Does pharo build not retry anymore?

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
10 messages Options
Reply | Threaded
Open this post in threaded view
|

Does pharo build not retry anymore?

Guillermo Polito
I see that we are retrying tests only once, which is causing most (if not all) pull requests to fail sporadically...

Maybe it got lost because of a mistake?

Also, I'd like that changes in the build process are discussed in the mailing list from time to time... I've put a lot of effort into making it work and if we have regressions I'd like to understand them...

Guille

--

   

Guille Polito

Research Engineer

Centre de Recherche en Informatique, Signal et Automatique de Lille

CRIStAL - UMR 9189

French National Center for Scientific Research - http://www.cnrs.fr


Web: http://guillep.github.io

Phone: +33 06 52 70 66 13

Reply | Threaded
Open this post in threaded view
|

Re: Does pharo build not retry anymore?

Pavel Krivanek-3
Yes, with Marcus we decreased the retryTimes variable two weeks ago to 1 because in that time the validations were in really bad shape and literally every validation failed on some random issue and the retries caused enormous build times. It was producing more troubles than profit. It probably can be reverted now because the main problem of the validation instability was suppressed. I will prepare a PR.

-- Pavel

2017-12-12 18:52 GMT+01:00 Guillermo Polito <[hidden email]>:
I see that we are retrying tests only once, which is causing most (if not all) pull requests to fail sporadically...

Maybe it got lost because of a mistake?

Also, I'd like that changes in the build process are discussed in the mailing list from time to time... I've put a lot of effort into making it work and if we have regressions I'd like to understand them...

Guille

--

   

Guille Polito

Research Engineer

Centre de Recherche en Informatique, Signal et Automatique de Lille

CRIStAL - UMR 9189

French National Center for Scientific Research - http://www.cnrs.fr


Web: http://guillep.github.io

Phone: <a href="tel:+33%206%2052%2070%2066%2013" value="+33652706613" target="_blank">+33 06 52 70 66 13


Reply | Threaded
Open this post in threaded view
|

Re: Does pharo build not retry anymore?

Pavel Krivanek-3
The variable value was reverted to the original value. Thanks.

-- Pavel

2017-12-12 20:13 GMT+01:00 Pavel Krivanek <[hidden email]>:
Yes, with Marcus we decreased the retryTimes variable two weeks ago to 1 because in that time the validations were in really bad shape and literally every validation failed on some random issue and the retries caused enormous build times. It was producing more troubles than profit. It probably can be reverted now because the main problem of the validation instability was suppressed. I will prepare a PR.

-- Pavel

2017-12-12 18:52 GMT+01:00 Guillermo Polito <[hidden email]>:
I see that we are retrying tests only once, which is causing most (if not all) pull requests to fail sporadically...

Maybe it got lost because of a mistake?

Also, I'd like that changes in the build process are discussed in the mailing list from time to time... I've put a lot of effort into making it work and if we have regressions I'd like to understand them...

Guille

--

   

Guille Polito

Research Engineer

Centre de Recherche en Informatique, Signal et Automatique de Lille

CRIStAL - UMR 9189

French National Center for Scientific Research - http://www.cnrs.fr


Web: http://guillep.github.io

Phone: <a href="tel:+33%206%2052%2070%2066%2013" value="+33652706613" target="_blank">+33 06 52 70 66 13



Reply | Threaded
Open this post in threaded view
|

Re: Does pharo build not retry anymore?

Stephane Ducasse-3
Pavel what was the problem?

Stef


On Tue, Dec 12, 2017 at 9:34 PM, Pavel Krivanek <[hidden email]> wrote:
The variable value was reverted to the original value. Thanks.

-- Pavel

2017-12-12 20:13 GMT+01:00 Pavel Krivanek <[hidden email]>:
Yes, with Marcus we decreased the retryTimes variable two weeks ago to 1 because in that time the validations were in really bad shape and literally every validation failed on some random issue and the retries caused enormous build times. It was producing more troubles than profit. It probably can be reverted now because the main problem of the validation instability was suppressed. I will prepare a PR.

-- Pavel

2017-12-12 18:52 GMT+01:00 Guillermo Polito <[hidden email]>:
I see that we are retrying tests only once, which is causing most (if not all) pull requests to fail sporadically...

Maybe it got lost because of a mistake?

Also, I'd like that changes in the build process are discussed in the mailing list from time to time... I've put a lot of effort into making it work and if we have regressions I'd like to understand them...

Guille

--

   

Guille Polito

Research Engineer

Centre de Recherche en Informatique, Signal et Automatique de Lille

CRIStAL - UMR 9189

French National Center for Scientific Research - http://www.cnrs.fr


Web: http://guillep.github.io

Phone: <a href="tel:+33%206%2052%2070%2066%2013" value="+33652706613" target="_blank">+33 06 52 70 66 13




Reply | Threaded
Open this post in threaded view
|

Re: Does pharo build not retry anymore?

