Oskar Linden Posted June 25, 2012 Share Posted June 25, 2012 This week we have infrastructure changes rolling to the main channel and then the Pathfinding beta on RC. UPDATE: There were issues with Pathfinding and we are putting AvTextures project on Magnum and LeTigre instead. Magnum will be rolled off of Pathfinding to the AvTextures project. Second Life Server (main channel) We are promoting the infrastructure project that was on Magnum. This contains new simulator changes to help with infrastructure. Bug Fixes: SVC-8000: Noticed 06-14-12 on LeTigre server in Cleary. LSL function llGetWallclock() displays number of seconds since Zulu (GMT) midnite instead of Pacific midnite SVC-8004: Region server names now include backslashes, apparently escaping the space that follows. SVC-7598: Error fetching server release notes URL. 2012-06-26, 5:00am: Release Notes: https://wiki.secondlife.com/wiki/Release_Notes/Second_Life_Server/12 Second Life RC BlueSteel This channel will have the simulator changes to process prebaked avatar textures for library outfits. This is the same code that was on here last week. This code is on all RC channels. 2012-06-27, 7-11:00am: Release Notes: https://wiki.secondlife.com/wiki/Release_Notes/Second_Life_RC_BlueSteel/12 Second Life RC LeTigre This channel will have the simulator changes to process prebaked avatar textures for library outfits. This is the same code as is on BlueSteel and Magnum. 2012-06-27, 7-11:00am: Release Notes: https://wiki.secondlife.com/wiki/Release_Notes/Second_Life_RC_LeTigre/12 Second Life RC Magnum This channel will have the simulator changes to process prebaked avatar textures for library outfits. This is the same code as is on BlueSteel and LeTigre. Magnum did have Pathfinding but it was removed when blocking issues were found. 2012-06-27, 7-11:00am: Release Notes: https://wiki.secondlife.com/wiki/Release_Notes/Second_Life_RC_Magnum/12 We will be monitoring this thread during the next week so please feel free to post issues that you feel have been introduced by the new code. Please file a JIRA for issues you find and post the JIRA link into this thread. It really helps us out. When determining if issues are relevant or not research is key. Tracking down exactly the right situation where an issue is occurring greatly speeds up the development process to get fixes in place. I appreciate your help. Have a good week! __Oskar p.s. If you are interested in helping test SecondLife in beta please join the group "Second Life Beta" in-world. We also have an email list where we communicate upcoming projects and how you can help. (https://lists.secondlife.com/cgi-bin/mailman/listinfo/server-beta ) Once a week we meet on ADITI to discuss new features, new bugs, new fixes, and other fun stuff. You are more than welcome. Information is here:https://wiki.secondlife.com/wiki/Server_Beta_User_Group Link to comment Share on other sites More sharing options...
Ayesha Askham Posted June 25, 2012 Share Posted June 25, 2012 Oskar Welcome back to the madhouse. I hope your break de-stressed you. Now... SVC-7598...that was supposed to be fixed on LeTigre last week. It wasn't, at least for Firestorm 4.0.1.27000 on Sim 21103 Woods of Heaven it wasn't. How come? Edit Tuesday 26th June 2:47pm BST: Hello? Re: SVC-7598,what part of "It's not fixed" do Linden Lab not understand, Oskar? Oh, and the software let me in again without having to log in. Not good. Link to comment Share on other sites More sharing options...
Lydia Craig Posted June 26, 2012 Share Posted June 26, 2012 Blue Steel servers desperately need relief from the crashes and teleport failures and lag that has been plauging them for over a month now, and if it does not come soon you are going to see a mass exodus of businesses closing in those areas and accounts closing with them. Right now this are is in desperate trouble and if it does not get help soon whatever project you are working on there will not matter because there will be few there to test it. Link to comment Share on other sites More sharing options...
Cincia Singh Posted June 26, 2012 Share Posted June 26, 2012 Lydia, I am intrigued. I haven't found anywhere in SL that behaves like you describe. Could you post some region names so I can go experience all these difficulties you describe? Link to comment Share on other sites More sharing options...
AnnMarie Otoole Posted June 26, 2012 Share Posted June 26, 2012 I'm seeing error messages:- Can't move object "Vehicle Name" to { "Vector here "} in region (Region Name) because the other region is running an older version which does not support receiving this object via region crossing. Notifications from this region are temporarily deactivated Is there a change coming in sim crossings? Link to comment Share on other sites More sharing options...
Ansariel Hiller Posted June 26, 2012 Share Posted June 26, 2012 That's caused by pathfinding. One of the list of long-known and still unfixed nasty bugs in pathfinding... Link to comment Share on other sites More sharing options...
