issue 12506 Vm crashes vm when testing updates [A-L]
My observations for issue 12506 - I can reproduce it on linux - segfault happens in primtiveNextInstance - tests A-L worked in 30663 - tests A-L segfault since 30664 - doesn't segfaults anymore if you open the image, change some code, save and close the image - it segfaults if you just run a subset of the A-L tests (pharo --nodisplay Pharo.image test --junit-xml-output 'FileSystem-Tests-Core' 'FreeTypeTests' 'FileSystem-Tests-Disk' 'FuelTests' 'FileSystem-Tests-Memory' 'KernelTests') - this "fail"-subset is different for the different images (30664, 30665, ...) - it does not segfault if you omit the junit-xml-output option And someone knows what happened with the image between version 30663 and 30666, because only 30664 and 30665 need the PharoV20.sources in the same directory as the VM, for all other versions it suffices if the sources are where the image file is. (Same as in http://forum.world.st/pharo-3-0-image-td4732890.html) Nicolai
On 05 Jan 2014, at 12:08, Nicolai Hess <nicolaihess@web.de> wrote:
My observations for issue 12506
- I can reproduce it on linux - segfault happens in primtiveNextInstance - tests A-L worked in 30663 - tests A-L segfault since 30664 - doesn't segfaults anymore if you open the image, change some code, save and close the image - it segfaults if you just run a subset of the A-L tests (pharo --nodisplay Pharo.image test --junit-xml-output 'FileSystem-Tests-Core' 'FreeTypeTests' 'FileSystem-Tests-Disk' 'FuelTests' 'FileSystem-Tests-Memory' 'KernelTests') - this "fail"-subset is different for the different images (30664, 30665, ...) - it does not segfault if you omit the junit-xml-output option
-> Opal Regression Tests do run all tests (after recompiling) â> no problem ever -> Monkey runs all tests for all âFix review neededâ â> no problem
And someone knows what happened with the image between version 30663 and 30666, because only 30664 and 30665 need the PharoV20.sources in the same directory as the VM, for all other versions it suffices if the sources are where the image file is. (Same as in http://forum.world.st/pharo-3-0-image-td4732890.html)
These are the changes: 30663: full diff: https://github.com/pharo-project/pharo-core/commit/c6adb033b0a42acbfe4519554... 12510 NautilusHistory entry for class methods loads instance methods https://pharo.fogbugz.com/f/cases/12510 12502 FileList MNU FileSystemPermission>><= https://pharo.fogbugz.com/f/cases/12502 12429 MessageBrowser loses changes https://pharo.fogbugz.com/f/cases/12429 12451 Error on Browse Full in Nautilus from class in group view https://pharo.fogbugz.com/f/cases/12451 30664 full diff: https://github.com/pharo-project/pharo-core/commit/b84651feddc39d4336329e223... 12517 SystemNavigation methodHierarchyBrowserForClass:selector: discards package information https://pharo.fogbugz.com/f/cases/12517 12476 Undeclared https://pharo.fogbugz.com/f/cases/12476 (remove Athens Tutorial as it introduces an Undeclared and was not in the image before) (in addition, I downloaded the image, recompiled and uploaded a new seed image as latest.zip). 30665 full diff: https://mail.google.com/mail/u/0/#search/30665/1435265c4bcdb3c2 - do not call #cleanUpForRelease on update (to check if this was the reason for the crash, it was not)
But what of these changes were responsible for that you have to have the sources file in your vm directory? (and only in 30664 and 30665!) What are the exact steps for recompiling the image, you did? (just OpalCompiler recompileAll) 2014/1/5 Marcus Denker <marcus.denker@inria.fr>
On 05 Jan 2014, at 12:08, Nicolai Hess <nicolaihess@web.de> wrote:
My observations for issue 12506
- I can reproduce it on linux - segfault happens in primtiveNextInstance - tests A-L worked in 30663 - tests A-L segfault since 30664 - doesn't segfaults anymore if you open the image, change some code, save and close the image - it segfaults if you just run a subset of the A-L tests (pharo --nodisplay Pharo.image test --junit-xml-output 'FileSystem-Tests-Core' 'FreeTypeTests' 'FileSystem-Tests-Disk' 'FuelTests' 'FileSystem-Tests-Memory' 'KernelTests') - this "fail"-subset is different for the different images (30664, 30665, ...) - it does not segfault if you omit the junit-xml-output option
-> Opal Regression Tests do run all tests (after recompiling) â> no problem ever -> Monkey runs all tests for all âFix review neededâ â> no problem
And someone knows what happened with the image between version 30663 and 30666, because only 30664 and 30665 need the PharoV20.sources in the same directory as the VM, for all other versions it suffices if the sources are where the image file is. (Same as in http://forum.world.st/pharo-3-0-image-td4732890.html)
These are the changes:
30663: full diff: https://github.com/pharo-project/pharo-core/commit/c6adb033b0a42acbfe4519554...
