- Pharo-dev - lists.pharo.org

Re: [Pharo-dev] filetree jenkins job

by Camillo Bruni

> | well that's what the fuel files are for... > | Can't you run the tests and copy the files to a safe location where > | you can analyze them offline? > > Nope. seems to be doable: http://sleepycoders.blogspot.fr/2013/03/sharing-travis-ci-generated-files.h… > > | Depending on stack traces to debug is definitely history ;) > > I guess I won't be porting Metacello to Pharo3.0 then either... Don't get me wrong, there still are stack traces, but most probably the bug you experience is more hidden. What I am trying to say is that should use the fuel files to debug failing tests instead of spending time deciphering dead text-based stack traces. And exactly in the case you describe theses fuel files can save you a lot of time since you can basically debug the same way you debug a local application. BTW: where do I see which tests fail under travis? clicking on the status badges on the filetree page doesn't lead me to the direct results. In any case I see 4 failing tests on our jenkins build: https://ci.inria.fr/pharo-contribution/job/FileTree/PHARO=30,VERSION=stable… are these the same you have on travis?

July 8, 2013

Re: [Pharo-dev] filetree jenkins job

by Dale K. Henrichs

----- Original Message ----- | From: "Camillo Bruni" <camillobruni(a)gmail.com> | To: "Pharo Development List" <pharo-dev(a)lists.pharo.org> | Sent: Sunday, July 7, 2013 7:02:26 PM | Subject: Re: [Pharo-dev] filetree jenkins job | | > Running on travisCI does not allow me to get access to serialized | > anything ... I get to see a log file (what do you expect for | > free:) ... a simple stack report printed is usually enough for me | > to figure out what's happening but the above is totally useless | > ... most cases reproduce locally, but the pharo3.0 issues that I'm | > trying to debug now are failing only on travis so I need to have | > something to point to what's wrong ... like I said I figured the | > headless operation had been jiggered in some way and will | > patiently wait for Christophe to come up with an acceptable | > solution for running on travis ... | | | well that's what the fuel files are for... | Can't you run the tests and copy the files to a safe location where | you can analyze them offline? Nope. | | Depending on stack traces to debug is definitely history ;) I guess I won't be porting Metacello to Pharo3.0 then either... Dale

July 8, 2013

Re: [Pharo-dev] filetree jenkins job

by Camillo Bruni

> Running on travisCI does not allow me to get access to serialized anything ... I get to see a log file (what do you expect for free:) ... a simple stack report printed is usually enough for me to figure out what's happening but the above is totally useless ... most cases reproduce locally, but the pharo3.0 issues that I'm trying to debug now are failing only on travis so I need to have something to point to what's wrong ... like I said I figured the headless operation had been jiggered in some way and will patiently wait for Christophe to come up with an acceptable solution for running on travis ... well that's what the fuel files are for... Can't you run the tests and copy the files to a safe location where you can analyze them offline? Depending on stack traces to debug is definitely history ;)

July 8, 2013

Re: [Pharo-dev] filetree jenkins job

by Dale K. Henrichs

----- Original Message ----- | From: "Camillo Bruni" <camillobruni(a)gmail.com> | To: "Pharo Development List" <pharo-dev(a)lists.pharo.org> | Sent: Sunday, July 7, 2013 5:26:45 PM | Subject: Re: [Pharo-dev] filetree jenkins job | | On 2013-07-08, at 01:59, Dale K. Henrichs | <dale.henrichs(a)gemtalksystems.com> wrote: | > Cami, | > | > Even worse, I don't get _any_ information about failures. When I | > run unit tests headless I get the following in my output: | > | > =============================================================================== | > Notice: Errors in script loaded from | > /home/travis/dalehenrich-builderCI-3062fdf/builds/travisCI/travisCI.st | > =============================================================================== | > | > No stacks in the debug log, so I am completely blind when I run | > Pharo3.0 builds ... been happening about 2 weeks ... I assume | > there was a radical change to the way the Pharo3.0 reports | > headless errors, either by accident or on purpose... | | that's usually not a bug, it currently is a side-effect of Opal | giving proper feedback about undefined variables. | With the new classbuilder introduce we might have some side-effects | as well (there is a reported case that Martin and Christophe found). | | > Christophe will be checking into this and I am sure he'll et me | > running again ... | > | > Until I can see what's happening with the builds I can't get the | > last little bits fixed for Pharo3.0 ... very close though:) | | sometimes what helps is running in jenkins mode, failing tests | produce a serialized stack which can be drag n dropped onto an image | for easier debugging. I use that on a regular basis when figuring | out non-deterministic tests :) | | You can see such files here: | https://ci.inria.fr/pharo-contribution/job/FileTree/PHARO=30,VERSION=stable… | Running on travisCI does not allow me to get access to serialized anything ... I get to see a log file (what do you expect for free:) ... a simple stack report printed is usually enough for me to figure out what's happening but the above is totally useless ... most cases reproduce locally, but the pharo3.0 issues that I'm trying to debug now are failing only on travis so I need to have something to point to what's wrong ... like I said I figured the headless operation had been jiggered in some way and will patiently wait for Christophe to come up with an acceptable solution for running on travis ... Dale

