[yocto] QA notification for completed autobuilder build (yocto-2.8_M1.rc2)

richard.purdie at linuxfoundation.org richard.purdie at linuxfoundation.org
Wed Jun 12 01:04:32 PDT 2019


On Wed, 2019-06-12 at 00:51 +0000, Jain, Sangeeta wrote:
> On Tue, 2019-06-11 at 03:32 +0000, pokybuild at debian9-ty-1.yocto.io
> > wrote:
> > > A build flagged for QA (yocto-2.8_M1.rc2) was completed on the
> > > autobuilder and is available at:
> > > 
> > > 
> > >     https://autobuilder.yocto.io/pub/releases/yocto-2.8_M1.rc2
> > > 
> > 
> > Some notes for QA:
> Thanks Richard for sharing this.
> > There was a failure in this build in qemux86 oe-selftest. It looks
> > to be a race issue where qemu was shutting down. This needs a bug
> > and looking into but is a minor issue which wouldn't block the
> > milestone release or further QA.
> 
> We are analysing the failures and will file the bugs as suggested.

Just for reference I did file some of the bugs.

> > There are three timeouts in ptest, acl, bluez5 (#13366) and mdadm
> > (#13368).
> > These are also not going to block the milestone release. The
> > bluez5 issue is gcc9 related, mdadm was exposed by wider testing in
> > master. The
> > acl issue needs a bug opening and investigation as that is a
> > regression.
> 
> Looks like some disconnect here. In test result report, no timeout
> issue reported for acl ptest.

https://autobuilder.yocto.io/pub/releases/yocto-2.8_M1.rc2/testresults/testresult-report.txt

acl                          | 184         | 9        | 0         | 302 T

I hadn't spotted that dbus-test also has a timeout :(.

This timeout is also in the testresults file:

ERROR: Exit status is 9\nDURATION: 302\nTIMEOUT: /usr/lib/acl/ptest\n

in:

https://autobuilder.yocto.io/pub/releases/yocto-2.8_M1.rc2/testresults/qemux86-64-ptest/testresults.json

> We are referring to test results at yocto-testresults under tag "
> master/54573-g9910a3631ce953b8dd8f3d57d6e122c7fd8cb462/1"

You are right, there is no mention of this in the log at:

http://git.yoctoproject.org/cgit.cgi/yocto-testresults/tree/runtime/poky/qemux86-64/core-image-sato-sdk-ptest/ptest-acl.log?id=56b1caa6044ce30e92a1895eae48d3935562b627
or
http://git.yoctoproject.org/cgit.cgi/yocto-testresults/commit/runtime/poky/qemux86-64/core-image-sato-sdk-ptest/ptest-acl.log?id=ac623e13e3bc3a348b60fe3b884e6ae567b8230a

I suspect this is due to the information being filtered out as part of
the log processing. We probably shouldn't do that.

Looking at the individual logs in https://autobuilder.yocto.io/pub/releases/yocto-2.8_M1.rc2/testresults/qemux86-64-ptest/testresults.json, the timeout information looks to be truncated off there too.

I do however note that:

            "ptestresult.sections": {
                "acl": {
                    "duration": "302",
                    "exitcode": "9",
                    "timeout": true
                },

is in http://git.yoctoproject.org/cgit.cgi/yocto-testresults/tree/runtime/poky/qemux86-64/core-image-sato-sdk-ptest/testresults.json?id=56b1caa6044ce30e92a1895eae48d3935562b627
so the timeout is recorded in the results.

Regards,

Richard




More information about the yocto mailing list