gaffer crash
197 views
Skip to first unread message
s.des...@gmail.com
Apr 22, 2021, 9:15:56 AM4/22/21
to gaffer-dev
Hello,
I get this error in my gaffer script and I can't find why it is crashing ? is someone can have a look at the logs ? I don't really understand what it says ;)
Thanks a lot,
here is the log :
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n1" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n2" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n3" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n1" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n2" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n3" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n1" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n2" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n3" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "uv" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n1" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n2" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n3" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "uv" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n1" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n2" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n3" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "uv" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n1" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n2" has unsupported array extent
WARNING : FromAlembicGeomBaseConverter::convertArbGeomParam : Param "n3" has unsupported array extent
ERROR | signal caught: SIGSEGV -- Invalid memory reference
****
* Arnold 6.0.2.1 [088e32fd] linux clang-5.0.0 oiio-2.1.4 osl-1.11.0 vdb-4.0.0 clm-1.1.1.118 rlm-12.4.2 optix-6.5.0 2020/03/06 10:17:06
* CRASHED in boost::intrusive_ptr<IECore::TypedData<Imath_2_4::Box<Imath_2_4::Vec3<float> > > const> Gaffer::ValuePlug::getObjectValue<IECore::TypedData<Imath_2_4::Box<Imath_2_4::Vec3<float> > > > at 00:00:00
* signal caught: SIGSEGV -- Invalid memory reference (address not mapped to object)
*
* backtrace:
* 0 0x00007fe9ec1e6242 [libai.so ] AiPromptADPDialog
* 1 0x00007fea62fb262f [libpthread.so.0 ] _L_unlock_13 [funlockfile.c : ?]
>> 2 0x00007fea4f2a7f80 [libGaffer.so ] boost::intrusive_ptr<IECore::TypedData<Imath_2_4::Box<Imath_2_4::Vec3<float> > > const> Gaffer::ValuePlug::getObjectValue<IECore::TypedData<Imath_2_4::Box<Imath_2_4::Vec3<float> > > >(IECore::MurmurHash const*)
* 3 0x00007fea4f2a7f80 [libGaffer.so ] boost::intrusive_ptr<IECore::TypedData<Imath_2_4::Box<Imath_2_4::Vec3<float> > > const> Gaffer::ValuePlug::getObjectValue<IECore::TypedData<Imath_2_4::Box<Imath_2_4::Vec3<float> > > >(IECore::MurmurHash const*)
* 4 0x00007fea1bd334a9 [libGafferScene.so] GafferScene::SceneReader::computeBound(std::vector<IECore::InternedString, std::allocator<IECore::InternedString> > const&, Gaffer::Context const*, GafferScene::ScenePlug const*)
* 5 0x00007fea1bd28d94 [libGafferScene.so] GafferScene::SceneNode::compute(Gaffer::ValuePlug*, Gaffer::Context const*)
* 6 0x00007fea4f2a8e9b [libGaffer.so ] boost::intrusive_ptr<IECore::TypedData<Imath_2_4::Box<Imath_2_4::Vec3<float> > > const> Gaffer::ValuePlug::getObjectValue<IECore::TypedData<Imath_2_4::Box<Imath_2_4::Vec3<float> > > >(IECore::MurmurHash const*) [ValuePlug.cpp : ?]
* 7 0x00007fea4f2b3cfa [libGaffer.so ] Gaffer::ValuePlug::ComputeProcess::value(Gaffer::ValuePlug const*, IECore::MurmurHash const*)
* 8 0x00007fea4f2ac9c8 [libGaffer.so ] Gaffer::ValuePlug::getValueInternal(IECore::MurmurHash const*)
* 9 0x00007fea4f2a7f40 [libGaffer.so ] boost::intrusive_ptr<IECore::TypedData<Imath_2_4::Box<Imath_2_4::Vec3<float> > > const> Gaffer::ValuePlug::getObjectValue<IECore::TypedData<Imath_2_4::Box<Imath_2_4::Vec3<float> > > >(IECore::MurmurHash const*)
* 10 0x00007fea4f2a81ff [libGaffer.so ] Gaffer::TypedPlug<Imath_2_4::Box<Imath_2_4::Vec3<float> > >::getValue(IECore::MurmurHash const*)
* 11 0x00007fea1bd2859e [libGafferScene.so] GafferScene::SceneNode::compute(Gaffer::ValuePlug*, Gaffer::Context const*) [SceneNode.cpp : ?]