12510 NautilusHistory entry for class methods loads instance methods https://pharo.fogbugz.com/f/cases/12510
12502 FileList MNU FileSystemPermission>><= https://pharo.fogbugz.com/f/cases/12502
12429 MessageBrowser loses changes https://pharo.fogbugz.com/f/cases/12429
12451 Error on Browse Full in Nautilus from class in group view https://pharo.fogbugz.com/f/cases/12451 30664 full diff: https://github.com/pharo-project/pharo-core/commit/b84651feddc39d4336329e223...
12517 SystemNavigation methodHierarchyBrowserForClass:selector: discards package information https://pharo.fogbugz.com/f/cases/12517
12476 Undeclared https://pharo.fogbugz.com/f/cases/12476 (remove Athens Tutorial as it introduces an Undeclared and was not in the image before)
(in addition, I downloaded the image, recompiled and uploaded a new seed image as latest.zip).
30665 full diff: https://mail.google.com/mail/u/0/#search/30665/1435265c4bcdb3c2
- do not call #cleanUpForRelease on update
(to check if this was the reason for the crash, it was not)
On 05 Jan 2014, at 12:37, Nicolai Hess <nicolaihess@web.de> wrote:
But what of these changes were responsible for that you have to have the sources file in your vm directory? (and only in 30664 and 30665!)
Or maybe because I did do a âSmalltalk cleanUpâ before uploading the recompiled image?
What are the exact steps for recompiling the image, you did? (just OpalCompiler recompileAll)
2014/1/5 Marcus Denker <marcus.denker@inria.fr>
On 05 Jan 2014, at 12:08, Nicolai Hess <nicolaihess@web.de> wrote:
My observations for issue 12506
- I can reproduce it on linux - segfault happens in primtiveNextInstance - tests A-L worked in 30663 - tests A-L segfault since 30664 - doesn't segfaults anymore if you open the image, change some code, save and close the image - it segfaults if you just run a subset of the A-L tests (pharo --nodisplay Pharo.image test --junit-xml-output 'FileSystem-Tests-Core' 'FreeTypeTests' 'FileSystem-Tests-Disk' 'FuelTests' 'FileSystem-Tests-Memory' 'KernelTests') - this "fail"-subset is different for the different images (30664, 30665, ...) - it does not segfault if you omit the junit-xml-output option
-> Opal Regression Tests do run all tests (after recompiling) â> no problem ever -> Monkey runs all tests for all âFix review neededâ â> no problem
And someone knows what happened with the image between version 30663 and 30666, because only 30664 and 30665 need the PharoV20.sources in the same directory as the VM, for all other versions it suffices if the sources are where the image file is. (Same as in http://forum.world.st/pharo-3-0-image-td4732890.html)
These are the changes:
30663: full diff: https://github.com/pharo-project/pharo-core/commit/c6adb033b0a42acbfe4519554...
12510 NautilusHistory entry for class methods loads instance methods https://pharo.fogbugz.com/f/cases/12510
12502 FileList MNU FileSystemPermission>><= https://pharo.fogbugz.com/f/cases/12502
12429 MessageBrowser loses changes https://pharo.fogbugz.com/f/cases/12429
12451 Error on Browse Full in Nautilus from class in group view https://pharo.fogbugz.com/f/cases/12451
30664 full diff: https://github.com/pharo-project/pharo-core/commit/b84651feddc39d4336329e223...
12517 SystemNavigation methodHierarchyBrowserForClass:selector: discards package information https://pharo.fogbugz.com/f/cases/12517
12476 Undeclared https://pharo.fogbugz.com/f/cases/12476 (remove Athens Tutorial as it introduces an Undeclared and was not in the image before)
(in addition, I downloaded the image, recompiled and uploaded a new seed image as latest.zip).
30665 full diff: https://mail.google.com/mail/u/0/#search/30665/1435265c4bcdb3c2
- do not call #cleanUpForRelease on update
(to check if this was the reason for the crash, it was not)
I tried to take the latest + latestVM and run my unreloader and now it never finishes but no crash. I have to investigate. Stef On 05 Jan 2014, at 12:51, Marcus Denker <marcus.denker@inria.fr> wrote:
On 05 Jan 2014, at 12:37, Nicolai Hess <nicolaihess@web.de> wrote:
But what of these changes were responsible for that you have to have the sources file in your vm directory? (and only in 30664 and 30665!)
Or maybe because I did do a âSmalltalk cleanUpâ before uploading the recompiled image?
What are the exact steps for recompiling the image, you did? (just OpalCompiler recompileAll)
2014/1/5 Marcus Denker <marcus.denker@inria.fr>
On 05 Jan 2014, at 12:08, Nicolai Hess <nicolaihess@web.de> wrote:
My observations for issue 12506
- I can reproduce it on linux - segfault happens in primtiveNextInstance - tests A-L worked in 30663 - tests A-L segfault since 30664 - doesn't segfaults anymore if you open the image, change some code, save and close the image - it segfaults if you just run a subset of the A-L tests (pharo --nodisplay Pharo.image test --junit-xml-output 'FileSystem-Tests-Core' 'FreeTypeTests' 'FileSystem-Tests-Disk' 'FuelTests' 'FileSystem-Tests-Memory' 'KernelTests') - this "fail"-subset is different for the different images (30664, 30665, ...) - it does not segfault if you omit the junit-xml-output option
-> Opal Regression Tests do run all tests (after recompiling) â> no problem ever -> Monkey runs all tests for all âFix review neededâ â> no problem
And someone knows what happened with the image between version 30663 and 30666, because only 30664 and 30665 need the PharoV20.sources in the same directory as the VM, for all other versions it suffices if the sources are where the image file is. (Same as in http://forum.world.st/pharo-3-0-image-td4732890.html)
These are the changes:
30663: full diff: https://github.com/pharo-project/pharo-core/commit/c6adb033b0a42acbfe4519554...