Pavel Krivanek-3
The problems were related UIManager initialization and consequences of the opening of the Welcome window. 

-- Pavel

2017-12-16 9:12 GMT+01:00 Stephane Ducasse <[hidden email]>:
Pavel what was the problem?

Stef


On Tue, Dec 12, 2017 at 9:34 PM, Pavel Krivanek <[hidden email]> wrote:
The variable value was reverted to the original value. Thanks.

-- Pavel

2017-12-12 20:13 GMT+01:00 Pavel Krivanek <[hidden email]>:
Yes, with Marcus we decreased the retryTimes variable two weeks ago to 1 because in that time the validations were in really bad shape and literally every validation failed on some random issue and the retries caused enormous build times. It was producing more troubles than profit. It probably can be reverted now because the main problem of the validation instability was suppressed. I will prepare a PR.

-- Pavel

2017-12-12 18:52 GMT+01:00 Guillermo Polito <[hidden email]>:
I see that we are retrying tests only once, which is causing most (if not all) pull requests to fail sporadically...

Maybe it got lost because of a mistake?

Also, I'd like that changes in the build process are discussed in the mailing list from time to time... I've put a lot of effort into making it work and if we have regressions I'd like to understand them...

Guille

--

   

Guille Polito

Research Engineer

Centre de Recherche en Informatique, Signal et Automatique de Lille

CRIStAL - UMR 9189

French National Center for Scientific Research - http://www.cnrs.fr


Web: http://guillep.github.io

Phone: <a href="tel:+33%206%2052%2070%2066%2013" value="+33652706613" target="_blank">+33 06 52 70 66 13





Reply | Threaded
Open this post in threaded view
|

Re: Does pharo build not retry anymore?

Stephane Ducasse-3
tx pavel

On Sat, Dec 16, 2017 at 9:25 AM, Pavel Krivanek <[hidden email]> wrote:
The problems were related UIManager initialization and consequences of the opening of the Welcome window. 

-- Pavel

2017-12-16 9:12 GMT+01:00 Stephane Ducasse <[hidden email]>:
Pavel what was the problem?

Stef


On Tue, Dec 12, 2017 at 9:34 PM, Pavel Krivanek <[hidden email]> wrote:
The variable value was reverted to the original value. Thanks.

-- Pavel

2017-12-12 20:13 GMT+01:00 Pavel Krivanek <[hidden email]>:
Yes, with Marcus we decreased the retryTimes variable two weeks ago to 1 because in that time the validations were in really bad shape and literally every validation failed on some random issue and the retries caused enormous build times. It was producing more troubles than profit. It probably can be reverted now because the main problem of the validation instability was suppressed. I will prepare a PR.

-- Pavel

2017-12-12 18:52 GMT+01:00 Guillermo Polito <[hidden email]>:
I see that we are retrying tests only once, which is causing most (if not all) pull requests to fail sporadically...

Maybe it got lost because of a mistake?

Also, I'd like that changes in the build process are discussed in the mailing list from time to time... I've put a lot of effort into making it work and if we have regressions I'd like to understand them...

Guille

--

   

Guille Polito

Research Engineer

Centre de Recherche en Informatique, Signal et Automatique de Lille

CRIStAL - UMR 9189

French National Center for Scientific Research - http://www.cnrs.fr


Web: http://guillep.github.io

Phone: <a href="tel:+33%206%2052%2070%2066%2013" value="+33652706613" target="_blank">+33 06 52 70 66 13






Reply | Threaded
Open this post in threaded view
|

Re: Does pharo build not retry anymore?

Pavel Krivanek-3
The UI Manager problem was fixed by Pablo and Guille

-- Pavel

2017-12-16 13:06 GMT+01:00 Stephane Ducasse <[hidden email]>:
tx pavel

On Sat, Dec 16, 2017 at 9:25 AM, Pavel Krivanek <[hidden email]> wrote:
The problems were related UIManager initialization and consequences of the opening of the Welcome window. 

-- Pavel

2017-12-16 9:12 GMT+01:00 Stephane Ducasse <[hidden email]>:
Pavel what was the problem?

Stef


On Tue, Dec 12, 2017 at 9:34 PM, Pavel Krivanek <[hidden email]> wrote:
The variable value was reverted to the original value. Thanks.

-- Pavel

2017-12-12 20:13 GMT+01:00 Pavel Krivanek <[hidden email]>:
Yes, with Marcus we decreased the retryTimes variable two weeks ago to 1 because in that time the validations were in really bad shape and literally every validation failed on some random issue and the retries caused enormous build times. It was producing more troubles than profit. It probably can be reverted now because the main problem of the validation instability was suppressed. I will prepare a PR.

