Problem
-------
There's a bug in ipcsocket cleanup logic that effectively
tries to wipe out /tmp.
Workaround
----------
We should fix the underlying bug, but we can start by
explicitly configuring the temp directories ipcsocket uses.
Ref https://github.com/sbt/sbt/issues/6931
The `XDG_RUNTIME_DIR` environment variable will be used to choose the base directory in which the sockets will be created by **sbt**. This can help when issues such as #6777 appear. There are other related issues, such as #6101, [Metals issue #2235](https://github.com/scalameta/metals/issues/2235), [Che issue #18394](https://github.com/eclipse/che/issues/18394). Those are closed issues, but there are some cases in which the solution does not work (such as the case in #6777). Furthermore, the solution given seems more like a workaround than an actual fix.
**What causes this issue?**
At least in my case, the **ServerAlreadyBootingException** is thrown after **sbt** tries to create a Unix domain socket and fails. In my specific environment, I was not able to create the sockets in some directories because they were mounted on an NFS. This prevented me from using any automated sbt command on any of those directories (Metals uses sbt under the hood, and it was not able to start because of the ServerAlreadyBootingException), which in turn resulted in me not being able to use Metals on a project that was created on any of those directories.
**How is the issue solved in this PR?**
If the `XDG_RUNTIME_DIR` environment variable is set, then sockets will be created in a folder with a hashed name, inside the directory `XDG_RUNTIME_DIR/.sbt/`. If this variable is not set, everything works exactly the same as always.
Let's see an example:
1. My environment variable `XDG_RUNTIME_DIR` is set to /home/alonso/.runtime-dir`
2. I create my project in `/home/alonso/workspace/hello-world`
3. I run `sbt compile`
4. The socket is created in `/home/alonso/.runtime-dir/.sbt/sbt-socket102030405060/sbt-load.sock`. The number **102030405060** is a hash that is computed from the base directory of the project (it is actually the same hash that is produced in order to create sockets in Windows) and it will be different depending on the location of the project
5. The sbt command executed correctly, which would not have happened if the socket had been created in the `home/alonso/workspace` directory or any of its subdirectories (in this example, `/home/BlackDiamond/workspace` corresponds to a network file share)
Co-authored-by: Alonso Montero <lumontero@mobilize.net>
In order to start the sbt server we must use the sbt script because
it is the only way to load the .sbtopts and the .jvmopts file properly.
To do so the sbt script can pass a -Dsbt.script prop to the java server.
It is used in the NetworkClient to start the server, and it is replicated
in the BuildServerConnection file (.bsp/sbt.json).
The problem was that -client was different from --client, which
makes for a confusing user experience. So, this change makes
them the same and re-names -client to --java-client. The value
of this is that hopefully -client and --client being the same
feels more logical to users.
The intellij bsp integration launches sbt with the launcher and runs
-bsp. This doesn't work if sbt is not on intellij's path. To try and
work around this, we can add an option --sbt-launch-jar that is
recognized by the network client and will make it use the launch jar
rather than the sbt script to launch a new server if needed.
Made the bspConfig task dependendant on the bspConfig value.
Changed the bspConfig setting to use a attributeKey so we can use it in the server as well.
The new version provides an apple silicon arm64 jni implementation of
the ipcsocket api. It also adds a jni implementation for getting the max
socket length which is necessary because jna isn't supported on the
apple arm64 platform yet.
There isn't yet a version of the jna available that works with the new
apple silicon using arm64. To workaround this, we can use the jni
implementation by default on arm64 macs. If the user wants to force the
jni implementation for any supported platform, they can opt in with the
`sbt.ipcsocket.jni` system property and/or by setting the serverUseJni
setting.
Network client already supports the -bsp command (since
65ab7c94d0). This commit reworks the
BspClient.run method so that it delegates to the NetworkClient. The
advantage to doing it this way is that improvements to starting up the
sbt server by the thin client will automatically propagate to the -bsp
command. The way that it is implemented, all of the output generated
during server startup will be redirected to System.err which is useful
for debugging without messing up the bsp protocol, which relies on only
bsp messages being written to System.out.
The boot server socket was not working correctly when the sbt server was
started by the thin client. This was because it is necessary for us to
create a ConsoleTerminal in order for System.out and System.err to be
properly forwarded to the clients connected over the boot server socket.
As a result, if you started a server instance of sbt with the thin
client, you wouldn't see any output util you connected to the server.
The fix is to just make sure that we create a console terminal if sbt is
run as a subprocess.
When a batch command is run with the thin client, it logs an info
message that the command completed. This is unnecessary given that
completion is implied by the success or failure method that follows. It
made the output look a little different in the thin client vs the
console.
I saw a stacktrace when exiting sbtn on windows due to an interrupted
exception being thrown during thread joining. We only want to throw this
exception if we didn't successfully join the thread. I also noticed that
we would try to join the thread forever. There was supposed to be a
timelimit so that we would eventually stop blocking even if we were
unable to join the thread. The limit was set but not respected.
With sbtn, the system properties are passed in as regular command
arguments. We need to parse them before we call Terminal.withStreams or
else system properties like -Dsbt.color=false are ignored.
EvaluateTask was holding references to SafeState that could be quite
large. This was reported as #5992. In that project, I ran the `ci` task
and observed the OOM as reported. I took a heap dump prior to OOM and
got the retained size graph from visualvm (which took hours to compute).
The lastEvaluatedState was holding a reference to SafeState that was
1.7GB. The project max heap size was set to 2GB. Instead of using the
lastEvaluatedState, we can just use StandardMain.exchange.withState.
The cached instances of state were used for task cancellation and
completions. While it is possible that early on in booting
StandardMain.exchange.withState could return a null state, in practice
this won't happen because it is set early on during the sbt boot
commands.
After this change, I successfully ran the `ci` task in the #5992 issue
project with the same memory parameters as their ci config.
The ConsoleAppender formatEnabledInEnv field was being used both as an
indicator that ansi codes were supported and that color codes are
enabled. There are cases in which general ansi codes are not supported
but color codes are and these use cases need to be handled separately.
To make things more explicit, this commit adds isColorEnabled and
isAnsiSupported to the Terminal companion object so that we can be more
specific about what the requirements are (general ansi escape codes or
just colors). There are a few cases in ConsoleAppender itself where
formatEnabledInEnv was used to set flags for both color and ansi codes.
When that is the case, we use Terminal.isAnsiSupported because when that
is true, colors should at least work but there are terminals that
support color but not general ansi escape codes.
Eugene Yokota
Alonso Montero
dmitrii.naumenko
Philippus
David Francoeur
xuwei-k
Adrien Piquerez
Samuel CLARENC
Samuel CLARENC
Amina Adewusi
Ondra Pelech
Sam Halliday
João Ferreira
Guillaume Martres
Ethan Atkins
Erlend Hamnaberg