[Opensim-users] Server causing viewers to crash
Drover, Matt
mattdrover at mun.ca
Wed Mar 26 10:53:26 UTC 2014
After some research, I think it happened when I transferred my database
form mssql to mysql. Almost positive.
Thanks for all the help Justin. We still want to hire you when you have
time, I'll check back in a few months.
-Matt
-----Original Message-----
From: opensim-users-bounces at lists.berlios.de
[mailto:opensim-users-bounces at lists.berlios.de] On Behalf Of Justin
Clark-Casey
Sent: Monday, March 24, 2014 10:36 PM
To: opensim-users at lists.berlios.de
Subject: Re: [Opensim-users] Server causing viewers to crash
I don't think there should be a priblem with nulling those entries. I'd
love to know how those became corrupt in the first place but that's
going to be hard to pin down after the fact. I think other people may
have suffered the same problem though I don't think it's very common.
On 21/03/14 14:16, Drover, Matt wrote:
> So it seems I do have curroption in my partical data. I tried to
> export my oar as saw the errors below.
>
> Looking at my db, I see that prims.ParticleSystem has the following
> lengths.
>
> 0 - these prims work fine
> 1 - these prims are curropted in my export and crashes my viewer
> 86 - these prims have actual particle scripts.
>
> I did a test and NULL'd a few of those 1-length prims and it fixed my
> issue. Do you see any problems with my NULL'ing all the 1-length
prims?
>
> -Matt
>
> Export Error:
>
> Line 9: 2014-03-21 11:12:09,173 WARN -
> OpenSim.Region.Framework.Scenes.UuidGatherer [UUID GATHERER]: Could
> not check particle system for part orange01
> d855a386-9ed9-4288-a1fb-4f02b2911b55 in object barrel_oranges
> 0b926d04-df71-4c46-9b9f-a5f11fea0bfd since it is corrupt. Continuing.
> Line 10: 2014-03-21 11:12:09,175 WARN -
> OpenSim.Region.Framework.Scenes.UuidGatherer [UUID GATHERER]: Could
> not check particle system for part orange02
> 6719bfd3-3b91-464c-9af3-5c6cfcc4ec1e in object barrel_oranges
> 0b926d04-df71-4c46-9b9f-a5f11fea0bfd since it is corrupt. Continuing.
> Line 11: 2014-03-21 11:12:09,178 WARN -
> OpenSim.Region.Framework.Scenes.UuidGatherer [UUID GATHERER]: Could
> not check particle system for part Primitive
> 5df5f8ed-de7e-42e8-b8f3-7e97af4e153c in object barrel_oranges
> 0b926d04-df71-4c46-9b9f-a5f11fea0bfd since it is corrupt. Continuing.
> Line 12: 2014-03-21 11:12:09,179 WARN -
> OpenSim.Region.Framework.Scenes.UuidGatherer [UUID GATHERER]: Could
> not check particle system for part Rug v4
> 131ce5d2-e5b1-4fd7-971f-a733d6cd3766 in object Rug v4
> 131ce5d2-e5b1-4fd7-971f-a733d6cd3766 since it is corrupt. Continuing.
> Line 13: 2014-03-21 11:12:09,183 WARN -
> OpenSim.Region.Framework.Scenes.UuidGatherer [UUID GATHERER]: Could
> not check particle system for part Cuboard_small
> 1cd66f19-9621-4ee2-82c8-9a309ab99009 in object Cuboard_small
> 1cd66f19-9621-4ee2-82c8-9a309ab99009 since it is corrupt. Continuing.
> Line 14: 2014-03-21 11:12:09,185 WARN -
> OpenSim.Region.Framework.Scenes.UuidGatherer [UUID GATHERER]: Could
> not check particle system for part Cuboard
> 331c99d3-8563-410b-bebf-4fea6f80b8cf in object Cuboard_small
> 1cd66f19-9621-4ee2-82c8-9a309ab99009 since it is corrupt. Continuing.
