Accidentally broken by 4fede53c09
("nixos manuals: bring back package references").
Without this fix, grafana won't start:
$ systemctl status grafana
...
systemd[1]: Starting Grafana Service Daemon...
systemd[1]: Started Grafana Service Daemon.
grafana[666]: 2016/03/06 19:57:32 [log.go:75 Fatal()] [E] Failed to detect generated css or javascript files in static root (%!s(MISSING)), have you executed default grunt task?
systemd[1]: grafana.service: Main process exited, code=exited, status=1/FAILURE
systemd[1]: grafana.service: Unit entered failed state.
systemd[1]: grafana.service: Failed with result 'exit-code'.
Broken by 17389e256f.
The description attributes of mkOption are parsed by XSLT, so we can
create a DocBook manual out of it.
Unfortunately, the passwordHash option had a description which includes
a <password> placeholder which is recognized by DocBook XSL as a valid
start tag. So as there is obviously no </password>, the build of the
manual bailed out with a parsing error.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Reported-by: devhell <"^"@regexmail.net>
Since we don't restart sysinit.service in switch-to-configuration, this
additionally overrides systemd-binfmt.service to depend on
proc-sys-fs-binfmt_misc.automount, which is normally provided by
sysinit.service.
In NixOS/nixpkgs@da6bc44 @thoughtpolice
made the Transmission NixOS module override the umask setting in the
Transmission config. This commit removes that override.
I want a different umask setting and I guess it is possible that other
people might want it to. Thus I think it is a good idea to respect the
umask settings in the Transmission config.
Add ability to do a more traditional bspwm startup (using the bspwm-session
script provided by nixpkgs.bspwm) as an alternative to directly starting
sxhkd & bspwm
Also added the ability to specify a custom startup script, instead of
relying on the provided bspwm-session
add '-f 100' as an argument to sxhkd to keep it from flooding bspwm
add SXHKD_SHELL=/bin/sh to help default to a faster shell than what may
be set in $SHELL (example: with zsh)
- Enforce that an option declaration has a "defaultText" if and only if the
type of the option derives from "package", "packageSet" or "nixpkgsConfig"
and if a "default" attribute is defined.
- Enforce that the value of the "example" attribute is wrapped with "literalExample"
if the type of the option derives from "package", "packageSet" or "nixpkgsConfig".
- Warn if a "defaultText" is defined in an option declaration if the type of
the option does not derive from "package", "packageSet" or "nixpkgsConfig".
- Warn if no "type" is defined in an option declaration.
Both Qt and GTK load plugins from the active profiles
automatically, so it is sufficient to install input methods
system-wide. Overriding the plugin paths may interfere with correct
operation of other plugins.
Updates gitlab to the current stable version and fixes a lot of features that
were broken, at least with the current version and our configuration.
Quite a lot of sweat and tears has gone into testing nearly all features and
reading/patching the Gitlab source as we're about to deploy gitlab for our
whole company.
Things to note:
* The gitlab config is now written as a nix attribute set and will be
converted to JSON. Gitlab uses YAML but JSON is a subset of YAML.
The `extraConfig` opition is also an attribute set that will be merged
with the default config. This way *all* Gitlab options are supported.
* Some paths like uploads and configs are hardcoded in rails (at least
after my study of the Gitlab source). This is why they are linked from
the Gitlab root to /run/gitlab and then linked to the configurable
`statePath`.
* Backup & restore should work out of the box from another Gitlab instance.
* gitlab-git-http-server has been replaced by gitlab-workhorse upstream.
Push & pull over HTTPS works perfectly. Communication to gitlab is done
over unix sockets. An HTTP server is required to proxy requests to
gitlab-workhorse over another unix socket at
`/run/gitlab/gitlab-workhorse.socket`.
* The user & group running gitlab are now configurable. These can even be
changed for live instances.
* The initial email address & password of the root user can be configured.
Fixes#8598.
NetworkManager needs an additional avahi-user to use link-local
IPv4 (and probably IPv6) addresses. avahi-autoipd also needs to be
patched to the right path.
Add a module to make options to pam_oath module configurable.
These are:
- enable - enable the OATH pam module
- window - number of OTPs to check
- digits - length of the OTP (adds support for two-factor auth)
- usersFile - filename to store OATH credentials in
We don't want to build all those things along with the manual, so that's
what the defaultText attribute is for.
