This was caused by derivations with 'allowSubstitutes = false'. Such
derivations will be built locally. However, if there is another
SubstitionGoal that has the output of the first derivation in its
closure, then the path will be simultaneously built and substituted.
There was a check to catch this situation (via pathIsLockedByMe()),
but it no longer worked reliably because substitutions are now done in
another thread. (Thus the comment 'It can't happen between here and
the lockPaths() call below because we're not allowing multi-threading'
was no longer valid.)
The fix is to handle the path already being locked in both
SubstitutionGoal and DerivationGoal.
Nix can now automatically run the garbage collector during builds or
while adding paths to the store. The option "min-free = <bytes>"
specifies that Nix should run the garbage collector whenever free
space in the Nix store drops below <bytes>. It will then delete
garbage until "max-free" bytes are available.
Garbage collection during builds is asynchronous; running builds are
not paused and new builds are not blocked. However, there also is a
synchronous GC run prior to the first build/substitution.
Currently, no old GC roots are deleted (as in "nix-collect-garbage
-d").
Since file locks are per-process rather than per-file-descriptor, the
garbage collector would always acquire a lock on its own temproots
file and conclude that it's stale.
Functions like copyClosure() had 3 bool arguments, which creates a
severe risk of mixing up arguments.
Also, implement copyClosure() using copyPaths().
Cygwin sqlite3 is patched to call SetDllDirectory("/usr/bin") on init, which
affects the current process and is inherited by child processes. It causes
DLLs to be loaded from /usr/bin/ before $PATH, which breaks all sorts of
things. A typical failures would be header/lib version mismatches (e.g.
openssl when running checkPhase on openssh). We'll just set it back to the
default value.
Note that this is a problem with the cygwin version of sqlite3 (currently
3.18.0). nixpkgs doesn't have the problematic patch.
In particular, UF_IMMUTABLE (uchg) needs to be cleared to allow the
path to be garbage-collected or optimised.
See https://github.com/NixOS/nixpkgs/issues/25819.
+ the file from being garbage-collected.
EAs/ACLs are not part of the NAR canonicalisation. Worse, setting an
ACL allows a builder to create writable files in the Nix store. So get
rid of them.
Closes#185.
It now means "paths that were built locally". It no longer includes
paths that were added locally. For those we don't need info.ultimate,
since we have the content-addressability assertion (info.ca).
The typical use is to inherit Config and add Setting<T> members:
class MyClass : private Config
{
Setting<int> foo{this, 123, "foo", "the number of foos to use"};
Setting<std::string> bar{this, "blabla", "bar", "the name of the bar"};
MyClass() : Config(readConfigFile("/etc/my-app.conf"))
{
std::cout << foo << "\n"; // will print 123 unless overriden
}
};
Currently, this is used by Store and its subclasses for store
parameters. You now get a warning if you specify a non-existant store
parameter in a store URI.
Previously, the Settings class allowed other code to query for string
properties, which led to a proliferation of code all over the place making
up new options without any sort of central registry of valid options. This
commit pulls all those options back into the central Settings class and
removes the public get() methods, to discourage future abuses like that.
Furthermore, because we know the full set of options ahead of time, we
now fail loudly if someone enters an unrecognized option, thus preventing
subtle typos. With some template fun, we could probably also dump the full
set of options (with documentation, defaults, etc.) to the command line,
but I'm not doing that yet here.
... and use this in Downloader::downloadCached(). This fixes
$ nix-build https://nixos.org/channels/nixos-16.09-small/nixexprs.tar.xz -A hello
error: cannot import path ‘/nix/store/csfbp1s60dkgmk9f8g0zk0mwb7hzgabd-nixexprs.tar.xz’ because it lacks a valid signature
Because config.h can #define things like _FILE_OFFSET_BITS=64 and not
every compilation unit includes config.h, we currently compile half of
Nix with _FILE_OFFSET_BITS=64 and other half with _FILE_OFFSET_BITS
unset. This causes major havoc with the Settings class on e.g. 32-bit ARM,
where different compilation units disagree with the struct layout.
E.g.:
diff --git a/src/libstore/globals.cc b/src/libstore/globals.cc
@@ -166,6 +166,8 @@ void Settings::update()
_get(useSubstitutes, "build-use-substitutes");
+ fprintf(stderr, "at Settings::update(): &useSubstitutes = %p\n", &nix::settings.useSubstitutes);
_get(buildUsersGroup, "build-users-group");
diff --git a/src/libstore/remote-store.cc b/src/libstore/remote-store.cc
+++ b/src/libstore/remote-store.cc
@@ -138,6 +138,8 @@ void RemoteStore::initConnection(Connection & conn)
void RemoteStore::setOptions(Connection & conn)
{
+ fprintf(stderr, "at RemoteStore::setOptions(): &useSubstitutes = %p\n", &nix::settings.useSubstitutes);
conn.to << wopSetOptions
Gave me:
at Settings::update(): &useSubstitutes = 0xb6e5c5cb
at RemoteStore::setOptions(): &useSubstitutes = 0xb6e5c5c7
That was not a fun one to debug!