12510 NautilusHistory entry for class methods loads instance methods https://pharo.fogbugz.com/f/cases/12510
12502 FileList MNU FileSystemPermission>><= https://pharo.fogbugz.com/f/cases/12502
12429 MessageBrowser loses changes https://pharo.fogbugz.com/f/cases/12429
12451 Error on Browse Full in Nautilus from class in group view https://pharo.fogbugz.com/f/cases/12451
30664 full diff: https://github.com/pharo-project/pharo-core/commit/b84651feddc39d4336329e223...
12517 SystemNavigation methodHierarchyBrowserForClass:selector: discards package information https://pharo.fogbugz.com/f/cases/12517
12476 Undeclared https://pharo.fogbugz.com/f/cases/12476 (remove Athens Tutorial as it introduces an Undeclared and was not in the image before)
(in addition, I downloaded the image, recompiled and uploaded a new seed image as latest.zip).
30665 full diff: https://mail.google.com/mail/u/0/#search/30665/1435265c4bcdb3c2
- do not call #cleanUpForRelease on update
(to check if this was the reason for the crash, it was not)
Marcus I could not get it crashed with the latest vm and latest image. Now Iâm confident that on linux it should crash. Pavel? could you try? I will ask damien. Stef
I tried to take the latest + latestVM and run my unreloader and now it never finishes but no crash. I have to investigate.
Stef
On 05 Jan 2014, at 12:51, Marcus Denker <marcus.denker@inria.fr> wrote:
On 05 Jan 2014, at 12:37, Nicolai Hess <nicolaihess@web.de> wrote:
But what of these changes were responsible for that you have to have the sources file in your vm directory? (and only in 30664 and 30665!)
Or maybe because I did do a âSmalltalk cleanUpâ before uploading the recompiled image?
What are the exact steps for recompiling the image, you did? (just OpalCompiler recompileAll)
2014/1/5 Marcus Denker <marcus.denker@inria.fr>
On 05 Jan 2014, at 12:08, Nicolai Hess <nicolaihess@web.de> wrote:
My observations for issue 12506
- I can reproduce it on linux - segfault happens in primtiveNextInstance - tests A-L worked in 30663 - tests A-L segfault since 30664 - doesn't segfaults anymore if you open the image, change some code, save and close the image - it segfaults if you just run a subset of the A-L tests (pharo --nodisplay Pharo.image test --junit-xml-output 'FileSystem-Tests-Core' 'FreeTypeTests' 'FileSystem-Tests-Disk' 'FuelTests' 'FileSystem-Tests-Memory' 'KernelTests') - this "fail"-subset is different for the different images (30664, 30665, ...) - it does not segfault if you omit the junit-xml-output option
-> Opal Regression Tests do run all tests (after recompiling) â> no problem ever -> Monkey runs all tests for all âFix review neededâ â> no problem
And someone knows what happened with the image between version 30663 and 30666, because only 30664 and 30665 need the PharoV20.sources in the same directory as the VM, for all other versions it suffices if the sources are where the image file is. (Same as in http://forum.world.st/pharo-3-0-image-td4732890.html)
These are the changes:
30663: full diff: https://github.com/pharo-project/pharo-core/commit/c6adb033b0a42acbfe4519554...
12510 NautilusHistory entry for class methods loads instance methods https://pharo.fogbugz.com/f/cases/12510
12502 FileList MNU FileSystemPermission>><= https://pharo.fogbugz.com/f/cases/12502
12429 MessageBrowser loses changes https://pharo.fogbugz.com/f/cases/12429
12451 Error on Browse Full in Nautilus from class in group view https://pharo.fogbugz.com/f/cases/12451
30664 full diff: https://github.com/pharo-project/pharo-core/commit/b84651feddc39d4336329e223...
12517 SystemNavigation methodHierarchyBrowserForClass:selector: discards package information https://pharo.fogbugz.com/f/cases/12517
12476 Undeclared https://pharo.fogbugz.com/f/cases/12476 (remove Athens Tutorial as it introduces an Undeclared and was not in the image before)
(in addition, I downloaded the image, recompiled and uploaded a new seed image as latest.zip).
30665 full diff: https://mail.google.com/mail/u/0/#search/30665/1435265c4bcdb3c2
- do not call #cleanUpForRelease on update
(to check if this was the reason for the crash, it was not)
participants (3)
-
Marcus Denker -
Nicolai Hess -
Stéphane Ducasse