> Line 15: 2014-03-21 11:12:09,187 WARN -
> OpenSim.Region.Framework.Scenes.UuidGatherer [UUID GATHERER]: Could
> not check particle system for part Cuboard
> bc3e158e-f9a1-4ae6-914b-dace73c4b4ea in object Cuboard_small
> 1cd66f19-9621-4ee2-82c8-9a309ab99009 since it is corrupt. Continuing.
> Line 16: 2014-03-21 11:12:09,189 WARN -
> OpenSim.Region.Framework.Scenes.UuidGatherer [UUID GATHERER]: Could
> not check particle system for part Cuboard
> cd584f59-2e04-4c62-bb9b-b8850e62fc46 in object Cuboard_small
> 1cd66f19-9621-4ee2-82c8-9a309ab99009 since it is corrupt. Continuing.
> Line 17: 2014-03-21 11:12:09,191 WARN -
> OpenSim.Region.Framework.Scenes.UuidGatherer [UUID GATHERER]: Could
> not check particle system for part Cuboard
> ebc9a798-8b8e-4977-a6c4-30df67ba10de in object Cuboard_small
> 1cd66f19-9621-4ee2-82c8-9a309ab99009 since it is corrupt. Continuing.
> Line 18: 2014-03-21 11:12:09,193 WARN -
> OpenSim.Region.Framework.Scenes.UuidGatherer [UUID GATHERER]: Could
> not check particle system for part Cuboard
> 4ef18c97-ed54-42d5-b4cd-cf977add797c in object Cuboard_small
> 1cd66f19-9621-4ee2-82c8-9a309ab99009 since it is corrupt. Continuing.
> Line 19: 2014-03-21 11:12:09,195 WARN -
> OpenSim.Region.Framework.Scenes.UuidGatherer [UUID GATHERER]: Could
> not check particle system for part Cuboard
> ede54df0-9d25-42c5-b33a-f7fe3d8047ce in object Cuboard_small
> 1cd66f19-9621-4ee2-82c8-9a309ab99009 since it is corrupt. Continuing.
> Line 20: 2014-03-21 11:12:09,197 WARN -
> OpenSim.Region.Framework.Scenes.UuidGatherer [UUID GATHERER]: Could
> not check particle system for part Cuboard
> 6ad2ce2e-069e-4cac-945e-13b5134219d3 in object Cuboard_small
> 1cd66f19-9621-4ee2-82c8-9a309ab99009 since it is corrupt. Continuing.
> Line 21: 2014-03-21 11:12:09,198 WARN -
> OpenSim.Region.Framework.Scenes.UuidGatherer [UUID GATHERER]: Could
> not check particle system for part Cuboard
> 06f25e74-2b93-4ab1-ac14-bfabdebba470 in object Cuboard_small
> 1cd66f19-9621-4ee2-82c8-9a309ab99009 since it is corrupt. Continuing.
>
> -----Original Message-----
> From: opensim-users-bounces at lists.berlios.de
> [mailto:opensim-users-bounces at lists.berlios.de] On Behalf Of Justin
> Clark-Casey
> Sent: Thursday, March 20, 2014 9:12 PM
> To: opensim-users at lists.berlios.de
> Subject: Re: [Opensim-users] Server causing viewers to crash
>
> 0.7.6 was September 2013 so those changes won't be there. It would be
> very odd that 500 objects are causing the issue, unless someone has
> duplicated a lot of stuff.
>
> Matt, have you definitely eliminated some kind of disk corruption?
> Your problem with mysql timeout during initial load of some regions
> made me think that you might have some underlying issue.
>
> Failing that, how do you know these objects have no particle system
> data? They have none in the database itself?