The store parameter "write-nar-listing=1" will cause BinaryCacheStore
to write a file ‘<store-hash>.ls.xz’ for each ‘<store-hash>.narinfo’
added to the binary cache. This file contains an XZ-compressed JSON
file describing the contents of the NAR, excluding the contents of
regular files.
E.g.
{
"version": 1,
"root": {
"type": "directory",
"entries": {
"lib": {
"type": "directory",
"entries": {
"Mcrt1.o": {
"type": "regular",
"size": 1288
},
"Scrt1.o": {
"type": "regular",
"size": 3920
},
}
}
}
...
}
}
(The actual file has no indentation.)
This is intended to speed up the NixOS channels programs index
generator [1], since fetching gazillions of large NARs from
cache.nixos.org is currently a bottleneck for updating the regular
(non-small) channel.
[1] https://github.com/NixOS/nixos-channel-scripts/blob/master/generate-programs-index.cc
The fact that queryPathInfo() is synchronous meant that we needed a
thread for every concurrent binary cache lookup, even though they end
up being handled by the same download thread. Requiring hundreds of
threads is not a good idea. So now there is an asynchronous version of
queryPathInfo() that takes a callback function to process the
result. Similarly, enqueueDownload() now takes a callback rather than
returning a future.
Thus, a command like
nix path-info --store https://cache.nixos.org/ -r /nix/store/slljrzwmpygy1daay14kjszsr9xix063-nixos-16.09beta231.dccf8c5
that returns 4941 paths now takes 1.87s using only 2 threads (the main
thread and the downloader thread). (This is with a prewarmed
CloudFront.)
The inner lambda was returning a SQLite-internal char * rather than a
std::string, leading to Hydra errors liks
Caught exception in Hydra::Controller::Root->narinfo "path âø£â is not in the Nix store at /nix/store/6mvvyb8fgwj23miyal5mdr8ik4ixk15w-hydra-0.1.1234.abcdef/libexec/hydra/lib/Hydra/Controller/Root.pm line 352."
This makes it easier to create a diverted store, i.e.
NIX_REMOTE="local?root=/tmp/root"
instead of
NIX_REMOTE="local?real=/tmp/root/nix/store&state=/tmp/root/nix/var/nix" NIX_LOG_DIR=/tmp/root/nix/var/log
This is primarily to subsume the functionality of the
copy-from-other-stores substituter. For example, in the NixOS
installer, we can now do (assuming we're in the target chroot, and the
Nix store of the installation CD is bind-mounted on /tmp/nix):
$ nix-build ... --option substituters 'local?state=/tmp/nix/var&real=/tmp/nix/store'
However, unlike copy-from-other-stores, this also allows write access
to such a store. One application might be fetching substitutes for
/nix/store in a situation where the user doesn't have sufficient
privileges to create /nix, e.g.:
$ NIX_REMOTE="local?state=/home/alice/nix/var&real=/home/alice/nix/store" nix-build ...
This allows commands like "nix verify --all" or "nix path-info --all"
to work on S3 caches.
Unfortunately, this requires some ugly hackery: when querying the
contents of the bucket, we don't want to have to read every .narinfo
file. But the S3 bucket keys only include the hash part of each store
path, not the name part. So as a special exception
queryAllValidPaths() can now return store paths *without* the name
part, and queryPathInfo() accepts such store paths (returning a
ValidPathInfo object containing the full name).
Caching path info is generally useful. For instance, it speeds up "nix
path-info -rS /run/current-system" (i.e. showing the closure sizes of
all paths in the closure of the current system) from 5.6s to 0.15s.
This also eliminates some APIs like Store::queryDeriver() and
Store::queryReferences().
These are content-addressed paths or outputs of locally performed
builds. They are trusted even if they don't have signatures, so "nix
verify-paths" won't complain about them.
This enables an optimisation in hydra-queue-runner, preventing a
download of a NAR it just uploaded to the cache when reading files
like hydra-build-products.
Also, move a few free-standing functions into StoreAPI and Derivation.
Also, introduce a non-nullable smart pointer, ref<T>, which is just a
wrapper around std::shared_ptr ensuring that the pointer is never
null. (For reference-counted values, this is better than passing a
"T&", because the latter doesn't maintain the refcount. Usually, the
caller will have a shared_ptr keeping the value alive, but that's not
always the case, e.g., when passing a reference to a std::thread via
std::bind.)
Previously files in the Nix store were owned by root or by nixbld,
depending on whether they were created by a substituter or by a
builder. This doesn't matter much, but causes spurious diffoscope
differences. So use root everywhere.
I.e., not readable to the nixbld group. This improves purity a bit for
non-chroot builds, because it prevents a builder from enumerating
store paths (i.e. it can only access paths it knows about).
Namely:
nix-store: derivations.cc:242: nix::Hash nix::hashDerivationModulo(nix::StoreAPI&, nix::Derivation): Assertion `store.isValidPath(i->first)' failed.
This happened because of the derivation output correctness check being
applied before the references of a derivation are valid.