* 12 0x00007fea5666f7ab [libtbb.so.2 ] tbb::internal::custom_scheduler<tbb::internal::IntelSchedulerTraits>::local_wait_for_all(tbb::task&, tbb::task*) [custom_scheduler.h:509]
* 13 0x00007fea56667f63 [libtbb.so.2 ] [arena.cpp :160]
* 14 0x00007fea56666a9a [libtbb.so.2 ] tbb::internal::arena::process(tbb::internal::generic_scheduler&) [market.cpp :693]
* 15 0x00007fea56662d45 [libtbb.so.2 ] [private_server.cpp:270]
* 16 0x00007fea56662f78 [libtbb.so.2 ] tbb::internal::market::process(rml::job&) [private_server.cpp:223]
* 17 0x00007fea62faaea4 [libpthread.so.0 ] [pthread_create.c : ?]
* 18 0x00007fea625ca96c [libc.so.6 ] clone
*
* loaded modules:
* 0x00007fe9eb7c9000 libai.so
* 0x00007fea62fa3000 libpthread.so.0
* 0x00007fea4f123000 libGaffer.so
* 0x00007fea1baf0000 libGafferScene.so
* 0x00007fea56640000 libtbb.so.2
* 0x00007fea624cc000 libc.so.6
*
* memory: VM 53526 MB, RSS 4262 MB, 2 page faults
****
Segmentation fault (core dumped)
John Haddon
Apr 22, 2021, 11:46:12 AM4/22/21
to gaffe...@googlegroups.com
Looks like it's crashing while trying to compute the bounding box for the Alembic scene. I did recently fix a crash triggered by loading invalid primitive variables from an Alembic file, but that hasn't made it into a release version yet. Those warning messages do make me wonder if there might be something else problematic in the file? I don't suppose you are able to share it are you? Or a cut-down version that still shows the problem?
--
You received this message because you are subscribed to the Google Groups "gaffer-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gaffer-dev+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/gaffer-dev/d8ef8ed4-6218-4b4f-b599-3c0312e84912n%40googlegroups.com.
s.des...@gmail.com
Apr 22, 2021, 3:29:30 PM4/22/21
to gaffer-dev
Hello John,
Unfortunatly I will not be able to share my gaffer script.
But as you mention the bounding box, I have lot of prune and isolate nodes. And I see that there is a "adjust bounds" on these nodes.
Is this option can fix this kind of issue ?
thx,
Seb
Andrew Kaufman
Apr 22, 2021, 3:36:22 PM4/22/21
to gaffe...@googlegroups.com
It seems like the crash is occurring when loading the Alembic via the SceneReader, rather than from one of the downstream nodes. Do `abcecho` or maybe `usdview` report any issues with the file?
Could you try making a cut-down version of the .abc without any sensitive geometry in it and see if that suffers the same problem?
Andrew
To view this discussion on the web visit https://groups.google.com/d/msgid/gaffer-dev/5039b349-65db-4f0c-9349-9b77ccac75e0n%40googlegroups.com.
|
|
William Eguienta
May 3, 2021, 4:46:31 AM5/3/21
to gaffer-dev
Hello,
I also have several random crashes with 0.59.x on centos 7
ERROR | signal caught: SIGSEGV -- Invalid memory reference
****
* signal caught: SIGSEGV -- Invalid memory reference (address not mapped to object)
*
* backtrace:
I also have several random crashes with 0.59.x on centos 7
mainly when I try to open a script directly from terminal, like :
gaffer /hdd/myscene.gfr
it result :
it result :
ERROR | signal caught: SIGSEGV -- Invalid memory reference
****
* Arnold 6.1.0.1 [07981301] linux clang-10.0.1 oiio-2.2.1 osl-1.11.6 vdb-4.0.0 clm-1.1.1.118 rlm-12.4.2 optix-6.6.0 2020/12/15 09:25:16
* CRASHED at 00:00:00
* CRASHED at 00:00:00
* signal caught: SIGSEGV -- Invalid memory reference (address not mapped to object)
*
* backtrace:
* 0 0x00007feb4c7b7b70 [libai.so ] AiADPDialogStrings
* 1 0x00007febeac5662f [libpthread.so.0] _L_unlock_13 [funlockfile.c:?]
