Re: [Question] Bug in SdkHarness

2021-07-15 Thread Miguel Hernández Sandoval
I hadn't seen it, but it seems that the server is not ready the second time it runs the tests. I'll take a look into it. On Thu, Jul 15, 2021 at 1:11 PM Luke Cwik wrote: > Were you able to see in the logs that the server was not yet > ready/launched when the client was trying to connect? > > On

Re: [Question] Bug in SdkHarness

2021-07-15 Thread Miguel Hernández Sandoval
Yes, I was able to reproduce the error locally and debugged it, but haven't found anything. On Wed, Jul 14, 2021 at 4:38 PM Luke Cwik wrote: > Repro locally and run with a Python debugger? > > On Wed, Jul 14, 2021 at 1:46 PM Miguel Hernández Sandoval < > rogelio.hernan...@wizeline.com> wrote: >

Re: [Question] Bug in SdkHarness

2021-07-14 Thread Miguel Hernández Sandoval
I think it is, but not sure what's causing it. Here's also the latest failure [1] of this kind. [1] https://github.com/apache/beam/runs/3063305519?check_suite_focus=true#step:6:358 On Wed, Jul 14, 2021 at 3:33 PM Luke Cwik wrote: > Is this a timing issue where the server isn't yet ready to

[Question] Bug in SdkHarness

2021-07-14 Thread Miguel Hernández Sandoval
Hi team, I am currently working on solving this bug [1]. The ticket describes a grpc.FutureTimeoutError in the SdkHarness.__init__ and it seems that it gets stuck in this line [2] causing the timeout and making some tests fail. Here [3] how the tests were run and the results. I was wondering if