Unfortunately a few of them were missing, so let's add them.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This commit adds the firmware for the Intel 2200BG wireless cards for
the ipw2200 kernel module. Also it changes the
networking.enableIntel2200BGFirmware option to set it as
hardware.firmware since firmware-linux-nonfree does not contain the
appropiate firmware anymore. Also hardware.enableAllFirmware does enable
the intel2200BGFirmware now.
This reverts commit b861bf8ddf, because according to @mdorman [1] this
change rendered his NixOS systems unbootable, and we probably don't want that.
[1] b861bf8ddf (commitcomment-16058598)
Allow usage of list of strings instead of a comma-separated string
for filesystem options. Deprecate the comma-separated string style
with a warning message; convert this to a hard error after 16.09.
15.09 was just released, so this provides a deprecation period during
the 16.03 release.
closes#10518
Signed-off-by: Robin Gloster <mail@glob.in>
The test only checked for existence of the rule file in the output path
of the rulefile generator.
However, we also need to check whether the basename of the file is also
the one we're currently searching for.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Partially reverts the following commits:
9f2a61c59c9c13fe6604
As @edolstra pointed out, it would make more sense to do this by default
instead of having that allowImpurePaths option. This of course might
break systems which add extra packages to udev, but on the upside it's
hard to miss one of these paths now because it won't get buried in the
ocean of build output lines.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
So far we were merely printing a warning if there are still references
to (/usr)/s?bin, but we actually want to make sure that we fix those
paths, especially on updates of packages that come with udev rules.
This adds a new option allowImpurePaths, which when set to false will
cause the "udev-rules" derivation to fail.
I've set this to true by default, to not break existing systems too much
and the intention is to set it to false for a few NixOS VM tests.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
We were trying to find FHS references in all of the rules found in
services.udev.packages. Unfortunately we're still fixing up paths in the
same derivation where we are checking those references, so for example
references to /sbin/modprobe were still printed to be needed to fixup
even though they were already fixed at the time.
So now we're printing a more helpful warning message which is also
conditional (before the warning message was printed regardless of
whether there are any rules that need fixup) and is based off the rules
that were already fixed up.
The new warning message not only contains the build-local rule files but
also the original files from other store paths and the FHS path
references that were still found.
With 8ecd3a5e1d reverted, we now get this:
/nix/store/...-udev-rules/63-md-raid-arrays.rules (originally from
/nix/store/...-mdadm-3.3.4/lib/udev/rules.d/63-md-raid-arrays.rules)
contains references to /usr/bin/readlink and /usr/bin/basename.
Which is now more accurate to what is not yet fixed and where it's
coming from.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
In 8ecd3a5, we fixed up the FHS paths for stage 1, but unfortunately we
have a similar udev rules generator twice one for the initrd and one
without. So we might need to refactor this in the future.
For now, let's just fix the references to readlink and basename in the
udev module as well until we have properly addressed this.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Fixes: #12722
It's not entirely clear why this happens, but sometimes /proc/1/exe
returns a bogus value, like
/ar3a3j6b9livhy5fcfv69izslhgk4gcz-systemd-217/lib/systemd/systemd. In
any case, we can just conservatively assume that we need to restart
systemd when this happens.
Fixes#10261.
This reverts most of 89e983786a, as those references are sanitized now.
Fixes#10039, at least most of it.
The `sane` case wasn't fixed, as it calls a *function* in pkgs to get
the default value.
This is an improved version of original #12357.
For the purpose of generating docs, evaluate options with each derivation
in `pkgs` (recursively) replaced by a fake with path "\${pkgs.attribute.path}".
It isn't perfect, but it seems to cover a vast majority of use cases.
Caveat: even if the package is reached by a different means,
the path above will be shown and not e.g. `${config.services.foo.package}`.
As before, defaults created by `mkDefault` aren't displayed,
but documentation shouldn't (mostly) be a reason to use that anymore.
Note: t wouldn't be enough to just use `lib.mapAttrsRecursive`,
because derivations are also (special) attribute sets.
Fixes references coming from the mdadm udev rules.
This addresses #12722 (mdadm udev rules have references to /usr/bin) but
still won't fix the warning, though (if we want to fix the warnings, we
will have to patch the udev rules generater in services/hardware/udev).
