bitbake exit codes?

bitbake exit codes?

[Steffen Sledz]

While setting up Jenkins/Hudson jobs for the testing-release-2011.03 branch based on the testscript.sh from the contrib subdir i hit the following problem.

bitbake (1.10.2) finally reports "NOTE: Task Summary: Attempted 1217 tasks of which 0 didn't need to be rerun and 0 failed.". So it seems that the build was successful. But bitbake did not return 0 but 1. So Jenkins marks the build as failed.

I guess the exit code 1 is coming from some QA Issues. How do we handle that if if my guess is right?

Do we mark such a build run as successful in our test matrix?

Has bitbake a defined set of exit codes that we can evaluate in the testscript.sh?

Steffen

-- 
DResearch Fahrzeugelektronik GmbH
Otto-Schmirgal-Str. 3, 10319 Berlin, Germany
Tel: +49 30 515932-237 mailto:sledz@DResearch.de
Fax: +49 30 515932-299
Geschäftsführer: Dr. Michael Weber, Werner Mögle;
Amtsgericht Berlin Charlottenburg; HRB 130120 B;
Ust.-IDNr. DE273952058

Re: bitbake exit codes?

[Frans Meulenbroeks]

2011/2/14 Steffen Sledz <sledz@dresearch.de>:
> While setting up Jenkins/Hudson jobs for the testing-release-2011.03 branch based on the testscript.sh from the contrib subdir i hit the following problem.
>
> bitbake (1.10.2) finally reports "NOTE: Task Summary: Attempted 1217 tasks of which 0 didn't need to be rerun and 0 failed.". So it seems that the build was successful. But bitbake did not return 0 but 1. So Jenkins marks the build as failed.
>
> I guess the exit code 1 is coming from some QA Issues. How do we handle that if if my guess is right?
>
> Do we mark such a build run as successful in our test matrix?
>
> Has bitbake a defined set of exit codes that we can evaluate in the testscript.sh?
>
> Steffen
>
>
There have been a few threads on this in the past.
One of the ways (and maybe the best one) to get around this is to fix
the QA errors.

Frans

Re: bitbake exit codes? (testing-release-2011.03 build results)

[Steffen Sledz]

Am 14.02.2011 09:11, schrieb Frans Meulenbroeks:
> 2011/2/14 Steffen Sledz <sledz@dresearch.de>:
>> While setting up Jenkins/Hudson jobs for the testing-release-2011.03 branch based on the testscript.sh from the contrib subdir i hit the following problem.
>>
>> bitbake (1.10.2) finally reports "NOTE: Task Summary: Attempted 1217 tasks of which 0 didn't need to be rerun and 0 failed.". So it seems that the build was successful. But bitbake did not return 0 but 1. So Jenkins marks the build as failed.
>>
>> I guess the exit code 1 is coming from some QA Issues. How do we handle that if if my guess is right?
>>
>> Do we mark such a build run as successful in our test matrix?
>>
>> Has bitbake a defined set of exit codes that we can evaluate in the testscript.sh?
>>
>> Steffen
>
> There have been a few threads on this in the past.
> One of the ways (and maybe the best one) to get around this is to fix
> the QA errors.

Of course. You are right.

But for the moment the question is: "Do we mark such a builds as successful in our test matrix or not?"

Steffen

-- 
DResearch Fahrzeugelektronik GmbH
Otto-Schmirgal-Str. 3, 10319 Berlin, Germany
Tel: +49 30 515932-237 mailto:sledz@DResearch.de
Fax: +49 30 515932-299
Geschäftsführer: Dr. Michael Weber, Werner Mögle;
Amtsgericht Berlin Charlottenburg; HRB 130120 B;
Ust.-IDNr. DE273952058

Re: bitbake exit codes? (testing-release-2011.03 build results)

[Frans Meulenbroeks]

2011/2/14 Steffen Sledz <sledz@dresearch.de>:
> Am 14.02.2011 09:11, schrieb Frans Meulenbroeks:
>> 2011/2/14 Steffen Sledz <sledz@dresearch.de>:
>>> While setting up Jenkins/Hudson jobs for the testing-release-2011.03 branch based on the testscript.sh from the contrib subdir i hit the following problem.
>>>
>>> bitbake (1.10.2) finally reports "NOTE: Task Summary: Attempted 1217 tasks of which 0 didn't need to be rerun and 0 failed.". So it seems that the build was successful. But bitbake did not return 0 but 1. So Jenkins marks the build as failed.
>>>
>>> I guess the exit code 1 is coming from some QA Issues. How do we handle that if if my guess is right?
>>>
>>> Do we mark such a build run as successful in our test matrix?
>>>
>>> Has bitbake a defined set of exit codes that we can evaluate in the testscript.sh?
>>>
>>> Steffen
>>
>> There have been a few threads on this in the past.
>> One of the ways (and maybe the best one) to get around this is to fix
>> the QA errors.
>
> Of course. You are right.
>
> But for the moment the question is: "Do we mark such a builds as successful in our test matrix or not?"
>
> Steffen
>

