Describe the bug A clear and concise description of what the bug is.

It cannot find sccache because it is in a custom directory. A good environment variable to look for is $CARGO_HOME.

To Reproduce Steps to reproduce the behavior:

1. Run fleet run
2. Sccache is installed but it warns about not being installed

Expected behavior A clear and concise description of what you expected to happen.

It should find sccache in the custom cargo directory.

Screenshots If applicable, add screenshots to help explain your problem.

Desktop (please complete the following information):

• OS: Windows
• Version: 11

Additional context Add any other context about the problem here.

Building the fleet executable through Cargo (in the natural way (eg, git clone; cd fleet; cargo build)doesn't work. Even after installing sccache and lld, the builds are broken, because your config.toml expects sccache at C:\Users\potti\.cargo\bin.

cargo build

 cargo build                                                                 
error: failed to run `rustc` to learn about target-specific information

Caused by:
  could not execute process `C:\Users\potti\.cargo\bin\sccache rustc - --crate-name ___ --print=file-names -Clink-arg=-fuse-ld=lld -Zshare-generics=y --crate-type bin --crate-type rlib --crate-type dylib --crate-type cdylib --crate-type staticlib --crate-type proc-macro --print=sysroot --print=cfg` (never executed)

Caused by:
  No such file or directory (os error 2)

this is fixable by simply deleting that line in the config, or by deleting the config entirely.

if you want people to take this project more seriously, I strongly suggest making some kind of basic CI that tests the builds on "bare" ubuntu, alpine, & windows, by following your readme exactly.

you should list your dependencies in the README, too: AFAIK, they're:

ALL:

sccache

cargo install sccache

lld (windows, ubuntu)

windows

winget install llvm

ubuntu

sudo apt-get install lld-12

zld (macos)

brew install zld

Describe the bug The website is unavailable right now.

To Reproduce Either go to the website ( https://fleet.rs ) or this URL https://www.isitdownrightnow.com/fleet.rs.html

This repo gitignores .cargo/config - a file that contains general configuration for the project. If you don't gitignore it, you will constantly have changes in the .cargo/config as it is updating paths to work on that specific machine. Are there anyways to make it so we don't have to gitignore .cargo/config, but continue to use fleet?

Describe the bug Fleet runs the apt command to install clang and lld even on Arch Linux which is a non-Debian/Ubuntu based Linux distro. It's offical package manager is called pacman. Also it is kinda bad practice to call sudo explicitly by a program.

To Reproduce

curl -L get.fleet.rs | sh

Expected behavior I think fleet should ask the user to install clang and lld by themselves rather doing it by itself.

Desktop

• OS: Arch Linux

on a linux server, where I do not have sudo access:

error: linker /usr/bin/clang not found | = note: No such file or directory (os error 2)

error: could not compile rayon-core due to previous error warning: build failed, waiting for other jobs to finish... error: could not compile log due to previous error error: could not compile crossbeam-epoch due to previous error error: could not compile parking_lot_core due to previous error error: could not compile memchr due to previous error error: could not compile getrandom due to previous error error: could not compile crossbeam-utils due to previous error error: could not compile proc-macro-hack due to previous error error: could not compile proc-macro2 due to previous error error: could not compile serde_derive due to previous error error: could not compile libc due to previous error error: could not compile serde due to previous error error: could not compile syn due to previous error => You have not installed lld. Run sudo apt install lld. => You have not installed clang. Run sudo apt install clang.

I do have those 2 installed but is not in /usr/bin:

/usr/local/pace-apps/spack/packages/0.13/linux-rhel7-x86_64/gcc-4.8.5/intel-parallel-studio-cluster.2019.5-rde2uvgb2mq2x6obqaklxppmycfngjgm/compilers_and_libraries_2019.5.281/linux/bin/lld [jzhao399@atl1-1-03-004-19-l hnswlib-rs]$ which clang /usr/local/pace-apps/spack/packages/0.13/linux-rhel7-x86_64/gcc-4.8.5/intel-parallel-studio-cluster.2019.5-rde2uvgb2mq2x6obqaklxppmycfngjgm/compilers_and_libraries_2019.5.281/linux/bin/clang

why depend on a system level clang and lld? Think about those who wants to install via conda et.al.

