This seems to be the root cause of the random page allocation failures
and @wizeman did a very good job on not only finding the root problem
but also giving a detailed explanation of it in #10828.
Here is an excerpt:
The problem here is that the kernel is trying to allocate a contiguous
section of 2^7=128 pages, which is 512 KB. This is way too much:
kernel pages tend to get fragmented over time and kernel developers
often go to great lengths to try allocating at most only 1 contiguous
page at a time whenever they can.
From the error message, it looks like the culprit is unionfs, but this
is misleading: unionfs is the name of the userspace process that was
running when the system ran out of memory, but it wasn't unionfs who
was allocating the memory: it was the kernel; specifically it was the
v9fs_dir_readdir_dotl() function, which is the code for handling the
readdir() function in the 9p filesystem (the filesystem that is used
to share a directory structure between a qemu host and its VM).
If you look at the code, here's what it's doing at the moment it tries
to allocate memory:
buflen = fid->clnt->msize - P9_IOHDRSZ;
rdir = v9fs_alloc_rdir_buf(file, buflen);
If you look into v9fs_alloc_rdir_buf(), you will see that it will try
to allocate a contiguous buffer of memory (using kzalloc(), which is a
wrapper around kmalloc()) of size buflen + 8 bytes or so.
So in reality, this code actually allocates a buffer of size
proportional to fid->clnt->msize. What is this msize? If you follow
the definition of the structures, you will see that it's the
negotiated buffer transfer size between 9p client and 9p server. On
the client side, it can be controlled with the msize mount option.
What this all means is that, the reason for running out of memory is
that the code (which we can't easily change) tries to allocate a
contiguous buffer of size more or less equal to "negotiated 9p
protocol buffer size", which seems to be way too big (in our NixOS
tests, at least).
After that initial finding, @lethalman tested the gnome3 gdm test
without setting the msize parameter at all and it seems to have resolved
the problem.
The reason why I'm committing this without testing against all of the
NixOS VM test is basically that I think we can only go better but not
worse than the current state.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
While debugging an issue with running NixOps tests, I found out that the
output from debClosureGenerator is not deterministic.
The reason behind this is the way how Provides and Replaces fields are
handled. I haven't yet found out what's the exact issue, but so far
packages "Provides" are more or less picked at random.
So, running the NixOps Hetzner tests we get either mawk, original-awk or
gawk altering on every invocation.
While for the test it isn't poisionous whether wi have mawk or gawk,
having original-awk certainly is, because live-build only works with
mawk or gawk.
The best solution would obviously be to make debClosureGenerator
deterministic, but in the case of "Provides: awk", we can safely pick
mawk by default, because the latter has a "Priority: required" in its
package description.
This also has the advantage that we can safely cherry-pick this to
release-15.09 because it's very unlikely that we'll break the
debClosureGenerator by adding a dependency to commonDebPackages.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This ensures that the intermediate machine is shut down only after the
migration has finished writing the memory dump to disk, to ensure we
don't end up with empty state files depending on how fast the migration
finished before we actually shut down the VM.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This ensures that the builder isn't waiting forever if the Windows VM
drops dead while we're waiting for the controller VM to signal that a
particular command has been executed on the Windows VM. It won't ever
happen in such cases so it doesn't make sense to wait for the timeout.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
So far, we determined this based on stdenv.is64bit, but there are cases
where you want to run/build a 32bit program on a 64 bit Windows.
This is now possible, by passing windowsImage.arch = "i686" | "x86_64"
to runInWindowsVM. Based an what was passed, the corresponding Cygwin
packages and setup.exe are bootstrapped.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Another very annoying part. Unfortunately, the only option we might have
here is to include it in nixpkgs or maybe make a fixed Hash on the
result of the closure fetcher.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
As the official Cygwin setup binary download doesn't come in snapshots
or even versioned, the fetchurl of setup.exe will frequently fail, which
in turn will annoy us as hell (or at least me).
One warning though: The fetchurl is currently broken and the cross-build
might not work yet for example on mingw32 (mingw-w64 branch on its way),
but the upstream URL has already changed and the new version contains a
bug (not yet tracked down) which breaks our Windows bootstrap process.
So to conclude: If it's already broken, make it at least "less broken".
"Not broken" is coming soon with the merge of the mingw-w64 branch.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