Personally I'm inclined to say no. Marking them as unsuccessful
increases the chance of getting them fixed.
Then again others may think differently & I don't know exactly how
many QA errors we're talking about.
(I seem to recall that my console-image build had four, for cups I saw
a patch, binutils is being looked at or maybe already resolved by
Khem, no idea about the other two).

Best regards, Frans.

Re: bitbake exit codes? (testing-release-2011.03 build results)

[Frans Meulenbroeks]

Got an error when sending the message below. Not sure if it made it to
the list (gmane does not have it yet), so resending.
Apologies if you get this twice.

Frans

2011/2/14 Frans Meulenbroeks <fransmeulenbroeks@gmail.com>:
> 2011/2/14 Steffen Sledz <sledz@dresearch.de>:
>> Am 14.02.2011 09:11, schrieb Frans Meulenbroeks:
>>> 2011/2/14 Steffen Sledz <sledz@dresearch.de>:
>>>> While setting up Jenkins/Hudson jobs for the testing-release-2011.03 branch based on the testscript.sh from the contrib subdir i hit the following problem.
>>>>
>>>> bitbake (1.10.2) finally reports "NOTE: Task Summary: Attempted 1217 tasks of which 0 didn't need to be rerun and 0 failed.". So it seems that the build was successful. But bitbake did not return 0 but 1. So Jenkins marks the build as failed.
>>>>
>>>> I guess the exit code 1 is coming from some QA Issues. How do we handle that if if my guess is right?
>>>>
>>>> Do we mark such a build run as successful in our test matrix?
>>>>
>>>> Has bitbake a defined set of exit codes that we can evaluate in the testscript.sh?
>>>>
>>>> Steffen
>>>
>>> There have been a few threads on this in the past.
>>> One of the ways (and maybe the best one) to get around this is to fix
>>> the QA errors.
>>
>> Of course. You are right.
>>
>> But for the moment the question is: "Do we mark such a builds as successful in our test matrix or not?"
>>
>> Steffen
>>
>
> Personally I'm inclined to say no. Marking them as unsuccessful
> increases the chance of getting them fixed.
> Then again others may think differently & I don't know exactly how
> many QA errors we're talking about.
> (I seem to recall that my console-image build had four, for cups I saw
> a patch, binutils is being looked at or maybe already resolved by
> Khem, no idea about the other two).
>
> Best regards, Frans.
>

Re: bitbake exit codes? (testing-release-2011.03 build results)

[Tom Rini]

On 02/14/2011 03:39 AM, Steffen Sledz wrote:
> Am 14.02.2011 09:11, schrieb Frans Meulenbroeks:
>> 2011/2/14 Steffen Sledz<sledz@dresearch.de>:
>>> While setting up Jenkins/Hudson jobs for the testing-release-2011.03 branch based on the testscript.sh from the contrib subdir i hit the following problem.
>>>
>>> bitbake (1.10.2) finally reports "NOTE: Task Summary: Attempted 1217 tasks of which 0 didn't need to be rerun and 0 failed.". So it seems that the build was successful. But bitbake did not return 0 but 1. So Jenkins marks the build as failed.
>>>
>>> I guess the exit code 1 is coming from some QA Issues. How do we handle that if if my guess is right?
>>>
>>> Do we mark such a build run as successful in our test matrix?
>>>
>>> Has bitbake a defined set of exit codes that we can evaluate in the testscript.sh?
>>>
>>> Steffen
>>
>> There have been a few threads on this in the past.
>> One of the ways (and maybe the best one) to get around this is to fix
>> the QA errors.
>
> Of course. You are right.
>
> But for the moment the question is: "Do we mark such a builds as successful in our test matrix or not?"

Well, as the code stands today (post tag) the change to make those 
failures kill the build have been merged.

-- 
Tom Rini
Mentor Graphics Corporation