Customize ADB commands timeout

[gaetanlgce]

Hi,

On old devices, ADB invocation can be quite long to be executed, more than 5 seconds.
It could be useful to allow overriding the value of 5 seconds. (cf issue #5).

Currently, I edit the dist version of the file (https://github.com/ionic-team/native-run/blob/develop/src/android/utils/adb.ts#L469) in my node_modules to force another value of timeout.

Is there a proper way to pass a param for that. I don't use native-run directly but I use the ionic/cordova stack that use native-run. Maybe an environment variable ?

[me-to-you93]

Same problem here, would be interested by this option !

[markemer]

I'll look into this - seems like we could do this in a non-breaking way.

I'm not sure when I can get to it - but I'll put it on my radar.

If you are interested in helping I'd be happy to review any PR you'd post.

[Dexus]

To bad that this never found its way to the code...

[piotr-cz]

The timeout argument would have to be read here:

native-run/src/android/run.ts

Line 35 in a6e5605

export async function run(args: readonly string[]): Promise<void> {

and passed all the way to here:

native-run/src/android/utils/adb.ts

Line 358 in a6e5605

export async function execAdb(sdk: SDK, args: string[], options: ExecADBOptions = {}): Promise<string> {

[piotr-cz]

For me, it looks like that:

• the startActivity function executes ADB command which longer than 5s
• the execAdb function times out
• retrying fails, as the first ADB command is still executing
• app is deployed and started on device
• native-run exits with an error

npx capacitor run --live-reload --port 8100
...
√ Running Gradle build in 4.87s
× Deploying app-debug.apk to Pixel_9_API_35 - failed!
[error] Selected hardware device emulator-5554
        Installing xxx\android\app\build\outputs\apk\debug\app-debug.apk...
        Starting application activity xxx/xxx.MainActivity...
        ADBs is unresponsive after 5000ms, killing server and retrying...
        Error: Command failed: C:\Users\xxx\AppData\Local\Android\Sdk\platform-tools\adb -s emulator-5554 shell am
        start -W -n xxx/xxx.MainActivity
        adb.exe: device offline

        at ChildProcess.exithandler (node:child_process:422:12)
        at ChildProcess.emit (node:events:517:28)
        at maybeClose (node:internal/child_process:1098:16)
        at ChildProcess._handle.onexit (node:internal/child_process:303:5)

[piotr-cz]

I prepared PR #393 that increases timeout to 15_000 ms, because it works for my case.

@gaetanlgce @me-to-you93 @Dexus please let me know in above PR which timeout duration works for you

[Dexus]

I have patched it my own thanks to patch-package so I was able to set it to 30_000 that was working for me

[thoro23]

We have the same problem, the starting of the activity takes longer than 5s when deploying to an emulator. When deploying to a device it takes less than 5s and works fine.

For now we patch it with patch-package like @Dexus suggested but a proper fix would be preferable.

[xEverth]

Device offline and device still authorizing errors from ADB should be caught and retried on by native-run - increasing the timeout via environment variable or patch to startActivity just moves the issue slightly further without addressing it.
Also native-run seems a little too trigger-happy with adb kill-server currently