>
> On 20/03/14 20:18, Dahlia Trimble wrote:
>> I added some code to support the new particle system features such as
>> ribbon and alpha blending and glow, back around last November 6th and
>> 7th. I don't remember when 0.7.6 was released but I thought those
>> changes made it in. There were also changes to most third-party
> viewers to support those features. I'm not aware of any crashes caused
> specifically by these changes but if you have an object or a particle
> script which can reproduce it, could you get it to me somehow?
>>
>> -dahlia
>>
>>
>> On Thu, Mar 20, 2014 at 10:32 AM, Drover, Matt <mattdrover at mun.ca
> <mailto:mattdrover at mun.ca>> wrote:
>>
>> Hi, your code fix of " update.PSBlock = Utils.EmptyBytes;"
worked.
> My
>> world is up and running again. To start to solve the problem
> properly, I
>> recoded some of singularity to output the prims causing the
issue.
> There
>> is about 500 of them. Oddly, none of them had any particle
>> system
> data.
>> I tested editing an item and closing it and it solves the
problem.
> But
>> how to do this for 500 items?
>>
>> I realize this started happening after the upgrade form 0.7.4 to
> 0.7.6.
>>
>> Any thoughts?
>>
>> -Matthew
>>
>> -----Original Message-----
>> From: opensim-users-bounces at lists.berlios.de
> <mailto:opensim-users-bounces at lists.berlios.de>
>> [mailto:opensim-users-bounces at lists.berlios.de
> <mailto:opensim-users-bounces at lists.berlios.de>] On Behalf Of Justin
>> Clark-Casey
>> Sent: Tuesday, March 11, 2014 5:58 PM
>> To: opensim-users at lists.berlios.de
> <mailto:opensim-users at lists.berlios.de>
>> Subject: Re: [Opensim-users] Server causing viewers to crash
>>
>> Hi Matthew. I haven't seen this before. However, a quick grep
>> of
> the
>> Singularity source code reveals that "syssize"
>> and "whippang" are currently unused and probably dummy names
>> used
> in the
>> unpacking of particle system data (llpartdata.cpp).
>>
>> This suggests to me that the particle system data in one of your
> prims
>> has somehow become corrupt, in such a way that it triggers a
>> crash
> in
>> the viewer.
>>
>> If you are building OpenSimulator, an extremely quick and dirty
> test
>> would be to replace
>>
>> update.PSBlock = data.ParticleSystem ?? Utils.EmptyBytes;
>>
>> in LLClientView.CreatePrimUpdateBlock() with just
>>
>> update.PSBlock = Utils.EmptyBytes;
>>
>> and see if the problem goes away. If it does, then it is a
> particle
>> system issue. Unfortunately, then it's a case of elimination of
>> anything with a particle system (I think you could get a list of
> such
>> prims by looking for non-empty ParticleSystem fields in the
>> prims
> db
>> table).
>>
>> Or if we're unlucky, it could be some other bad data in building
> the
>> prim packet and this is only showing up when the viewer tries to
> unpack
>> the particle system section.
>>
>> On 11/03/14 11:27, Drover, Matt wrote:
>> > Somehow, something has changed on my server and all viewers
are
>> crashing when they connect to my grid.
>> >
>> > I'm running opensim v0.7.6 in grid mode. I have 15 regions
> spanned
>> > across 5 opensim instances. All on the same pc. The pc is
> windows
>> server 2008 r2 and is a vm.
>> >
>> > I tested many viewers, all with the same results. Here are
>> the
> errors
>> from the singulatiry viewer log.
>> >
>> > 2014-03-10T14:20:23Z WARNING:
> LLDataPackerBinaryBuffer::verifyLength:
>> > Buffer overflow in BinaryBuffer length verify, field name
crc!
>> >
>> > 2014-03-10T14:20:23Z WARNING:
> LLDataPackerBinaryBuffer::verifyLength:
>> > Current pos: 0 Buffer size: 1 Data size: 4
>> >
>> > 2014-03-10T14:20:23Z WARNING:
> LLDataPackerBinaryBuffer::verifyLength:
>> > Buffer overflow in BinaryBuffer length verify, field name
> syssize!
