EPICS Controls Argonne National Laboratory

Experimental Physics and
Industrial Control System

2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  <20192020  2021  2022  2023  2024  Index 2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  <20192020  2021  2022  2023  2024 
<== Date ==> <== Thread ==>

Subject: RE: EPICS 7 build failures on Appveyor/Windows
From: Freddie Akeroyd - UKRI STFC via Core-talk <[email protected]>
To: Michael Davidsaver <[email protected]>, Ralph Lange <[email protected]>, "EPICS core-talk ([email protected])" <[email protected]>
Date: Mon, 23 Sep 2019 11:10:32 +0000
I'm happy to take a look at this

Freddie

-----Original Message-----
From: Core-talk <[email protected]> On Behalf Of Michael Davidsaver via Core-talk
Sent: 21 September 2019 01:43
To: Ralph Lange <[email protected]>; EPICS Core-Talk <[email protected]>
Subject: Re: EPICS 7 build failures on Appveyor/Windows

> On Fri, 20 Sep 2019 at 14:28, Michael Davidsaver <[email protected] <mailto:[email protected]>> wrote:
>     Any hint at where/why these hang?  Why not at APS as well?  Troubleshooting stuck tests in CI is hard enough on Linux.  I have no idea how to start on Windows.

I have a couple of ideas.

Keep all the .tap files as an artifact file in the hope that the tap parser is hiding something useful (like a stack trace).

https://www.appveyor.com/docs/packaging-artifacts/

It also seems that there is a way to pause the build and allow interactive inspection via RDP?

https://www.appveyor.com/docs/how-to/rdp-to-build-worker/


Unfortunately I haven't been able to try either as my appveyor builds seem to be perpetually Queued now, and my windows vm has expired.  I'm done with M$ for today...


If would be excellent if someone can run the unit tests locally and post a stack trace from a hung or crashing test.


On 9/20/19 6:06 AM, Ralph Lange via Core-talk wrote:
> On Fri, 20 Sep 2019 at 14:28, Michael Davidsaver <[email protected] <mailto:[email protected]>> wrote:
> 
> 
>     On Fri, Sep 20, 2019, 4:53 AM Ralph Lange via Core-talk <[email protected] <mailto:[email protected]>> wrote:
> 
>         So... the winners are
>         Commit c0a04eae <https://github.com/epics-base/epics-base/commit/c0a04eae219d3f3f8cf9ce9c1507cb31a6149034> was hanging the mingw builds (build 423 <https://ci.appveyor.com/project/epics-base/epics-base/builds/27544080>)
>         Commit d5899699 
> <https://github.com/epics-base/epics-base/commit/d58996992de2699354a4a
> 61dc2acb7126476392c> was hanging one of the VS builds (build 454 
> <https://ci.appveyor.com/project/epics-base/epics-base/builds/27549354
> >)
> 
> 
>     Any hint at where/why these hang?  Why not at APS as well?  Troubleshooting stuck tests in CI is hard enough on Linux.  I have no idea how to start on Windows.
> 
> 
> Well, looking at the build log, one thing is a clear failure in modules/libCom/test:
> 
> 8292 <https://ci.appveyor.com/project/epics-base/epics-base/builds/27544080/job/4h7liuuh38dta380#L8292>epicsStackTraceTest.t ......
> 8293 
> <https://ci.appveyor.com/project/epics-base/epics-base/builds/27544080
> /job/4h7liuuh38dta380#L8293>Failed 1/5 subtests
> 8294 
> <https://ci.appveyor.com/project/epics-base/epics-base/builds/27544080
> /job/4h7liuuh38dta380#L8294>(less 3 skipped subtests: 1 okay)
> 8295 
> <https://ci.appveyor.com/project/epics-base/epics-base/builds/27544080
> /job/4h7liuuh38dta380#L8295>ipAddrToAsciiTest.t ........ ok
> 8296 
> <https://ci.appveyor.com/project/epics-base/epics-base/builds/27544080
> /job/4h7liuuh38dta380#L8296>osiSockTest.t .............. ok
> 8297 
> <https://ci.appveyor.com/project/epics-base/epics-base/builds/27544080
> /job/4h7liuuh38dta380#L8297>
> 8298 
> <https://ci.appveyor.com/project/epics-base/epics-base/builds/27544080
> /job/4h7liuuh38dta380#L8298>Test Summary Report
> 8299 
> <https://ci.appveyor.com/project/epics-base/epics-base/builds/27544080
> /job/4h7liuuh38dta380#L8299>-------------------
> 8300 
> <https://ci.appveyor.com/project/epics-base/epics-base/builds/27544080
> /job/4h7liuuh38dta380#L8300>epicsStackTraceTest.t (Wstat: 0 Tests: 5 
> Failed: 1)
> 8301 
> <https://ci.appveyor.com/project/epics-base/epics-base/builds/27544080
> /job/4h7liuuh38dta380#L8301>Failed test: 5
> 8302 
> <https://ci.appveyor.com/project/epics-base/epics-base/builds/27544080
> /job/4h7liuuh38dta380#L8302>Files=42, Tests=4116, 190 wallclock secs ( 
> 0.19 usr + 0.09 sys = 0.28 CPU)
> 8303 
> <https://ci.appveyor.com/project/epics-base/epics-base/builds/27544080
> /job/4h7liuuh38dta380#L8303>Result: FAIL
> 8304 <https://ci.appveyor.com/project/epics-base/epics-base/builds/27544080/job/4h7liuuh38dta380#L8304>Failed 1/42 test programs. 1/4116 subtests failed.
> 
> and then I see further down
> 
> 8486 
> <https://ci.appveyor.com/project/epics-base/epics-base/builds/27544080
> /job/4h7liuuh38dta380#L8486>mingw32-make -C std/rec runtests
> 
> without ever seen make actually descend into that directory. Looks fishy.
> 
> Cheers,
> ~Ralph
> 



References:
EPICS 7 build failures on Appveyor/Windows Ralph Lange via Core-talk
Re: EPICS 7 build failures on Appveyor/Windows Michael Davidsaver via Core-talk
Re: EPICS 7 build failures on Appveyor/Windows Ralph Lange via Core-talk
Re: EPICS 7 build failures on Appveyor/Windows Michael Davidsaver via Core-talk

Navigate by Date:
Prev: Re: How to resolve git merge conflict in submodules? Zimoch Dirk (PSI) via Core-talk
Next: Re: [Merge] ~dirk.zimoch/epics-base:iocLogClientFixesTry2 into epics-base:7.0 mdavidsaver via Core-talk
Index: 2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  <20192020  2021  2022  2023  2024 
Navigate by Thread:
Prev: Re: EPICS 7 build failures on Appveyor/Windows Michael Davidsaver via Core-talk
Next: [Merge] ~epics-core/epics-base/+git/Com:errlog2syslog into epics-base:7.0 mdavidsaver via Core-talk
Index: 2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  <20192020  2021  2022  2023  2024 
ANJ, 23 Sep 2019 Valid HTML 4.01! · Home · News · About · Base · Modules · Extensions · Distributions · Download ·
· Search · EPICS V4 · IRMIS · Talk · Bugs · Documents · Links · Licensing ·