There is no risk of getting an inconsistent result here: if the ID
returned by queryValidPathId() is deleted from the database
concurrently, subsequent queries involving that ID will simply fail
(since IDs are never reused).
In the Hydra build farm we fairly regularly get SQLITE_PROTOCOL errors
(e.g., "querying path in database: locking protocol"). The docs for
this error code say that it "is returned if some other process is
messing with file locks and has violated the file locking protocol
that SQLite uses on its rollback journal files." However, the SQLite
source code reveals that this error can also occur under high load:
if( cnt>5 ){
int nDelay = 1; /* Pause time in microseconds */
if( cnt>100 ){
VVA_ONLY( pWal->lockError = 1; )
return SQLITE_PROTOCOL;
}
if( cnt>=10 ) nDelay = (cnt-9)*238; /* Max delay 21ms. Total delay 996ms */
sqlite3OsSleep(pWal->pVfs, nDelay);
}
i.e. if certain locks cannot be not acquired, SQLite will retry a
number of times before giving up and returing SQLITE_PROTOCOL. The
comments say:
Circumstances that cause a RETRY should only last for the briefest
instances of time. No I/O or other system calls are done while the
locks are held, so the locks should not be held for very long. But
if we are unlucky, another process that is holding a lock might get
paged out or take a page-fault that is time-consuming to resolve,
during the few nanoseconds that it is holding the lock. In that case,
it might take longer than normal for the lock to free.
...
The total delay time before giving up is less than 1 second.
On a heavily loaded machine like lucifer (the main Hydra server),
which often has dozens of processes waiting for I/O, it seems to me
that a page fault could easily take more than a second to resolve.
So, let's treat SQLITE_PROTOCOL as SQLITE_BUSY and retry the
transaction.
Issue NixOS/hydra#14.
On a system with multiple CPUs, running Nix operations through the
daemon is significantly slower than "direct" mode:
$ NIX_REMOTE= nix-instantiate '<nixos>' -A system
real 0m0.974s
user 0m0.875s
sys 0m0.088s
$ NIX_REMOTE=daemon nix-instantiate '<nixos>' -A system
real 0m2.118s
user 0m1.463s
sys 0m0.218s
The main reason seems to be that the client and the worker get moved
to a different CPU after every call to the worker. This patch adds a
hack to lock them to the same CPU. With this, the overhead of going
through the daemon is very small:
$ NIX_REMOTE=daemon nix-instantiate '<nixos>' -A system
real 0m1.074s
user 0m0.809s
sys 0m0.098s
For instance, it's pointless to keep copy-from-other-stores running if
there are no other stores, or download-using-manifests if there are no
manifests. This also speeds things up because we don't send queries
to those substituters.
Otherwise subsequent invocations of "--repair" will keep rebuilding
the path. This only happens if the path content differs between
builds (e.g. due to timestamps).
It is surprisingly impossible to check if a mountpoint is a bind mount
on Linux, and in my previous commit I forgot to check if /nix/store was
even a mountpoint at all. statvfs.f_flag is not populated with MS_BIND
(and even if it were, my check was wrong in the previous commit).
Luckily, the semantics of mount with MS_REMOUNT | MS_BIND make both
checks unnecessary: if /nix/store is not a mountpoint, then mount will
fail with EINVAL, and if /nix/store is not a bind-mount, then it will
not be made writable. Thus, if /nix/store is not a mountpoint, we fail
immediately (since we don't know how to make it writable), and if
/nix/store IS a mountpoint but not a bind-mount, we fail at first write
(see below for why we can't check and fail immediately).
Note that, due to what is IMO buggy behavior in Linux, calling mount
with MS_REMOUNT | MS_BIND on a non-bind readonly mount makes the
mountpoint appear writable in two places: In the sixth (but not the
10th!) column of mountinfo, and in the f_flags member of struct statfs.
All other syscalls behave as if the mount point were still readonly (at
least for Linux 3.9-rc1, but I don't think this has changed recently or
is expected to soon). My preferred semantics would be for MS_REMOUNT |
MS_BIND to fail on a non-bind mount, as it doesn't make sense to remount
a non bind-mount as a bind mount.
/nix/store could be a read-only bind mount even if it is / in its own filesystem, so checking the 4th field in mountinfo is insufficient.
Signed-off-by: Shea Levy <shea@shealevy.com>
It turns out that in multi-user Nix, a builder may be able to do
ln /etc/shadow $out/foo
Afterwards, canonicalisePathMetaData() will be applied to $out/foo,
causing /etc/shadow's mode to be set to 444 (readable by everybody but
writable by nobody). That's obviously Very Bad.
Fortunately, this fails in NixOS's default configuration because
/nix/store is a bind mount, so "ln" will fail with "Invalid
cross-device link". It also fails if hard-link restrictions are
enabled, so a workaround is:
echo 1 > /proc/sys/fs/protected_hardlinks
The solution is to check that all files in $out are owned by the build
user. This means that innocuous operations like "ln
${pkgs.foo}/some-file $out/" are now rejected, but that already failed
in chroot builds anyway.
