bitbox-bridge
bitbox-bridge copied to clipboard
digitalbitbox
Installation & Launch issue on Mac with M1
- MacBook with M1
- Mac OS 11.2
Issues
- Installation: Can install the BitBoxBridge via right-click > open.
- Without right-click > "open" he get's an "developer cannot be identified" error.
- After installation there is a "bitbox-bridge can't be opened" error that keeps poping up.
- Launching BitBox-Bridge: When performing normal double click on the bitbox-bridge executable the terminal shows the following error:
/opt/shiftcrypto/bitbox-bridge/bin/bitbox-bridge ; exit;
~ % /opt/shiftcrypto/bitbox-bridge/bin/bitbox-bridge ; exit;
zsh: killed /opt/shiftcrypto/bitbox-bridge/bin/bitbox-bridge
Saving session...
...saving history...truncating history files...
...completed.
Deleting expired sessions...none found.
- Launching BitBox-Bridge: Performing a right-click > open on the bitbox-bridge executable opens terminal and shows the following error message:
/opt/shiftcrypto/bitbox-bridge/bin/bitbox-bridge ; exit;
~ % /opt/shiftcrypto/bitbox-bridge/bin/bitbox-bridge ; exit;
Set RUST_LOG=<filter> to enable logging. Example RUST_LOG=debug
listening on http://127.0.0.1:8178
[2021-04-06T18:08:58Z INFO bitbox_bridge::usb] Notified!
thread 'main' panicked at 'error binding to 127.0.0.1:8178: error creating server listener: Address already in use (os error 48)', /tmp/cargo/registry/src/github.com-1ecc6299db9ec823/warp-0.2.1/src/server.rs:211:27
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
Saving session...
...copying shared history...
...saving history...truncating history files...
...completed.
- Checking localhost in the browser, the bridge is running despite this error
cc @x1ddos
It looks correctly installed. It runs in the background which explains why the port is already busy. Iirc there is a web address you can enter to see if it is running. Let me check
@NickeZ this link http://127.0.0.1:8178/?
Yeah after launching via right-click > open it's running, just didn't expect the terminal to show an error
It should be running even if you didn't "right-click -> open". It should be running in the background after the laptop has booted.
edit: yeah, that is the link to check if it is running
FTR I believe all these troubles are solved by producing signed and notarized binaries or macOS.
re-reading it I realized that this is a new ARM based macbook, so signing/notarizing wouldn't solve the issue. I thought M1 should be able to run x86 executables, but maybe not. We can compile it for M1, but I'll have a hard time testing. So we need to find a user willing to test for us.
signing/notarizing wouldn't solve the issue
why not? why would a CPU architecture affect signing and notarization?
M1 should be able to run x86 executables, but maybe not
it cannot run x86 natively but via their Rosetta 2 thing. but yes, we should compile the bridge for native M1 too and make it a universal binary that works for both architectures.
signing/notarizing wouldn't solve the issue
why not? why would a CPU architecture affect signing and notarization?
(Since I don't have apple aarch64 hardware I cannot replicate the issue, but I assume the issue "After installation there is a "bitbox-bridge can't be opened" error that keeps poping up." is unrelated to signing/notarization.
M1 should be able to run x86 executables, but maybe not
it cannot run x86 natively but via their Rosetta 2 thing. but yes, we should compile the bridge for native M1 too and make it a universal binary that works for both architectures.
Yeah, doesn't seem that hard to create a universal binary. The only problem I have is with testing it :/
On 4/12/21 3:09 PM, Niklas Dusenlund wrote:
signing/notarizing wouldn't solve the issue
why not? why would a CPU architecture affect signing and notarization?
(Since I don't have apple aarch64 hardware I cannot replicate the issue, but I assume the issue "After installation there is a "bitbox-bridge can't be opened" error that keeps poping up." is unrelated to signing/notarization.
I have access to a M1 mac mini here. A BitBoxApp built for x86 behaves in the same way. If unsigned or not notarized, I can't run it easily - need to right click > open first, same as here in this issue. When signed and notarized, I'm running it via the Rosetta 2 without issues.
M1 should be able to run x86 executables, but maybe not
it cannot run x86 natively but via their Rosetta 2 thing. but yes, we should compile the bridge for native M1 too and make it a universal binary that works for both architectures.
Yeah, doesn't seem that hard to create a universal binary. The only problem I have is with testing it :/
I have an M1. Maybe I should take this issue and send you some pull requests.