*
* loaded modules:
* 0x00007feb4b7b5000 libai.so
* 0x00007febeac47000 libpthread.so.0
*
* memory: VM 36322 MB, RSS 429 MB, 1 page faults
****
Segmentation fault
* 1 0x00007febeac5662f [libpthread.so.0] _L_unlock_13 [funlockfile.c:?]
*
* loaded modules:
* 0x00007feb4b7b5000 libai.so
* 0x00007febeac47000 libpthread.so.0
*
* memory: VM 36322 MB, RSS 429 MB, 1 page faults
****
Segmentation fault
same crash no matter the script and no problem with 0.58.x versions, so I guess it's from gaffer itself and not the scripts
Bests
Daniel Dresser
May 4, 2021, 4:33:54 PM5/4/21
to gaffer-dev
Hmm, I regularly open scripts that way on Gaffer 0.59 on Centos 7 ( though in the IE env we use a lot of custom libs instead of default Centos 7 ). It could be about the environment, but the first thing to narrow down is whether it's the script.
You say it crashes "no matter the script" - have you tried making a new empty script and immediately saving that and opening it, to narrow down whether it could possibly have anything to do with the script context?
The other thing to try would be setting the env var GAFFER_DEBUG to 1, so that it starts in GDB ( you then have to type "run" to start executing it ). That will hopefully generate a more informative stack trace.
William Eguienta
May 12, 2021, 8:02:48 AM5/12/21
to gaffer-dev
Hi,
here is the feedback with gaffer debug on
also, when I try to open alembic file with custom attribute from blender, I have this seg fault :
here is the log : https://gist.github.com/Kaiz3rTool/43a4f476c2e2f22e45cd3fe6899979f1
here is the log : https://gist.github.com/Kaiz3rTool/43a4f476c2e2f22e45cd3fe6899979f1
the alembic works well in maya and others
here is an example abc https://fromsmash.com/pSjNZOJkFm-ct
here is an example abc https://fromsmash.com/pSjNZOJkFm-ct
Cheers
John Haddon
May 12, 2021, 10:19:31 AM5/12/21
to gaffe...@googlegroups.com
Thanks William,
Unfortunately those stack traces aren't particularly illuminating. I tried loading the Alembic file here and it worked fine in 0.59.6.0 (I also tested 0.59.7.0 and 0.58.6.0, which were both OK as well). Could you confirm that it still crashes at your end with a completely vanilla installation of Gaffer, without custom plugins etc?
Cheers...
John
To view this discussion on the web visit https://groups.google.com/d/msgid/gaffer-dev/656d06bf-9887-4a46-a9b1-468d793e96abn%40googlegroups.com.
William Eguienta
May 13, 2021, 5:50:13 AM5/13/21
to gaffer-dev
Hi John,
yes I have tested with vanilla install (only active plugin is arnold)
and I still have the same problem
I've verified the abc sended and this one is not problematic, sorry, here is the faulty one from blender
and I still have the same problem
I've verified the abc sended and this one is not problematic, sorry, here is the faulty one from blender
John Haddon
May 13, 2021, 1:03:28 PM5/13/21
to gaffe...@googlegroups.com
On Thu, May 13, 2021 at 10:50 AM William Eguienta <kaiz3...@gmail.com> wrote:
I've verified the abc sended and this one is not problematic, sorry, here is the faulty one from blender
Thanks, the new file crashes for me perfectly, so hopefully I'll have a fix before too long...
William Eguienta
May 13, 2021, 1:07:58 PM5/13/21
to gaffe...@googlegroups.com
Thanks John !
