bpt=1
However, some stack traces are better than *no* stack traces when we get
mystery exceptions falling out the bottom of the program. I've also
promoted the path for "mystery exceptions falling out the bottom of the
program" to hard crash and generate a core dump because although there's
been some months since the last one of these, these are nonetheless
always *atrociously* diagnosed.
We can't improve the crash handling further until either we use Crashpad
(which involves more C++ deps, no thanks) or we put in the ostensibly
work in progress Rust minidump infrastructure, in which case we need to
finish full support for Rust in libutil first.
Sample report:
Lix crashed. This is a bug. We would appreciate if you report it at https://git.lix.systems/lix-project/lix/issues with the following information included:
Exception: std::runtime_error: lol
Stack trace:
0# nix::printStackTrace() in /home/jade/lix/lix3/build/src/nix/../libutil/liblixutil.so
1# 0x000073C9862331F2 in /home/jade/lix/lix3/build/src/nix/../libmain/liblixmain.so
2# 0x000073C985F2E21A in /nix/store/p44qan69linp3ii0xrviypsw2j4qdcp2-gcc-13.2.0-lib/lib/libstdc++.so.6
3# 0x000073C985F2E285 in /nix/store/p44qan69linp3ii0xrviypsw2j4qdcp2-gcc-13.2.0-lib/lib/libstdc++.so.6
4# nix::handleExceptions(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::function<void ()>) in /home/jade/lix/lix3/build/src/nix/../libmain/liblixmain.so
5# 0x00005CF65B6B048B in /home/jade/lix/lix3/build/src/nix/nix
6# 0x000073C985C8810E in /nix/store/dbcw19dshdwnxdv5q2g6wldj6syyvq7l-glibc-2.39-52/lib/libc.so.6
7# __libc_start_main in /nix/store/dbcw19dshdwnxdv5q2g6wldj6syyvq7l-glibc-2.39-52/lib/libc.so.6
8# 0x00005CF65B610335 in /home/jade/lix/lix3/build/src/nix/nix
Change-Id: I1a9f6d349b617fd7145a37159b78ecb9382cb4e9
This caused an infinite loop before since it would just keep asking the
underlying source for more data.
In practice this happened because an HTTP server served a
response to a HEAD request (for which curl will not retrieve any body or
call our write callback function) with Content-Encoding: br, leading to
decompressing nothing at all and going into an infinite loop.
This adds a test to make sure none of our compression methods do that
again, as well as just patching the HTTP client to never feed empty data
into a compression algorithm (since they absolutely have the right to
throw CompressionError on unexpectedly-short streams!).
Reported on Matrix: https://matrix.to/#/!lymvtcwDJ7ZA9Npq:lix.systems/$8BWQR_zKxCQDJ40C5NnDo4bQPId3pZ_aoDj2ANP7Itc?via=lix.systems&via=matrix.org&via=tchncs.de
Change-Id: I027566e280f0f569fdb8df40e5ecbf46c211dad1
This test suite was in desperate need of using the parameterization
available with gtest, and was a bunch of useless duplicated code. At
least now it's not duplicated code, though it still probably should be
more full of property tests.
Change-Id: Ia8ccee7ef4f02b2fa40417b79aa8c8f0626ea479
eldritch horrors
raito
alois31
jade
piegames
V.
Maximilian Bosch
Lulu
Rebecca Turner
Jonas Chevalier
puck