Thanks, Jianshu

@suptejas really impressed with your rust skill and product development at large for a 16yr old. Really impressed with your products. I also started programming at age 13. You've done what I never had the opportunity to do that young. Keep it up and the fleet is great I use it daily.

Team,

After install and uninstall fleet, I always have the following error no matter what I compile:

error: failed to run rustc to learn about target-specific information

Caused by: could not execute process /Users/jianshuzhao/.cargo/bin/sccache rustc - --crate-name ___ --print=file-names --crate-type bin --crate-type rlib --crate-type dylib --crate-type cdylib --crate-type staticlib --crate-type proc-macro --print=sysroot --print=cfg (never executed)

Caused by: No such file or directory (os error 2)

I believe sccache is not required before. What happened to fleet and what configuration was changed?

This is really confusing and not acceptable.

Thanks,

Jianshu

After install on Mac(M1), get an error when fleet run:

main-2588b02f17a95c9e" "-Wl,-dead_strip" "-nodefaultlibs" "-fuse-ld=/usr/local/bin/zld"
  = note: clang: error: invalid linker name in argument '-fuse-ld=/usr/local/bin/zld'

and also the cargo command fail after installing fleet:

8b02f17a95c9e" "-Wl,-dead_strip" "-nodefaultlibs" "-fuse-ld=/usr/local/bin/zld"
  = note: clang: error: invalid linker name in argument '-fuse-ld=/usr/local/bin/zld'

It means both fleet and cargo is broken.

Is there any way to fix it? Thank you!

Describe the bug using fleet to build wezterm on a chromebook Linux penguin 5.10.114-16023-g49367ce3348c #1 SMP PREEMPT Mon May 30 19:31:09 PDT 2022 aarch64 GNU/Linux

debian 11 bullseye cargo 1.61.0 (a028ae4 2022-04-29) rustc 1.61.0 (fe5b13d68 2022-05-18)

To Reproduce Follow build steps here https://wezfurlong.org/wezterm/install/source.html

used fleet build -- --release and fleet build produced crash files, attached.

Expected behavior A rust build

The dump files say permission denied but I can't figure out what file.

Archive.zip (GitHub doesn't allow .toml uploads)

For what it's worth, cargo build --release didn't work either. So, I'm just submitting the crash report.

Describe the bug

      - name: Cache .cargo/bin
        uses: actions/cache@v3
        env:
          cache-name: cache-node-modules
        with:
          # npm cache files are stored in `~/.npm` on Linux/macOS
          path: ~/.cargo/bin
          key: ${{ runner.os }}-build-${{ env.cache-name }}-${{ hashFiles('**/cargo.lock') }}
          restore-keys: |
            ${{ runner.os }}-build-${{ env.cache-name }}-
            ${{ runner.os }}-build-
            ${{ runner.os }}-

To Reproduce Linux CI/CD

Expected behavior We are caching for cargo, not npm node modules.

Screenshots

Describe the bug The fleet installer panicks when installing the build tool in Windows 11.

To Reproduce Steps to reproduce the behavior: In an elevated PowerShell prompt in Windows 11:

#  % natalie-perret @ nppc in ~ [15:59:04]
$ $Env:RUST_BACKTRACE='full'

#  % natalie-perret @ nppc in ~ [15:59:13]
$ iwr -useb windows.fleet.rs | iex
Installing Fleet
thread 'main' panicked at 'called `Result::unwrap()` on an `Err` value: Error { kind: NotFound, message: "program not found" }', src\main.rs:29:10
stack backtrace:
   0:     0x7ff67776f70f - <unknown>
   1:     0x7ff677785c3a - <unknown>
   2:     0x7ff67776c089 - <unknown>
   3:     0x7ff677771c0b - <unknown>
   4:     0x7ff6777717fe - <unknown>
   5:     0x7ff677772201 - <unknown>
   6:     0x7ff6777720bd - <unknown>
   7:     0x7ff6777702c7 - <unknown>
   8:     0x7ff677771d99 - <unknown>
   9:     0x7ff677798c85 - <unknown>
  10:     0x7ff677798d23 - <unknown>
  11:     0x7ff67776189c - <unknown>
  12:     0x7ff677761b72 - <unknown>
  13:     0x7ff677762116 - <unknown>
  14:     0x7ff677761c1c - <unknown>
  15:     0x7ff677768aa9 - <unknown>
  16:     0x7ff677761c07 - <unknown>
  17:     0x7ff67778a344 - <unknown>
  18:     0x7ffdeb79244d - BaseThreadInitThunk
  19:     0x7ffdec92df78 - RtlUserThreadStart
Successfully Installed fleet

Expected behavior It ain't supposed to panic.