July 8, 2013

Re: [Pharo-dev] NeoCSV doc is now on Gutenberg :)

by Sean P. DeNigris

StÃ©phane Ducasse wrote > I love reading documentation :) Yes, and the NeoCSV doc is really good :) It felt like having an expert mentor me... ----- Cheers, Sean -- View this message in context: http://forum.world.st/Pharo-dev-NeoCSV-doc-is-now-on-Gutenberg-tp4697733p46… Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

July 8, 2013

Re: [Pharo-dev] filetree jenkins job

by Camillo Bruni

On 2013-07-08, at 01:59, Dale K. Henrichs <dale.henrichs(a)gemtalksystems.com> wrote: > Cami, > > Even worse, I don't get _any_ information about failures. When I run unit tests headless I get the following in my output: > > =============================================================================== > Notice: Errors in script loaded from /home/travis/dalehenrich-builderCI-3062fdf/builds/travisCI/travisCI.st > =============================================================================== > > No stacks in the debug log, so I am completely blind when I run Pharo3.0 builds ... been happening about 2 weeks ... I assume there was a radical change to the way the Pharo3.0 reports headless errors, either by accident or on purpose... that's usually not a bug, it currently is a side-effect of Opal giving proper feedback about undefined variables. With the new classbuilder introduce we might have some side-effects as well (there is a reported case that Martin and Christophe found). > Christophe will be checking into this and I am sure he'll et me running again ... > > Until I can see what's happening with the builds I can't get the last little bits fixed for Pharo3.0 ... very close though:) sometimes what helps is running in jenkins mode, failing tests produce a serialized stack which can be drag n dropped onto an image for easier debugging. I use that on a regular basis when figuring out non-deterministic tests :) You can see such files here: https://ci.inria.fr/pharo-contribution/job/FileTree/PHARO=30,VERSION=stable…

