Conversation
|
I'm not sure why Travis is failing here, but this looks good to me. Thanks! |
|
@rubyist Travis is failing, because of one of two reasons (or both):
|
|
Btw. travis is failing for some pull requests due to this two issues. Once the fixes are upstream, the PR tests can be rerun. |
|
@michael-k are you experiencing tests hanging locally? So far I am unable to reproduce this outside of Travis. Looking forward to seeing what you found in the transfer queue code. While it's definitely bad form (and I will merge that PR, too), I don't think the defered closes are relevant outside of that particular SSH PR (#378), which toggles mutex states inside of a |
|
I experienced the
Nope and that was a big problem. Endless adjustments and my own travis tests (#393 would have been helpful) until I found the lock. Upside: I know the test environment better. ;) But it is possible to reproduce it locally. See #396 for details (and fix). |
No description provided.