systemd gets sad otherwise and it is very difficult to console it
Change-Id: Ic6405489532c407273e5634474185f2947420b37
Reviewed-on: https://cl.tvl.fyi/c/depot/+/851
Reviewed-by: glittershark <grfn@gws.fyi>
Reviewed-by: BuildkiteCI
Tested-by: BuildkiteCI
it's not glittershark because grfn is the username I have on my laptop
and I want to be able to ssh without an `@`.
Change-Id: Ie1fb6f5e12f3ac52a44680704179bd27a00a7768
Reviewed-on: https://cl.tvl.fyi/c/depot/+/850
Reviewed-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
Tested-by: BuildkiteCI
I generated a new one for whitby, so it's called whitby because that's
the only thing I'm going to be using it for.
Change-Id: Ie8ebdb1e58c76e447ee3a3ea0b1b244fe000e3e4
Reviewed-on: https://cl.tvl.fyi/c/depot/+/849
Reviewed-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
Tested-by: BuildkiteCI
When visiting buffers that represent home-manager or nixos modules, set
compile-command to the relevant switch command.
Change-Id: I565e092d1e6b97eb7fee9d88211c0ee6ebb30b53
Reviewed-on: https://cl.tvl.fyi/c/depot/+/846
Reviewed-by: glittershark <grfn@gws.fyi>
This adds NixOS configuration for the machine whitby.tvl.fyi.
No interesting services are configured yet, so this configuration is
quite plain.
Change-Id: I67b7c75ebd6e298719b52e6b3bd83cc3be3c45d8
Reviewed-on: https://cl.tvl.fyi/c/depot/+/843
Tested-by: BuildkiteCI
Reviewed-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Reviewed-by: lukegb <lukegb@tvl.fyi>
Ensure that $NIX_PATH is empty using a .envrc file. nix.dev considers references
to <nixpkgs>, <briefcase>, <depot>, <unstable> to be an anti-pattern because it
hinders reproducibility by introducing an implicit environment.
This repository is chock-full of references to
<{nixpkgs,unstable,briefcase,depot}>, so right now everything is broken.
I don't know why-- but I don't want it breaking this command.
Change-Id: I90c28cd21f3a5bc9f0155a88b50b13434fda4c5d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/838
Reviewed-by: glittershark <grfn@gws.fyi>
This does not work for ARGON2 hashes.
Change-Id: I1e070fa0ff17ef21632e94e6777da637deb6f54f
Reviewed-on: https://cl.tvl.fyi/c/depot/+/834
Reviewed-by: Kane York <rikingcoding@gmail.com>
Reviewed-by: BuildkiteCI
Tested-by: BuildkiteCI
This wrapper script correctly invokes slappasswd for generating ARGON2
hashes.
People without Nix will need to figure this out on their own.
Change-Id: I264715ba4932855b868bef70cc9102515b3d129e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/833
Tested-by: BuildkiteCI
Reviewed-by: BuildkiteCI
Reviewed-by: Kane York <rikingcoding@gmail.com>
This makes it possible to use {ARGON2} hashes instead of the current
salted SHA hashes, which is a much better idea.
Unfortunately the nixpkgs module does not have an option for
overridding the package used, so it is overlaid into the system
package set - this causes widespread rebuilds.
This is fine for us for now, but I have opened a PR upstream to add a
package option: https://github.com/NixOS/nixpkgs/pull/91963
Change-Id: Ib4be931d88e74b91566639f8656742cf096f6cc3
Reviewed-on: https://cl.tvl.fyi/c/depot/+/831
Reviewed-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Tested-by: BuildkiteCI
This enables support for the Argon2 password hashing mechanism in
OpenLDAP. Note that we also need to configure the LDAP module to load
this, so this change is not yet sufficient for actually using Argon2
hashes.
Change-Id: I151b854b777daa924b22224a43851432a88a2760
Reviewed-on: https://cl.tvl.fyi/c/depot/+/830
Reviewed-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Tested-by: BuildkiteCI
... and remove a package that doesn't exist anymore (at this location)
from the nixpkgs allowlist.
Change-Id: I663c84c387fb04bb3b47448132ad768ed5352474
Reviewed-on: https://cl.tvl.fyi/c/depot/+/829
Reviewed-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Tested-by: BuildkiteCI
This makes the code slightly more readable. For users that use editors
without semantic navigation, this also makes it easier to jump around
between items in the files.
I looked into whether a rustfmt setting exists for this, but
unfortunately the answer is currently no.
Change-Id: I37b19fa6ab038c71b924c45dbc12b298e660e8cf
Reviewed-on: https://cl.tvl.fyi/c/depot/+/827
Reviewed-by: BuildkiteCI
Reviewed-by: eta <eta@theta.eu.org>
Tested-by: BuildkiteCI
Before this change, besadii would skip further processing of meta refs (which happen for every CL metadata change), but it would still schedule a build by returning an update - which would then inevitably fail.
This change makes besadii skip meta refs the same way it skips non-depot builds, i.e. completely.
Move *on* from meta refs, do *not* collect $100.
Change-Id: I269d2299f4d3cb1f9c041da8c92fa00ae7794b38
Reviewed-on: https://cl.tvl.fyi/c/depot/+/825
Reviewed-by: eta <eta@theta.eu.org>
Reviewed-by: BuildkiteCI
Tested-by: BuildkiteCI
Refactor the caching policy for the Memo by evicting the elements that have been
the least-recently-accessed.
Python's heapq module default to a min-heap. By storing our heap elements
as (UnixTime, a), we can guarantee that when we call heappop, we will get the
element with the lowest UnixTime value in heap (i.e. the oldest). When we call
heappush, we use (time.time(), key) and these values -- by having the largest
UnixTime, will propogate to the bottom of the min-heap.
Bound the size of the memo by creating a BoundedQueue. Whenever we add elements
to the BoundedQueue, we remove the oldest elements. We use the BoundedQueue to
control the size of our dictionary that we're using to store our key-value pairs.
After hearing from a Jane Street recruiter, I decided to dust off some of the
DS&As knowledge. I found this article online, which outlines an example problem
called "Memo":
https://blog.janestreet.com/what-a-jane-street-dev-interview-is-like/
Here's part 1 of the solution in Python.
It's beautiful how State is just Reader that returns a tuple of (a, r) instead
of just a, allowing you to modify the environment (i.e. state).
```haskell
newtype Reader r a = Reader { runReader :: r -> a }
newtype State s a = State { runState :: s -> (a, s) }
```
Setting `authMode = "Registered"` prevents me from running the `register
<password>` command from inside of `bitlbee`, which I *believe* I need to
run...
I'm having trouble getting PAM auth to work, so I'm temporarily disabling it.
TIL that I can use the following to verify that PAM is properly setup for a
program (e.g. `bitlbee`).
```
pamtester -v bitlbee $(whoami) authenticate
```
...but despite this succeeding, I still cannot use the `identify` command in
`bitlbee` to successfully authenticate. It just tells me "Incorrect password"
even though I'm providing it the same password that I type when doing the
`pamtester` command from above.
Computers!
I'm still not entirely sure what bitlbee does, but I know this: I want as many
messengers in the same place as possible: IRC, Slack, Telegram. @tazjin tells me
that Bitlbee will help me get to the promised land. This is hopefully one step
of many in that direction.
Adds a Naersk-based build to check that this compiles, with a Lockfile
based on the ~2018 crate versions.
Change-Id: I0460a476d3b983fcf71e35e6b480f4a526118b58
Reviewed-on: https://cl.tvl.fyi/c/depot/+/803
Reviewed-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
Tested-by: BuildkiteCI