For common mdraid functionality, this shouldn't fix anything, because
the wrong references seem to only apply to containers, see these
(wrapped) lines from ${mdadm}/lib/udev/rules.d/63-md-raid-arrays.rules:
# Tell systemd to run mdmon for our container, if we need it.
ENV{MD_LEVEL}=="raid[1-9]*",
ENV{MD_CONTAINER}=="?*",
PROGRAM="/usr/bin/readlink $env{MD_CONTAINER}",
ENV{MD_MON_THIS}="%c"
ENV{MD_MON_THIS}=="?*",
PROGRAM="/usr/bin/basename $env{MD_MON_THIS}",
ENV{SYSTEMD_WANTS}+="mdmon@%c.service"
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Partially reverts commit 901163c0c7.
This has broken remote SSH into initrd because ${cfg.shell} is not
expanded. Also, nsswitch is useless without libnss_files.so which
are installed by initrd-ssh.
This is a regression introduced by merging the EBS and S3 images. The
EBS images had a special marker /.ebs to prevent the initrd from using
ephemeral storage for the unionfs, but this marker was missing in the
consolidated image.
The fix is to check the file ami-manifest-path on the metadata server
to see if we're an S3-based instance. This does require networking in
the initrd.
Issue #12613.
The default behavior with an m3.medium instance is to relocate
/nix and /tmp to /disk0 because an assumption is made that any
ephemeral disk is larger than the root volume. Rather than make
that assumption, add a check to see if the disk is larger, and
only then relocate /nix and /tmp.
This addresses https://github.com/NixOS/nixpkgs/issues/12613
Regression introduced by 5184aaa1ea.
The fix was intended to remove the "x == true/false" assertions, but by
accident a "x == false" was made "x == true" instead of "(!x)".
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Reported-by: devhell <"^"@regexmail.net>
Reportedly some ARM boards need some boot code at the start of a SD card
that could be larger than a megabyte. Change it to 8M, and while at it
reduce the /boot size such that the root partition should now start on a
128M boundary (the flash on SD cards really don't like non-aligned
writes these days).
Before the error if the wrong default desktop was chosen would be:
/nixpkgs-channels/lib/modules.nix:282:11:
Default desktop manager ($(defaultDM)) not found.
which has the string interpolation done incorreclty. Now that is fixed
and it is more user-friendly as:
/nixpkgs-channels/lib/modules.nix:282:11:
Default desktop manager (gnome) not found.
Probably you want to change
services.xserver.desktopManager.default = "gnome";
to one of
services.xserver.desktopManager.default = "gnome3";
services.xserver.desktopManager.default = "none";
- Fixed a bug in bluedevil (link to a .js file)
- Made bluez5 the default bluetooth service except for kde4
- created org.bluez.obex systemd dbus service
- Patched bluez5 using bluez-5.37-obexd_without_systemd-1.patch
in order to enable obex when using either the bluedevil plasmoid
or dolpin file manager within plasma workspaces 5.
The functionality was tested using a Sony Xperia Z, the machine
and the handset paired and two different files were sent in both
directions successfully.
- Now `pkg.outputUnspecified = true` but this attribute is missing in
every output, so we can recognize whether the user chose or not.
If (s)he didn't choose, we put `pkg.bin or pkg.out or pkg` into
`systemPackages`.
- `outputsToLink` is replaced by `extraOutputsToLink`.
We add extra outputs *regardless* of whether the user chose anything.
It's mainly meant for outputs with docs and debug symbols.
- Note that as a result, some libraries will disappear from system path.
- added numlock on boot switch
- simply add :
services.xserver.displayManager.sddm.autoNumlock = true;
to configuration.nix and sddm will start
with numlock enabled.
If we limit SSH_ASKPASS to interactive shells, users are unable to trigger
the ssh-passphrase dialog from their desktop environment autostart scripts.
Usecase: I call ssh-add during my desktop environment autostart and want to have
the passphrase dialog immediately after startup.
For this to work, SSH_ASKPASS needs to be propagated properly on
non-interactive shells.
See http://nixos.org/nixpkgs/manual/#sec-package-naming
I've added an alias for multipath_tools to make sure that we don't break
existing configurations referencing the old name.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Make top level /var/lib/postfix as root:root 0755
After generating custom configs in /var/lib/postfix/conf,
`postfix set-permissions` called, to perform all required tricks
related to queue handling (postfix use file mode bits to keep
some internal statuses, so `chmod -R` not recommended by authors,
see comments in $out/libexec/postfix/post-install for details)
Also post-install script was patched, to skip permission check/update
for files inside $out, as well as symlinks following to $NIX_STORE.
