Arpad Borsos
Arpad Borsos
We will be looking into it in the near future, yes.
Another NOTE: we don’t depend really depend on "external dependencies", so it might not be necessary to run this on a weekly cadance.
Capturing crashes currently only works when the executable (and crash handler) are running natively, and not when run through rosetta.
Yes universal binaries should work well. We don’t have plans to support Rosetta. As times moves on, demand will be lower and lower considering that people move to universal/native binaries...
As mentioned before, we currently have no plans to make crash reporting on Rosetta emulation work.
> The issue however is the created file doesn't have write permissions open to every user on the system. Read permissions are open to every user not write. > We...
crashpad raises a fatal error when starting its handler fails: https://github.com/getsentry/crashpad/blob/c5c78676424d13589390a36cc271a5ec9b3b0004/util/posix/double_fork_and_exec.cc#L118-L131 I think it should be possible to build the handler as a completely static executable, but I must admit...
We plan to expose a method to manually capture a minidump. But I don’t think we detect hangs right now.
I have no idea to be honest, will have to look at all this myself first.
> a mini dump "trumps" all other exception information? I think it does. Essentially the minidump will create an `event` internally, and the rest of the envelope is only used...