[Fuego] [Fuego 1.2.1] Functional.bc test aborted : PROGRAM_BC is not defined

Bird, Timothy Tim.Bird at sony.com
Mon Dec 4 06:20:50 UTC 2017


> -----Original Message-----
> From: Dhinakar Kalyanasundaram on Thursday, November 30, 2017 11:39 PM
> Dear All,
> 
> In the latest Fuego v1.2.1 Functional.bc test gets aborted with the 'Fuego
> error reason: PROGRAM_BC is not defined.'
> 
> I have attached the log below. Please check.

We added dependency checking in lots of tests in the 1.2.1 release.
I'll comment below on the errors, but basically the test
could not find the program 'bc' on the target board.

Do you have 'bc' on your board?
 
> 
> ......
> 
> Started by user
> ha:AAAAlh+LCAAAAAAAAP9b85aBtbiIQTGjNKU4P08vOT+vOD8nVc83Py
> U1x6OyILUoJzMv2y+/JJUBAhiZGBgqihhk0NSjKDWzXb3RdlLBUSYGJk8GtpzUv
> PSSDB8G5tKinBIGIZ+sxLJE/ZzEvHT94JKizLx0a6BxUmjGOUNodHsLgAzOEgYu/
> dLi1CL9vNKcHACFIKlWvwAAAA==anonymous
> Building remotely on
> ha:AAAAph+LCAAAAAAAAP9b85aBtbiIQTGjNKU4P08vOT+vOD8nVc83P
> yU1x6OyILUoJzMv2y+/JJUBAhiZGBgqihhk0NSjKDWzXb3RdlLBUSYGJk8GtpzU
> vPSSDB8G5tKinBIGIZ+sxLJE/ZzEvHT94JKizLx0a6BxUmjGOUNodHsLgAyBEgYp/
> eT83ILSktQi/ZAgxzDd4MS8zJJK3ZDU4hIA7ynRAs8AAAA=TRAV-Sanity-
> Test in workspace /fuego-rw/buildzone
> [buildzone] $ /bin/sh -xe /tmp/hudson606150706127208992.sh
> + export Reboot=false
> + Reboot=false
> + export Rebuild=false
> + Rebuild=false
> + export Target_PreCleanup=true
> + Target_PreCleanup=true
> + export Target_PostCleanup=true
> + Target_PostCleanup=true
> + export TESTDIR=Functional.bc
> + TESTDIR=Functional.bc
> + export TESTSPEC=default
> + TESTSPEC=default
> + timeout --signal=9 6m /bin/bash /fuego-core/engine/scripts/main.sh
> Using nosyslogd.dist overlay
> ##### doing fuego phase: pre_test ########
> Logged in as user root
> 
> *** ABORTED ***
> 
> Fuego error reason: PROGRAM_BC is not defined. Make sure you use the
> correct overlays and specs for this test/benchmark.

This is not a great error message for this.  'assert_define' (the function
that aborted the test in this case) can take an argument for the message
to use if it decides to abort the test, and I think we should use one
in Functional.bc/fuego_test.sh to show a better message.

I'm away from my development machine at the moment.
Can you take a look at this and submit a patch with a better error message
for this?

> in signal_handler
> Teardown board link
> WARNING: program returned exit code
> Log evaluation may be invalid
> 
> Fuego error reason: '/fuego-rw/logs/Functional.bc/TRAV-Sanity-
> Test.default.1.1/testlog.txt' is missing.
> 
> ERROR: test_processing returned an error

This is a cascade* error, which we need to fix.  We aborted so early that the log file
was not created.  Fuego should detect this situation, and avoid this error
(and the rest of this log processing code, and additional cascade errors).
I'll add an issue for this and try to take a look at it when I get a chance.

* a cascade error is something that fails because something earlier failed.

> Going to check 2 fail cases for TRAV-Sanity-Test.default.Functional.bc
> grep: /fuego-rw/logs/Functional.bc/TRAV-Sanity-Test.default.1.1/testlog.txt:
> No such file or directory
> diff: /fuego-rw/logs/Functional.bc/TRAV-Sanity-
> Test.default.1.1/syslog.before.txt: No such file or directory
> No fail cases detected for TRAV-Sanity-Test.default.Functional.bc
> running python with
> PATH=/usr/local/bin:/usr/local/bin:/usr/bin:/bin:/usr/local/games:/usr/gam
> es
> WARNING: missing or faulty criteria.json file (/fuego-
> core/engine/tests/Functional.bc/criteria.json), looking for reference.log
> WARNING: No reference.log found. Using default criteria.
> No reference.json available
> ref={'test_sets': [{'test_cases': [{'status': 'FAIL', 'name': 'bc'}], 'name':
> 'default'}]}
> Applying criterion {'max_fail': 0, 'tguid': 'default'}
> Applying criterion {'max_fail': 0, 'tguid': 'Functional.bc'}
> Writing run data to /fuego-rw/logs/Functional.bc/TRAV-Sanity-
> Test.default.1.1/run.json
> reference.json not available
> WARNING: Unable to parse /fuego-rw/logs/Functional.bc/TRAV-System-
> Test.default.1.1/run.json
> Writing merged results to  /fuego-rw/logs/Functional.bc/results.json
> Writing flat results to /fuego-rw/logs/Functional.bc/flat_plot_data.txt
> chart_config.json not available or is wrong format, using default values
> Writing chart data to /fuego-rw/logs/Functional.bc/flot_chart_data.json
> ERROR: results did not satisfy the threshold
> Build step 'Execute shell' marked build as failure
> [description-setter] Description set:  <a
> href="/storm/userContent/fuego.logs/Functional.bc/TRAV-Sanity-
> Test.default.1.1/testlog.txt">testlog</a> <a
> href="/storm/userContent/fuego.logs/Functional.bc/TRAV-Sanity-
> Test.default.1.1/run.json">run.json</a> <a
> href="/storm/userContent/fuego.logs/Functional.bc/TRAV-Sanity-
> Test.default.1.1/consolelog.txt">fuegolog</a> <a
> href="/storm/userContent/fuego.logs/Functional.bc/TRAV-Sanity-
> Test.default.1.1/devlog.txt">devlog</a> <a
> href="/storm/userContent/fuego.logs/Functional.bc/TRAV-Sanity-
> Test.default.1.1/prolog.sh">prolog.sh</a>
> Finished: FAILURE
>

Thanks for the report!
 -- Tim



More information about the Fuego mailing list