Constantine Peresypkin
Constantine Peresypkin
Nope it's a success, and as far as I can see the method is there. 😆
Pretty strange behavior, it looks like it depends on a system. One test fails another one succeeds. I cannot reproduce it now on my dev machine (i.e. code from `master`...
Aha, here we go. Works: ``` Mitmproxy: 9.0.0.dev (+86, commit 57c71ea) Python: 3.9.14 OpenSSL: OpenSSL 3.0.5 5 Jul 2022 Platform: Linux-5.15.0-46-generic-x86_64-with-glibc2.31 ``` Fails with `GnuTLS recv error`: ``` Mitmproxy: 9.0.0.dev...
Okay, changed it to match on both, still the same behavior. KVM guest - fails, KVM host - works. There's something fishy on connection close: Works: ``` 17:14:03.444624 IP my-host.51586...
> > Nope it's a success, and as far as I can see the method is there. laughing > > That would be very surprising - we use PyOpenSSL, and...
The only difference right now is that it works on `Ubuntu` host and fails on the same `Ubuntu` but as kvm guest. While all other stuff is the same right...
Okay it looks like `RENAME_WHITEOUT` is out of the question right now, till the bind mount/CLONE_NEWNS is implemented
Tests added. Should be ok. @hbhasker @avagin
@avagin I can add to other fs too, if that's needed. I can also discuss adding `RENAME_WHITEOUT` to all of them, if needed, but it looks like MKNOD capability is...
Sorry for the delay, will rework it and resubmit with the proposed changes