Config file `main.cf` extended with all default directory locations,
to prevent post-install script from guessing and overwrite them.
And finally all actions in activation script snippets performed
by postmap/postalias/postfix tools from current build, not random one
from paths.
When using `--ensure-unique-name`, don't needlessly append `"-0"` if the
container name is already unique.
This is especially helpful with NixOps since when it deploys to a
container it uses `--ensure-unique-name`. This means that the container
name will never match the deployment host due to the `"-0"`. Having the
container name and the host name match isn't exactly a requirement, but
it's nice to have and a small change.
Set this option to 'true' (default: 'false') to enable extension mechanisms for
DNS (EDNS) in your local glibc resolver. This is required for supporting
DNSSEC, for example.
Implementation detail: the patch changes assignments to "resolv_conf_options"
to use "+=" instead of "=" to ensure that multiple users of that variable don't
overwrite each other. The generated config file is a shell script, after all,
so this should work fine.
Closes https://github.com/NixOS/nixpkgs/issues/12470.
- add missing types in module definitions
- add missing 'defaultText' in module definitions
- wrap example with 'literalExample' where necessary in module definitions
Building config.system.build.isoImage would fail with the following
error using the channel:
ln: failed to create symbolic link
'/nix/store/zz0hzi5imrg4927v6f8mv281qs6v6pbq-nixos-16.03pre69762.e916273/nixos/nixpkgs/.': File exists
The fix skips symlink as it already exists if the channel
nixpkgs copy is used.
Fixes#10367
Regression introduced by 3891d3e654.
Merging multiple options with type "str" won't work and give an
evaluation error. For extra configuration lines in the Postfix config it
really should be "lines", especially because even the description
mentions "extra lines".
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Modifies libvirt package to search for configs in /var/lib and changes
libvirtd service to copy the default configs to the new location.
This enables the user to change e.g. the networking configuration with
virsh or virt-manager and keep those settings.
This reverts most of 89e983786a, as those references are sanitized now.
Fixes#10039, at least most of it.
The `sane` case wasn't fixed, as it calls a *function* in pkgs to get
the default value.
Sadly, we can't instruct systemd to properly restart device-name.swap when this service restarts (or I haven't found the way to do so). As of now blindly restarting it would only get you a bunch of errors about device already used -- let's avoid it.
This reverts commit 6353f580f9.
Unfortunately cache=none doesn't work with all filesystem options.
Hydra tests error out with: file system may not support O_DIRECT
See http://hydra.nixos.org/build/30323625/
Setting nixosVersion to something custom is useful for meaningful GRUB
menus and /nix/store paths, but actuallly changing it rebulids the
whole system path (because of `nixos-version` script and manual
pages). Also, changing it is not a particularly good idea because you
can then be differentitated from other NixOS users by a lot of
programs that read /etc/os-release.
This patch introduces an alternative option that does all you want
from nixosVersion, but rebuilds only the very top system level and
/etc while using your label in the names of system /nix/store paths,
GRUB and other boot loaders' menus, getty greetings and so on.
This hopefully fixes intermittent initrd failures where udevd cannot
create a Unix domain socket:
machine# running udev...
machine# error getting socket: Address family not supported by protocol
machine# error initializing udev control socket
machine# error getting socket: Address family not supported by protocol
The "unix" kernel module is supposed to be loaded automatically, and
clearly that works most of the time, but maybe there is a race
somewhere. In any case, no sane person would run a kernel without Unix
domain sockets, so we may as well make it builtin.
http://hydra.nixos.org/build/30001448
Two concurrent tarsnap backups cannot be run at the same time with the
same keys - completely separate sets of keys must be generated for each
archive in this case, if you want backups to overlap.
This extends the archives attrset to support a 'keyfile' option, which
defaults to /root/tarsnap.key like the top-level attribute.
With this change, if you generate two keys with tarsnap-keygen(1) and
use each of those separately for each archive, you can backup
concurrently.
Signed-off-by: Austin Seipp <aseipp@pobox.com>
Tarsnap locks the cachedir during backup, meaning if you specify
multiple backups with a shared cache that might overlap (for example,
one backup may take an hour), secondary backups will fail. This isn't
very nice behavior for the obvious reasons.
This splits the cache dirs for each archive appropriately. Note that
this will require a rebuild of your archive caches (although if you were
only using one archive for your whole system, you can just move the
directory).