>> >
>> > 2014-03-10T14:20:23Z WARNING:
> LLDataPackerBinaryBuffer::verifyLength:
>> > Current pos: 0 Buffer size: 1 Data size: 4
>> >
>> > 2014-03-10T14:20:23Z WARNING:
> LLDataPackerBinaryBuffer::verifyLength:
>> > Buffer overflow in BinaryBuffer length verify, field name
> whippang!
>> >
>> > 2014-03-10T14:20:23Z WARNING:
> LLDataPackerBinaryBuffer::verifyLength:
>> > Current pos: 4 Buffer size: 1 Data size: 1
>> >
>> > 2014-03-10T14:20:23Z WARNING:
> LLDataPackerBinaryBuffer::verifyLength:
>> > Buffer overflow in BinaryBuffer length verify, field name
> whippang!
>> >
>> > 2014-03-10T14:20:23Z WARNING:
> LLDataPackerBinaryBuffer::verifyLength:
>> > Current pos: 5 Buffer size: 1 Data size: 1
>> >
>> > 2014-03-10T14:20:23Z WARNING:
> LLDataPackerBinaryBuffer::verifyLength:
>> > Buffer overflow in BinaryBuffer length verify, field name
> whippang!
>> >
>> > 2014-03-10T14:20:23Z WARNING:
> LLDataPackerBinaryBuffer::verifyLength:
>> > Current pos: 6 Buffer size: 1 Data size: 1
>> >
>> > 2014-03-10T14:20:23Z WARNING:
> LLDataPackerBinaryBuffer::verifyLength:
>> > Buffer overflow in BinaryBuffer length verify, field name
> whippang!
>> >
>> > The last 2 lines repeat about 1000 more times then the viewer
> crashes.
>> >
>> > Any ideas?
>> >
>> > *Matthew Drover
>> > *Programmer Consultant
>> >
>> > Immersive Technology
>> > Distance Education, Learning and Teaching Support
>> >
>> >
>> >
>> > _______________________________________________
>> > Opensim-users mailing list
>> > Opensim-users at lists.berlios.de
> <mailto:Opensim-users at lists.berlios.de>
>> > https://lists.berlios.de/mailman/listinfo/opensim-users
>> >
>>
>>
>> --
>> Justin Clark-Casey (justincc)
>> OSVW Consulting
>> http://justincc.org
>> http://twitter.com/justincc
>> _______________________________________________
>> Opensim-users mailing list
>> Opensim-users at lists.berlios.de
> <mailto:Opensim-users at lists.berlios.de>
>> https://lists.berlios.de/mailman/listinfo/opensim-users
>> _______________________________________________
>> Opensim-users mailing list
>> Opensim-users at lists.berlios.de
> <mailto:Opensim-users at lists.berlios.de>
>> https://lists.berlios.de/mailman/listinfo/opensim-users
>>
>>
>>
>>
>> _______________________________________________
>> Opensim-users mailing list
>> Opensim-users at lists.berlios.de
>> https://lists.berlios.de/mailman/listinfo/opensim-users
>>
>
>
> --
> Justin Clark-Casey (justincc)
> OSVW Consulting
> http://justincc.org
> http://twitter.com/justincc
> _______________________________________________
> Opensim-users mailing list
> Opensim-users at lists.berlios.de
> https://lists.berlios.de/mailman/listinfo/opensim-users
> _______________________________________________
> Opensim-users mailing list
> Opensim-users at lists.berlios.de
> https://lists.berlios.de/mailman/listinfo/opensim-users
>
--
Justin Clark-Casey (justincc)
OSVW Consulting
http://justincc.org
http://twitter.com/justincc
_______________________________________________
Opensim-users mailing list
Opensim-users at lists.berlios.de
https://lists.berlios.de/mailman/listinfo/opensim-users
More information about the Opensim-users
mailing list