-- Pavel

2017-12-12 18:52 GMT+01:00 Guillermo Polito <[hidden email]>:
I see that we are retrying tests only once, which is causing most (if not all) pull requests to fail sporadically...

Maybe it got lost because of a mistake?

Also, I'd like that changes in the build process are discussed in the mailing list from time to time... I've put a lot of effort into making it work and if we have regressions I'd like to understand them...

Guille

--

   

Guille Polito

Research Engineer

Centre de Recherche en Informatique, Signal et Automatique de Lille

CRIStAL - UMR 9189

French National Center for Scientific Research - http://www.cnrs.fr


Web: http://guillep.github.io

Phone: <a href="tel:+33%206%2052%2070%2066%2013" value="+33652706613" target="_blank">+33 06 52 70 66 13







Reply | Threaded
Open this post in threaded view
|

Re: Does pharo build not retry anymore?

Stephane Ducasse-3
tx them too. 
This is great to be surrounded by all you guys!


On Sat, Dec 16, 2017 at 8:22 PM, Pavel Krivanek <[hidden email]> wrote:
The UI Manager problem was fixed by Pablo and Guille

-- Pavel

2017-12-16 13:06 GMT+01:00 Stephane Ducasse <[hidden email]>:
tx pavel

On Sat, Dec 16, 2017 at 9:25 AM, Pavel Krivanek <[hidden email]> wrote:
The problems were related UIManager initialization and consequences of the opening of the Welcome window. 

-- Pavel

2017-12-16 9:12 GMT+01:00 Stephane Ducasse <[hidden email]>:
Pavel what was the problem?

Stef


On Tue, Dec 12, 2017 at 9:34 PM, Pavel Krivanek <[hidden email]> wrote:
The variable value was reverted to the original value. Thanks.

-- Pavel

2017-12-12 20:13 GMT+01:00 Pavel Krivanek <[hidden email]>:
Yes, with Marcus we decreased the retryTimes variable two weeks ago to 1 because in that time the validations were in really bad shape and literally every validation failed on some random issue and the retries caused enormous build times. It was producing more troubles than profit. It probably can be reverted now because the main problem of the validation instability was suppressed. I will prepare a PR.

-- Pavel

2017-12-12 18:52 GMT+01:00 Guillermo Polito <[hidden email]>:
I see that we are retrying tests only once, which is causing most (if not all) pull requests to fail sporadically...

Maybe it got lost because of a mistake?

Also, I'd like that changes in the build process are discussed in the mailing list from time to time... I've put a lot of effort into making it work and if we have regressions I'd like to understand them...

Guille

--

   

Guille Polito

Research Engineer

Centre de Recherche en Informatique, Signal et Automatique de Lille

CRIStAL - UMR 9189

French National Center for Scientific Research - http://www.cnrs.fr


Web: http://guillep.github.io

Phone: <a href="tel:+33%206%2052%2070%2066%2013" value="+33652706613" target="_blank">+33 06 52 70 66 13








Reply | Threaded
Open this post in threaded view
|

Re: Does pharo build not retry anymore?

Guillermo Polito
Hi, just to not mistake my intentions :)

I don't care WHO removed the retry or WHO fixed it. I care about the WHY.

If the WHY is not clear nor documented, we risk to make the mistake again. Please, let all of us be strict on the issue reports we do, and even stricter on ourselves. We need to raise the quality bar, not move it down!

And regarding this particular issue, I'd like to raise a warning. Let's not fix symptoms, fix causes. If we fix the symptom, the problem is still there. And moreover, if we fix the symptom and we lose track of it, we may forget unless somebody realizes weeks afterwards and it is too late (not this case fortunately).

On Sat, Dec 16, 2017 at 9:47 PM, Stephane Ducasse <[hidden email]> wrote:
tx them too. 
This is great to be surrounded by all you guys!


On Sat, Dec 16, 2017 at 8:22 PM, Pavel Krivanek <[hidden email]> wrote:
The UI Manager problem was fixed by Pablo and Guille

-- Pavel

2017-12-16 13:06 GMT+01:00 Stephane Ducasse <[hidden email]>:
tx pavel

On Sat, Dec 16, 2017 at 9:25 AM, Pavel Krivanek <[hidden email]> wrote:
The problems were related UIManager initialization and consequences of the opening of the Welcome window. 

-- Pavel

2017-12-16 9:12 GMT+01:00 Stephane Ducasse <[hidden email]>:
Pavel what was the problem?

Stef


On Tue, Dec 12, 2017 at 9:34 PM, Pavel Krivanek <[hidden email]> wrote:
The variable value was reverted to the original value. Thanks.

