[sldev] [PATCH] changes I made to run snowglobe on Fedora 11 x86_64

I was able to get snowglobe-src-viewer-1.0.2-r2451.tar.gz to compile and run on a Fedora 11 x86_64 system with the following patches. Fixed compiler warnings since I didn't know how to turn off gcc -Werror in cmake. Fixed viewer_manifest.py to copy secondlife-stripped to the right location for x8

Re: [sldev] Emerging top crashers in Snowglobe 1.1

Robert may be on to something here. The snowglobe client can use the same directory as the production viewer. However all settings for all clients are all in the same roaming folder and changes to stuff in one account affects all accounts. This tends to make me think the SL client is only safe

Re: [sldev] More crash rate statistics

Can I say not again on this list? In reference to another teleport memory leak Not again. I've also had issues with having certain textures being able to crash the viewer although in my crashes I think it was accidental rather than griefer attacks but if there isn't already there should be a J

Re: [sldev] Something is up with crash rate

On Tue, Aug 4, 2009 at 4:30 PM, Rob Lanphier wrote: > > #*** 0 = Previous session ended normally > #*** 1 = Froze > #*** 2 = Fatal Error > #*** 3 = Crash detected > #*** 4 = Froze after logout > #*** 5 = Crash after logout If this is only data being sent through official reports from the crash

Re: [sldev] More crash rate statistics

I am using the snowglobe test build 1.1 and have had it running over 40 hours non stop. but the crash i have had was do to this list of types. 1. bulk load time out crash.. trying to import textures to SL after texture edit out of sl. 2. missing or unknown texture attack by a griefer 3. invisi g

Re: [sldev] Emerging top crashers in Snowglobe 1.1

The sample sizes differ very much. The differences could be (partially) caused by different populations, e.g. developers may crash (more often) on purpose by checking a whether a bug (still) exists/ crash less often by avoiding a known crash bug because they are interested in something else. Ea

Re: [sldev] Emerging top crashers in Snowglobe 1.1

Sideways issue to this is How is the profile data folder controlled?? Some of the issue may be from folks running multiple clients with a single profile folder (on my own system i have SL 1.23 , SnowGlobe 1.1.2?? and Emerald all pointed at the main secondlife profile data folder (oh and until vwr-

Re: [sldev] Emerging top crashers in Snowglobe 1.1

What I'd like to know if is those people cleared their cache before starting the new viewer. Or, if the crashes persist if they do. On Wed, Aug 05, 2009 at 10:21:21AM -0700, Rob Lanphier wrote: > Hi folks, > > More crash data has come in, and these two are emerging as the top crashers: > > LLTe

Re: [sldev] Something is up with crash rate

Thanks for the links and correcting me, Rob :) On Wed, Aug 05, 2009 at 10:27:51AM -0700, Rob Lanphier wrote: > A couple of comments: > 1. You don't need to compile your own viewer to help in this way. > Older versions are available here: > https://lists.secondlife.com/pipermail/sldev-commits/200

Re: [sldev] Something is up with crash rate

On 08/05/2009 03:53 AM, Carlo Wood wrote: > On Wed, Aug 05, 2009 at 10:42:15AM +, tinselsilv...@frontiernet.net wrote: > >> I can only speak of my own experience. I had no reportable issues with >> Release 1.0.2.2451 when I first used it. With the next two releases my >> experience got pr

[sldev] Emerging top crashers in Snowglobe 1.1

Hi folks, More crash data has come in, and these two are emerging as the top crashers: LLTextureFetchWorker::callbackCacheRead: ASSERT (imagesize >= 0) 10 crashes by 6 agents: indra/newview/lltexturefetch.cpp:1233 http://jira.secondlife.com/browse/SNOW-170 LLViewerImage::doLoadedCallbacks: ASSER

[sldev] SLURL to SLurl - an easy task?

Hi there! Maybe this is an easy task for someone who can run a search and replace on the sourcecode and ensure that it doesn't bork something: VWR-14997 - Change SLURL to SLurl in the UI to sync documentation with viewer --zai __

Re: [sldev] Something is up with crash rate

On Wed, Aug 05, 2009 at 10:42:15AM +, tinselsilv...@frontiernet.net wrote: > I can only speak of my own experience. I had no reportable issues with > Release 1.0.2.2451 when I first used it. With the next two releases my > experience got progressively worse. Crashing out, slow rezz, slow loa

Re: [sldev] Something is up with crash rate

I can only speak of my own experience. I had no reportable issues with Release 1.0.2.2451 when I first used it. With the next two releases my experience got progressively worse. Crashing out, slow rezz, slow load, etc. I went back to the Release 1.0.2.2451 and the only two issues I'm currently

Re: [sldev] Something is up with crash rate

On Tue, 04 Aug 2009 13:30:48 -0700 Rob Lanphier wrote: > Crash rates per 100 hours: > > Snowglobe Release 1.0.2.2451 12.26 > Snowglobe Release 1.0.3.2537 11.25 > Snowglobe Release 1.1.2.2584 13.76 > Snowglobe Test Build 1.1.0.2558 5.82 > Snowglobe Test Build 1.1.0.2565