Signed-off-by: Austin Seipp <aseipp@pobox.com>
A machine may not always be active (or online!) when a backup timer
triggers, meaning backups can be missed - now we properly set the
tarsnap timer's Persistent option so systemd will run the command even
when the machine wasn't online at that exact time.
However, we also need to make sure that we can contact the tarsnap
server reliably before we start the backup. So, we attempt to ping the
access endpoint in a loop with a sleep, before continuing.
This fixes#8823.
Signed-off-by: Austin Seipp <aseipp@pobox.com>
Tarsnap locks the cachedir during backup, meaning if you specify
multiple backups with a shared cache that might overlap (for example,
one backup may take an hour), secondary backups will fail. This isn't
very nice behavior for the obvious reasons.
This splits the cache dirs for each archive appropriately. Note that
this will require a rebuild of your archive caches (although if you were
only using one archive for your whole system, you can just move the
directory).
Signed-off-by: Austin Seipp <aseipp@pobox.com>
The Bitmessage protocol v3 became mandatory on 16 Nov 2014 and notbit does not support it, nor has there been any activity in the project repository since then.
Part of the way towards #11864. We still don't have the auditd
userland logging daemon, but journald also tracks audit logs so we
can already use this.
This hopefully fixes intermittent test failures like
http://hydra.nixos.org/build/29962437
router# [ 240.128835] INFO: task mke2fs:99 blocked for more than 120 seconds.
router# [ 240.130135] Not tainted 3.18.25 #1-NixOS
router# [ 240.131110] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
assuming that these are caused by high load on the host.
... because we make it built-in by default.
I can't imagine anyone who wanted to purge this module from his/her system,
so let's keep it simple, at least for now.
This commit adds the options --build-host and --target-host to nixos-rebuild.
--build-host instructs nixos-rebuild to perform all nix builds on the
specified host (via ssh). Build results are then copied back to the
local machine and used when activating the system.
--build-target instructs nixos-rebuild to activate the configuration
not on the local machine but on the specified remote host. Build
results are copied to the target machine and then activated there (via ssh).
It is possible to combine the usage of --build-host and --target-host,
in which case you can perform the build on one remote machine and deploy
the configuration to another remote machine. The only requirement is that
the build host has a working ssh connection to the target host (if the
target is not local), and that the local machine can connect to both
the target and the build host. Also, your user must be allowed to copy
nix closures between the local machine and the target and host machines.
At no point in time are the configuration sources (the nix files) copied
anywhere. Instead, nix evaluation always happens locally
(with nix-instantiate). The drv-file is then copied and realised remotely
(with nix-store).
As a convenience, if only --target-host is specified, --build-host is
implicitly set to that host too. So if you want to build locally and deploy
remotely you have to explicitly set "--build-host localhost".
To activate (test, boot or switch) you need to have root access to the
target host. You can specify this by "--target-host root@myhost".
I have tested the obvious scenarios and they are working. Some of the
combinations of --build-host and --target-host and the various actions might
not make much sense, and should maybe be forbidden (like setting a remote
target host when building a VM), and some combinations might not work at all.
Previously this barfed with:
updating GRUB 2 menu...
fileparse(): need a valid pathname at /nix/store/zldbbngl0f8g5iv4rslygxwp0dbg1624-install-grub.pl line 391.
warning: error(s) occured while switching to the new configuration
* Patched fish to load /etc/fish/config.fish if it exists (by default,
it only loads config relative to itself)
* Added fish-foreign-env package to parse the system environment
closes#5331
We seem to be in an unfortunate situation: booting without 'nomodeset'
causes hangs when booting on some NVIDIA cards (6948c3ab80), but on the
other hand adding 'nomodeset' prevents X from starting on other hardware
(e.g. issue #10381 and my Thinkpad X250 with an integrated Broadwell GPU).
Attempt to remedy this situation a bit by adding a separate entry in the
ISOLINUX menu (with the non-'nomodeset' being the default).
The docker module used different code for socket-activated docker daemon than for the non-socket activated daemon.
In particular, if the socket-activated daemon is used, then modprobe wasn't set up to be usable and in PATH for
the docker daemon, which resulted in a failure to start the daemon with overlayfs as storageDriver if the
`overlay` kernel module wasn't already loaded. This commit fixes that bug (which only appears if socket
activation is used), and also reduces the duplication between code paths so that it's easier to keep
both in sync in future.
I think the name 'listenAddress' is more descriptive. Other NixOS
modules that define 'host' either use it as listen address or as address
a client connects to. listenAddress is unambiguous.