Bests
William Eguienta
Creative Director / VFX supervisor
06 88 700 364
De : gaffe...@googlegroups.com <gaffe...@googlegroups.com> de la part de John Haddon <jo...@gafferhq.org>
Envoyé : Thursday, May 13, 2021 7:03:15 PM
À : gaffe...@googlegroups.com <gaffe...@googlegroups.com>
Objet : Re: [gaffer] gaffer crash
Envoyé : Thursday, May 13, 2021 7:03:15 PM
À : gaffe...@googlegroups.com <gaffe...@googlegroups.com>
Objet : Re: [gaffer] gaffer crash
--
You received this message because you are subscribed to a topic in the Google Groups "gaffer-dev" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/gaffer-dev/1SyrmxrW9FU/unsubscribe.
To unsubscribe from this group and all its topics, send an email to gaffer-dev+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/gaffer-dev/CAB8pVgKWRDuOhMsjMYHi-b5qSXWxjRrbyVmvKOE1Koo16eWp%3DA%40mail.gmail.com.
You received this message because you are subscribed to a topic in the Google Groups "gaffer-dev" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/gaffer-dev/1SyrmxrW9FU/unsubscribe.
To unsubscribe from this group and all its topics, send an email to gaffer-dev+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/gaffer-dev/CAB8pVgKWRDuOhMsjMYHi-b5qSXWxjRrbyVmvKOE1Koo16eWp%3DA%40mail.gmail.com.
John Haddon
May 13, 2021, 1:17:17 PM5/13/21
to gaffe...@googlegroups.com
On Thu, May 13, 2021 at 6:07 PM William Eguienta <kaiz3...@gmail.com> wrote:
Thanks John !
The crash is triggered by non-scalar properties in the `.userProperties` of an object, so for instance `/Cube.userProperties.static_int` which is an ArrayProperty. It should be trivial to avoid the crash, but we don't currently have support for loading non-scalar properties. Is the crash fix useful on its own, or will it be useless if we don't add the loading code as well?
William Eguienta
May 13, 2021, 1:25:56 PM5/13/21
to gaffe...@googlegroups.com
I guess that just avoiding the crash can help to understand what Works or not from blender to gaffer.
The same alembic works in maya but will not shows all userProperties
I will write to blender dev about this to see if they can fix it on their side
Cheers
William Eguienta
Creative Director / VFX supervisor
06 88 700 364
De : gaffe...@googlegroups.com <gaffe...@googlegroups.com> de la part de John Haddon <jo...@gafferhq.org>
Envoyé : jeudi, mai 13, 2021 7:17 PM
À : gaffe...@googlegroups.com
À : gaffe...@googlegroups.com
Objet : Re: [gaffer] gaffer crash
On Thu, May 13, 2021 at 6:07 PM William Eguienta <kaiz3...@gmail.com> wrote:
Thanks John !
The crash is triggered by non-scalar properties in the `.userProperties` of an object, so for instance `/Cube.userProperties.static_int` which is an ArrayProperty. It should be trivial to avoid the crash, but we don't currently have support for loading
non-scalar properties. Is the crash fix useful on its own, or will it be useless if we don't add the loading code as well?
--
You received this message because you are subscribed to the Google Groups "gaffer-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
gaffer-dev+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/gaffer-dev/CAB8pVgJazc8%3D68c7aMDBEvv8Cwg57kH_NXnBWoj-jgGkUo7WAg%40mail.gmail.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/gaffer-dev/CAB8pVgJazc8%3D68c7aMDBEvv8Cwg57kH_NXnBWoj-jgGkUo7WAg%40mail.gmail.com.
John Haddon
May 14, 2021, 10:59:03 AM5/14/21
to gaffe...@googlegroups.com
OK, I have a fix for this now - the bug is in Cortex so it may not necessarily make it into the next Gaffer release though. You can follow its progress here : https://github.com/ImageEngine/cortex/pull/1156.
To view this discussion on the web visit https://groups.google.com/d/msgid/gaffer-dev/DB6PR03MB3080A1F1A712A2384EA71DB1A2519%40DB6PR03MB3080.eurprd03.prod.outlook.com.
Reply all
Reply to author
Forward
0 new messages