can reproduce this with 1.61.105 - downgrade to 1.58.2 as temp. measure, this one doesn’t exhibit crashes.
E Go : panic: java.lang.NoSuchMethodError: no non-static method "Lcom/tailscale/ipn/App;.getSyspolicyStringValue(Ljava/lang/String;)Ljava/lang/String;"
E Go :
E Go : goroutine 22 [running]:
E Go : github.com/tailscale/tailscale-android/cmd/jni.GetMethodID(0x7093a57110, 0xd, {0x6f54ce588c?, 0x4000cccbd8?}, {0x6f54cfff49, 0x26})
E Go : /home/vagrant/build/com.tailscale.ipn/cmd/jni/jni.go:451 +0x114
NoSuchMethodError on App getSyspolicyStringValue … reads like a compile+packaging issue originating from f-droid? maybe someone noted, as I see a 1.64 being in the build pipeline - https://monitor.f-droid.org/builds/log/com.tailscale.ipn/202
you’d see much more error reports in tailscales issue tracker or play store site if this wouldn’t be due to f-droid.
Edit: there’s an explanation by tailscale folk - https://github.com/tailscale/tailscale/issues/11719#issuecomment-2054073651 - the gist of it being: f-droid did react to version-tags and made releases, that tailscale themselves deemed unstable and didn’t yet publish to the play store. Interesting case for package distributors