The addition of 'host' was added earlier today[1], so not bothering with
./nixos/modules/rename.nix.
[1]: 44ea184997 ("jenkins ci enhancement: add port and prefix option")
As named these options enable to specify a bind host and url prefix
to be used by jenkins. Adding these options in the config rather than
using extra arguments allows us to re-use those information in other
services using jenkins such as jenkins-job-builder or a reverse proxy.
Add new option declarations to control what information is published
by the avahi daemon. The default values are chosen to respect the
privacy of the user over the connectivity of the system.
The kernel default for `link_power_management_policy` is `"max_performance"`.
This commit:
f169f60575
set the NixOS default to `"min_performance"`.
This issue (https://github.com/NixOS/nixpkgs/issues/11276) details my long
journey to discover this after several file system failures incorrectly
attributed to `TRIM` and `NCQ` settings.
I think we should use the kernel default of `"max_performance"` to assure
the best experience for new users with SSDs and to conform to the defaults of
the kernel and other distros.
The three KDE package sets now have circular dependencies between them,
so they can only be built if they are merged into a single package set
during evaluation.
- if xserver.tty and/or display are set to null, then don't specify
them, or the -logfile argument in the xserverArgs
- For lightdm, we set default tty and display to null and we determine
those at runtime based on arguments passed. This is necessary because
we run multiple X servers so they can't all be on the same display
This reverts commit 02b568414d.
With a5bc11f and 6353f58 in place, we really don't need this anymore.
After running about 500 VM tests on my Hydra, it still didn't improve
very much.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
As @domenkozar noted in #10828, cache=writeback seems to do more harm
than good:
https://github.com/NixOS/nixpkgs/issues/10828#issuecomment-164426821
He has tested it using the openstack NixOS tests and found that
cache=none significantly improves startup performance.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
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>
- Add new service for `clamd`, the ClamAV daemon.
- Replace the old upstart "jobs" section with systemd.services
- Remove unnecessary config options.
- Use `mkEnableOption`
We hit page allocation failures a lot at random for VM tests, in case of
my own Hydra when it comes to the installer tests. The reason for this
is that once the memory of the VM gets heavily fragmented the kernel is
unable to allocate new pages.
Setting vm.min_free_kbytes to 16MB forces the kernel to keep a minimum
of 16 MB free.
I've done some testing accross repeated runs of the installer tests with
and without vm.min_free_kbytes set. So accross 30 test runs for each
settings, all of the tests with the option being set passed while 14
tests without that sysctl option triggered page allocation failures.
Sure, running 30 tests is not a guarantee that 16MB is enough, but we'll
see how it turns out in the long run across all VM tests.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Adds a new service module for shairport-sync. Tested with a local
and remote pulseaudio server. Needs to be run as a user in the pulse group
to access pulseaudio.
These options allow setting the start and stop scripts for the display
manager. Making these configurable is necessary to allow some hardware
configurations. Upstream ships empty scripts by default, anyway.
This new service invokes `simp_le` for a defined set of certs on a regular
basis with a systemd timer. `simp_le` is smart enough to handle account
registration, domain validation and renewal on its own. The only thing
required is an existing HTTP server that serves the path
`/.well-known/acme-challenge` from the webroot cert parameter.
Example:
services.simp_le.certs."foo.example.com" = {
webroot = "/var/www/challenges";
extraDomains = [ "www.example.com" ];
email = "foo@example.com";
validMin = 2592000;
renewInterval = "weekly";
};
Example Nginx vhost:
services.nginx.appendConfig = ''
http {
server {
server_name _;
listen 80;
listen [::]:80;
location /.well-known/acme-challenge {
root /var/www/challenges;
}
location / {
return 301 https://$host$request_uri;
}
}
}
'';
Now the default way to define NixOS window manager modules is to use
mkEnableOption to describe the module itself.
In this commit, all files on nixos/modules/services/x11/window-managers
are changed.
Commit changes default version to 7.0.10, 7.0.5 version is kept for
people reluctant to update. Needed info has also been added for
versions 8.0, 8.1 and 8.2 only the latest minor version of each
major version is included.
This option allows user to specify a url prefix for owncloud.
By default it is set to "" and the document root will be set
to owncloud's dir.
If a prefix is set, e.g. urlPrefix = "/owncloud"
an alias will be created using that prefix to point to owncloud's
dir and owncloud will be available at http://localhost/owncloud