-- Pavel

2017-12-12 20:13 GMT+01:00 Pavel Krivanek <[hidden email]>:
Yes, with Marcus we decreased the retryTimes variable two weeks ago to 1 because in that time the validations were in really bad shape and literally every validation failed on some random issue and the retries caused enormous build times. It was producing more troubles than profit. It probably can be reverted now because the main problem of the validation instability was suppressed. I will prepare a PR.

-- Pavel

2017-12-12 18:52 GMT+01:00 Guillermo Polito <[hidden email]>:
I see that we are retrying tests only once, which is causing most (if not all) pull requests to fail sporadically...

Maybe it got lost because of a mistake?

Also, I'd like that changes in the build process are discussed in the mailing list from time to time... I've put a lot of effort into making it work and if we have regressions I'd like to understand them...

Guille

--

   

Guille Polito

Research Engineer

Centre de Recherche en Informatique, Signal et Automatique de Lille

CRIStAL - UMR 9189

French National Center for Scientific Research - http://www.cnrs.fr


Web: http://guillep.github.io

Phone: <a href="tel:+33%206%2052%2070%2066%2013" value="+33652706613" target="_blank">+33 06 52 70 66 13











--

   

Guille Polito

Research Engineer

Centre de Recherche en Informatique, Signal et Automatique de Lille

CRIStAL - UMR 9189

French National Center for Scientific Research - http://www.cnrs.fr


Web: http://guillep.github.io

Phone: +33 06 52 70 66 13

Reply | Threaded
Open this post in threaded view
|

Re: Does pharo build not retry anymore?

Stephane Ducasse-3
agreed :)


On Mon, Dec 18, 2017 at 11:43 AM, Guillermo Polito <[hidden email]> wrote:
Hi, just to not mistake my intentions :)

I don't care WHO removed the retry or WHO fixed it. I care about the WHY.

If the WHY is not clear nor documented, we risk to make the mistake again. Please, let all of us be strict on the issue reports we do, and even stricter on ourselves. We need to raise the quality bar, not move it down!

And regarding this particular issue, I'd like to raise a warning. Let's not fix symptoms, fix causes. If we fix the symptom, the problem is still there. And moreover, if we fix the symptom and we lose track of it, we may forget unless somebody realizes weeks afterwards and it is too late (not this case fortunately).

On Sat, Dec 16, 2017 at 9:47 PM, Stephane Ducasse <[hidden email]> wrote:
tx them too. 
This is great to be surrounded by all you guys!


On Sat, Dec 16, 2017 at 8:22 PM, Pavel Krivanek <[hidden email]> wrote:
The UI Manager problem was fixed by Pablo and Guille

-- Pavel

2017-12-16 13:06 GMT+01:00 Stephane Ducasse <[hidden email]>:
tx pavel

On Sat, Dec 16, 2017 at 9:25 AM, Pavel Krivanek <[hidden email]> wrote:
The problems were related UIManager initialization and consequences of the opening of the Welcome window. 

-- Pavel

2017-12-16 9:12 GMT+01:00 Stephane Ducasse <[hidden email]>:
Pavel what was the problem?

Stef


On Tue, Dec 12, 2017 at 9:34 PM, Pavel Krivanek <[hidden email]> wrote:
The variable value was reverted to the original value. Thanks.

-- Pavel

2017-12-12 20:13 GMT+01:00 Pavel Krivanek <[hidden email]>:
Yes, with Marcus we decreased the retryTimes variable two weeks ago to 1 because in that time the validations were in really bad shape and literally every validation failed on some random issue and the retries caused enormous build times. It was producing more troubles than profit. It probably can be reverted now because the main problem of the validation instability was suppressed. I will prepare a PR.

-- Pavel

2017-12-12 18:52 GMT+01:00 Guillermo Polito <[hidden email]>:
I see that we are retrying tests only once, which is causing most (if not all) pull requests to fail sporadically...

Maybe it got lost because of a mistake?

Also, I'd like that changes in the build process are discussed in the mailing list from time to time... I've put a lot of effort into making it work and if we have regressions I'd like to understand them...

Guille

--

   

Guille Polito

Research Engineer

Centre de Recherche en Informatique, Signal et Automatique de Lille

CRIStAL - UMR 9189

French National Center for Scientific Research - http://www.cnrs.fr


Web: http://guillep.github.io

Phone: <a href="tel:+33%206%2052%2070%2066%2013" value="+33652706613" target="_blank">+33 06 52 70 66 13











--

   

Guille Polito

Research Engineer

Centre de Recherche en Informatique, Signal et Automatique de Lille

CRIStAL - UMR 9189

French National Center for Scientific Research - http://www.cnrs.fr


Web: http://guillep.github.io

Phone: <a href="tel:+33%206%2052%2070%2066%2013" value="+33652706613" target="_blank">+33 06 52 70 66 13