July 8, 2013

Re: [Pharo-dev] filetree jenkins job

by Dale K. Henrichs

Cami, Even worse, I don't get _any_ information about failures. When I run unit tests headless I get the following in my output: =============================================================================== Notice: Errors in script loaded from /home/travis/dalehenrich-builderCI-3062fdf/builds/travisCI/travisCI.st =============================================================================== =============================================================================== Notice: Errors in script loaded from /home/travis/dalehenrich-builderCI-3062fdf/builds/travisCI/travisCI.st =============================================================================== =============================================================================== Notice: Errors in script loaded from /home/travis/dalehenrich-builderCI-3062fdf/builds/travisCI/travisCI.st =============================================================================== =============================================================================== Notice: Errors in script loaded from /home/travis/dalehenrich-builderCI-3062fdf/builds/travisCI/travisCI.st =============================================================================== =============================================================================== Notice: Errors in script loaded from /home/travis/dalehenrich-builderCI-3062fdf/builds/travisCI/travisCI.st =============================================================================== =============================================================================== Notice: Errors in script loaded from /home/travis/dalehenrich-builderCI-3062fdf/builds/travisCI/travisCI.st =============================================================================== No stacks in the debug log, so I am completely blind when I run Pharo3.0 builds ... been happening about 2 weeks ... I assume there was a radical change to the way the Pharo3.0 reports headless errors, either by accident or on purpose... Christophe will be checking into this and I am sure he'll et me running again ... Until I can see what's happening with the builds I can't get the last little bits fixed for Pharo3.0 ... very close though:) Dale ----- Original Message ----- | From: "Camillo Bruni" <camillobruni(a)gmail.com> | To: "Pharo Development List" <pharo-dev(a)lists.pharo.org> | Sent: Sunday, July 7, 2013 3:11:38 PM | Subject: Re: [Pharo-dev] filetree jenkins job | | I will change SUnit first produce proper error messages for the | #shouldnt:raise: methods since I cannot properly debug them. | Additionally at the point where the #:assert: happens the wrong | exception is not in the context which means it does not get properly | serialized by our test framework. All in all, with a little effort | we could debug much easier. | | BTW: There are so many strange methods on TAssertable that I do not | understand | - #shouldFix: | - #executeShould:inScopeOf: | | and some that I honestly think they hurt more than they help: | - #shouldnt:raise:whoseDescriptionIncludes:description: and friends? | | On 2013-07-07, at 23:47, "Dale K. Henrichs" | <dale.henrichs(a)gemtalksystems.com> wrote: | | > Cami, | > | > I've been having trouble with getting any useful information from | > Pharo3.0 failures on travis-ci[1] for a week or so now ... | > Christophe will be looking into this for me... | > | > Dale | > | > [1] | > https://github.com/dalehenrich/filetree/issues/80#issuecomment-20162125 | > ----- Original Message ----- | > | From: "Camillo Bruni" <camillobruni(a)gmail.com> | > | To: pharo-dev(a)lists.pharo.org | > | Sent: Sunday, July 7, 2013 2:00:10 PM | > | Subject: [Pharo-dev] filetree jenkins job | > | | > | after I tried to get filetree with the tests installed and I | > | didn't | > | get it | > | the first time I decided to create a jenkins job. | > | | > | https://ci.inria.fr/pharo-contribution/job/FileTree/ | > | | > | | > | | |

July 7, 2013

Re: [Pharo-dev] filetree jenkins job

by Camillo Bruni

I will change SUnit first produce proper error messages for the #shouldnt:raise: methods since I cannot properly debug them. Additionally at the point where the #:assert: happens the wrong exception is not in the context which means it does not get properly serialized by our test framework. All in all, with a little effort we could debug much easier. BTW: There are so many strange methods on TAssertable that I do not understand - #shouldFix: - #executeShould:inScopeOf: and some that I honestly think they hurt more than they help: - #shouldnt:raise:whoseDescriptionIncludes:description: and friends? On 2013-07-07, at 23:47, "Dale K. Henrichs" <dale.henrichs(a)gemtalksystems.com> wrote: > Cami, > > I've been having trouble with getting any useful information from Pharo3.0 failures on travis-ci[1] for a week or so now ... Christophe will be looking into this for me... > > Dale > > [1] https://github.com/dalehenrich/filetree/issues/80#issuecomment-20162125 > ----- Original Message ----- > | From: "Camillo Bruni" <camillobruni(a)gmail.com> > | To: pharo-dev(a)lists.pharo.org > | Sent: Sunday, July 7, 2013 2:00:10 PM > | Subject: [Pharo-dev] filetree jenkins job > | > | after I tried to get filetree with the tests installed and I didn't > | get it > | the first time I decided to create a jenkins job. > | > | https://ci.inria.fr/pharo-contribution/job/FileTree/ > | > | >

July 7, 2013

Re: [Pharo-dev] filetree jenkins job

by Dale K. Henrichs

Cami, I've been having trouble with getting any useful information from Pharo3.0 failures on travis-ci[1] for a week or so now ... Christophe will be looking into this for me... Dale [1] https://github.com/dalehenrich/filetree/issues/80#issuecomment-20162125 ----- Original Message ----- | From: "Camillo Bruni" <camillobruni(a)gmail.com> | To: pharo-dev(a)lists.pharo.org | Sent: Sunday, July 7, 2013 2:00:10 PM | Subject: [Pharo-dev] filetree jenkins job | | after I tried to get filetree with the tests installed and I didn't | get it | the first time I decided to create a jenkins job. | | https://ci.inria.fr/pharo-contribution/job/FileTree/ | |

July 7, 2013

Re: [Pharo-dev] filetree jenkins job

by Benjamin

Cool :) Ben On Jul 7, 2013, at 11:00 PM, Camillo Bruni <camillobruni(a)gmail.com> wrote: > after I tried to get filetree with the tests installed and I didn't get it > the first time I decided to create a jenkins job. > > https://ci.inria.fr/pharo-contribution/job/FileTree/ >

July 7, 2013