Since ca1dc3f70bf98e2424b7b2666ee2180675b67451, the NAR parser has moved
the preallocate & receive steps into the file handle class to remove the
assumption that only one file can be handled at a time.
The third argument to `open()` in `-|` mode is passed to a shell if it's
a string. In my case the store URI contains
`?secret-key=${signingKey.directory}/secret&compression=zstd`
For the `nix store cat` case this means that
* until `&` the process will be started in the background. This fails
immediately because no path to cat is specified.
* `compression=zstd` is a variable assignment
* the `$path` argument to `store cat` is attempted to be executed as
another command
Passing just the list solves the problem.
`dev-notes` are severely outdated. I dropped everything except one note
that I moved to hacking.md. The parts about creating users are also
covered elsewhere.
The `update-dbix` part got a just command to make it discoverable again.
When an artifact is requested from hydra the output is first copied
from the nix store into memory and then sent as a response, delaying
the download and taking up significant amounts of memory.
As reported in https://github.com/NixOS/hydra/issues/1357
Instead of calling a command and blocking while reading in the entire
output, this adds read_into_socket(). the function takes a
command, starting a subprocess with that command, returning a file
descriptor attached to stdout.
This file descriptor is then by responsebuilder of Catalyst to steam
the output directly
Lost in the h-e-j -> n-e-j migration, causing evaluation to always be
single threaded and limited to 4GiB RAM. Follow the config settings like
h-e-j used to do (via C++ code).
There are some known regressions regarding local testing setups - since
everything was kinda half written with the expectation that build dir =
source dir (which should not be true anymore). But everything builds and
the test suite runs fine, after several hours spent debugging random
crashes in libpqxx with MALLOC_PERTURB_...
The current way this whole build works is incompatible with having a
separate build dir, or at least with having a separate build dir. To be
improved in the future - maybe minimize the dependencies a bit. But this
isn't so much data that we really have to care.
New jobs have their "new" status take precedence over them being
"failed" or "queued", which means actions that can act on "failed" or
"queued" jobs weren't shown to the user when they could only act on
"new" jobs.
nix-eval-jobs streams output, unlike hydra-eval-jobs. Now that we've
migrated, we can use this to:
1. Use less RAM by avoiding buffering a whole eval's worth of metadata
into a Perl string and an array of JSON objects.
2. Make evals latency a bit lower by allowing the queue runner to start
ingesting builds faster.
The feature cannot easily be ported to nix-eval-jobs since it requires
deep integration into the evaluator, and h.n.o doesn't use it. Later
more of this will be ripped out.
This was the source of a flaky test because sometimes hydra-notify was
quick enough to send out `buildStarted` and sometimes it apparently
wasn't which was quickly spottable with `nix build --rebuild`.
Removing that status update doesn't make a difference functionally,
gitea doesn't differentiate between "queued" and "running", so we send
the same status ("pending") out on both events, so we'd even safe one
avoidable request.
(cherry picked from commit 806c375c338b4e6a1d276b96994018908784bf11)
Instead of just going for "whatever is the oldest build we know of",
use the following first:
- Is the step more constrained? If so, schedule it first to avoid
filling up "more desirable" build slots with less constrained builds.
- Does the step have more dependents? If so, schedule it first to try
and maximize open parallelism and breadth of scheduling options.
This allows for better builder usage when the queue runner is busy. To
avoid running into uncontrollable imbalances between builder/queue
runner, we only release the machine reservation after the local
throttler has found a slot to start copying the outputs for that build.
We don't rely on sequential / monotonic build IDs processing anymore, so
randomizing actually has the advantage of mixing builds for different
systems together, to avoid only one chunk of builds for a single system
getting processed while builders for other systems are starved.
Each output for a given step being ingested is looked up in parallel,
which should basically multiply the speed of builds ingestion by the
average number of outputs per derivation.
Running the query with/without it shows that it makes no difference to
postgres, since there's an index on finished=0 already. This allows a
few simplifications, but also paves the way towards running multiple
parallel monitor threads in the future.
By looking at the ratio of running vs. waiting for the dispatcher and
the queue monitor, we should get better visibility into what hydra is
currently bottlenecked on.
There are other side effects we can try to measure to get to the same
result, but having a simple way doesn't cost us much.
This is implement in an extremely hacky way due to poor DBIx feature
support. Ideally, what we'd need is a way to tell DBIx to ignore the
errormsg column unless explicitly requested, and to automatically add a
computed 'errormsg IS NULL' column in others. Since it does not support
that, this commit instead hacks some support via method overrides while
taking care to not break anything obvious.
My current theory is that running more parallel xz than available CPU
cores is reducing our overall throughput by requiring more scheduling
overhead and more cache thrashing.
When building e.g. nixpkgs, the "Running builds" view will mostly look
like this
hello.x86_64-linux (Build of hello-X.Y)
exa.x86_64-linux (Build of exa-X.Y)
...
This doesn't provide any useful information. Showing the step name only
makes sense if it's not a child of the job's derivation. With this
patch, that information will only be shown if the drv name (i.e. w/o
`/nix/store/` prefix, .drv ext & hash) is not equal to the drv name of
the job itself (build.nixname).
When using Hydra to build machine configurations, you'll often see
"nixosConfigurations.foo" five times, i.e. for each build step being
run. This isn't very helpful I think because in such a case, a single
build step can also be compiling the Linux kernel.
This change also fetches the `drvpath` and `type` from the `buildsteps`
relation. We're already joining it, so this doesn't make much difference
(confirmed via query logging that this doesn't cause extra SQL queries).
Unfortunately build steps don't have a human readable name, so I'm
deriving it from the drvpath by stripping away the hash (assuming that
it'll never contain a `-` and that `/nix/store/` is used as prefix). I
decided against using the Nix bindings for that to avoid too much
overhead due to store operations for each build step.
In 73694087a0 I gave builds that failed
because of a timeout or exceeded log limit a stop sign and I stand by
that reasoning: with that it's possible to distinguish between actual
build failures and rather transient things such as timeouts.
Back then I considered it a feature that these are shown in a different
tab, but I don't think that's a good idea anymore. When using a jobset to
e.g. track the regressions from a mass rebuild (like a compiler or gcc
update), "Newly failed builds" should exclusively display regressions (and
flaky builds of course, not much I can do about that).
Also, when a bunch of builds fail in such a jobset because of e.g. a
broken connection to a builder that results in a timeout, I want to be
able to restart them all w/o rebuilding actual regressions.
To make it clear that we not only have "Aborted" builds in the tab, I
renamed the label to "Aborted / Timed out".
Maximilian Bosch
Ilya K
Pierre Bourdon
emily
71rd
eldritch horrors
leo60228