Screenshots If applicable, add screenshots to help explain your problem.

Desktop (please complete the following information):

Copyright (c) Microsoft Corporation.

https://aka.ms/powershell
Type 'help' to get help.

   A new PowerShell stable release is available: v7.2.7
   Upgrade now, or check out the release page at:
     https://aka.ms/PowerShell-Release?tag=v7.2.7


 lllllllllllllll   lllllllllllllll  natalie-perret@NPPC
 lllllllllllllll   lllllllllllllll  -------------------
 lllllllllllllll   lllllllllllllll  OS: Windows 11 Pro [64-bit]
 lllllllllllllll   lllllllllllllll  Host: Dell Inc. Precision 5550
 lllllllllllllll   lllllllllllllll  Kernel: 10.0.22621.0
 lllllllllllllll   lllllllllllllll  Motherboard: Dell Inc. 0V6K79
 lllllllllllllll   lllllllllllllll  Uptime: 7 minutes
                                    Packages: 128 (choco), 2 (scoop)
 lllllllllllllll   lllllllllllllll  Shell: PowerShell v7.2.6
 lllllllllllllll   lllllllllllllll  Resolution: 1280x720, 1280x720
 lllllllllllllll   lllllllllllllll  Terminal: Windows Terminal
 lllllllllllllll   lllllllllllllll  CPU: Intel(R) Core(TM) i7-10850H CPU @ 2.70GHz
 lllllllllllllll   lllllllllllllll  GPU: Intel(R) UHD Graphics
 lllllllllllllll   lllllllllllllll  GPU: NVIDIA Quadro T1000
 lllllllllllllll   lllllllllllllll  Memory: 12.73 GiB / 31.75 GiB (40%)
                                    Disk (C:): 345 GiB / 475 GiB (72%)





Loading personal and system profiles took 2023ms.

Additional context Add any other context about the problem here.

When I try to compile I get it: error: incorrect value `...` for codegen option `split-debuginfo` - one of supported split-debuginfo modes (`off`, `packed`, or `unpacked`) was expected

Describe the bug

Just wanted to try fleet out, but getting an error even though I'm using nightly:

📝 Generated Fleet Config
🚀 Fleet is ready!
error: failed to run `rustc` to learn about target-specific information

Caused by:
  process didn't exit successfully: `rustc - --crate-name ___ --print=file-names -C -Zshare-generics=y -Csplit-debuginfo=unpacked -C -Zshare-generics=y -Csplit-debuginfo=unpacked --crate-type bin --crate-type rlib --crate-type dylib --crate-type cdylib --crate-type staticlib --crate-type proc-macro --print=sysroot --print=cfg` (exit status: 1)
  --- stderr
  error: unknown codegen option: `-Zshare-generics`

To Reproduce fleet build

Info:

• OS: macOS 10.15
• rustc: rustc 1.64.0-nightly (4d6d601c8 2022-07-26)
• Fleet: 8dce51c

Describe the bug NApi.rs release build failed because of https://github.com/dimensionhq/fleet/blob/9855e5fb9ce8d369e978386b69cefa7f0b13e103/src/core/config/cargo.rs#L113

I found this issue in below discussion (additional context)

Desktop (please complete the following information):

• OS: MacOS

Additional context Add any other context about the problem here.

• napi-rs/napi-rs#1246

https://github.com/dimensionhq/fleet/blob/974d781cc4b09a6617cc7bebedb22dbb48984cf7/installer/src/main.rs#L22

This code just updated my nightly rustc, while I already have a nightly rustc at a specific day. Is it possible to reuse the existing nightly rustc if there is one?