Darien Caldwell Posted June 26, 2012 Share Posted June 26, 2012 Ansariel Hiller wrote: That's caused by pathfinding. One of the list of long-known and still unfixed nasty bugs in pathfinding... That's not a bug, it's just an incompatibility between two server versions. Once the whole grid has Pathfinding, it will magically disappear. :matte-motes-big-grin: Link to comment Share on other sites More sharing options...
Ansariel Hiller Posted June 26, 2012 Share Posted June 26, 2012 But the bugs will stay... Link to comment Share on other sites More sharing options...
Qie Niangao Posted June 26, 2012 Share Posted June 26, 2012 Bug or not, it's immensely disruptive, and it's new. At least in Bay City, I can point to exactly when it started: yesterday, at 5pm Pacific. Are we to take it that the main server release to be deployed this week will also not be compatible with their PF neighbors on the main grid? If so, is there some way we should be okay with that? [ETA: Mentioned this at the Server UG, and opened a jira for tracking. See also parallel thread.] Link to comment Share on other sites More sharing options...
Abigail Merlin Posted June 26, 2012 Share Posted June 26, 2012 My sim Safe Kitten is not comming back up, please check. Link to comment Share on other sites More sharing options...
Perrie Juran Posted June 26, 2012 Share Posted June 26, 2012 Cincia Singh wrote: Lydia, I am intrigued. I haven't found anywhere in SL that behaves like you describe. Could you post some region names so I can go experience all these difficulties you describe? Interestingly enough, I don't believe I have ever seen Lydia (and Lydia, please correct me if I am wrong) post her system specs, which viewer she uses, her internet connection, results of speed and ping tests, etc etc. It's always, "I am having major Linden Lab caused problems" and/or "I am hearing reports of others having major problems." I don't believe I have ever seen a shred of documentation from her. For all we know she could be trying to log in on an old Tandy computer with a 300 Baud modem. Link to comment Share on other sites More sharing options...
Oskar Linden Posted June 26, 2012 Author Share Posted June 26, 2012 Ansariel Hiller wrote: That's caused by pathfinding. One of the list of long-known and still unfixed nasty bugs in pathfinding... It's not a bug. It is an incompatability between havok versions. Havok is what models physics on the simulator side. Mesh objects (vehicles made of mesh or maybe like mesh bullets) will give that error when going 'backwards in versions'. So if a mesh object goes from a Pathfinding region to a non Pathfinding region the older version of Havok cannot handle it. This issue will be resolved once the Pathfinding code is live everywhere. Until then it will be an awkward experience and we apologize for that. __Oskar Link to comment Share on other sites More sharing options...
Ayesha Askham Posted June 26, 2012 Share Posted June 26, 2012 @Oskar What Anne-Marie and Ansariel are talking about may not be a bug to you guys at The Lab, but that is not how it is seen out here in virtual land. I suspect that by the end of tomorrow's LeTigre roll, we will have a whole host of bugs and unexpected behaviours to relate. Still there's no gain without pain, they say. Soon Norman Wisdom will be whizzing around our sim like there's no tomorrow. "Mr Grimsdale????" (with apologies to Inara Pey). Oh Yippee. It would've been nice to see some acknowledgement of the failure of the SVC-7598 fix, but I guess there is another agenda at play there. If it has NOT failed, why do we not see any change in our viewers? Link to comment Share on other sites More sharing options...
Qie Niangao Posted June 27, 2012 Share Posted June 27, 2012 If it were limited to Mesh vehicles only, that would be understandable, and indeed it may have been working that way in the past. But now, every scripted object is trapped at the border, Mesh or no. Link to comment Share on other sites More sharing options...
Darien Caldwell Posted June 27, 2012 Share Posted June 27, 2012 I have found a new issue which has arisen in the last two days, I can only guess it's related to today's Deploy, but it seems to affect all sims, even those not rolled today (Verified on Le Tigre, Bluesteel, Magnum, and the Release Channel): https://jira.secondlife.com/browse/SVC-8024 It's not a serious issue by any means, but certainly causing some confusion. FYI Link to comment Share on other sites More sharing options...
WolfBaginski Bearsfoot Posted June 27, 2012 Share Posted June 27, 2012 According to the Status report, it took 12 hours to do the Main Channel roll-out.Even allowing for some possible delay in posting the Completion message, that's a long time. I also saw regions indicated as down on the map, which is unusual. The in-world Map system doesn't detect these things rapidly. Most slip between the scane. I can't be certain, but it looked like they had a bad day on Tuesday. I am not lookinf forward to what is looming Link to comment Share on other sites More sharing options...
Ayesha Askham Posted June 27, 2012 Share Posted June 27, 2012 @Wolf ahh, there is no end to their power. The Main Server roll took 12 hours, it seems, and yet that post was not visible to me until a few moments ago. They did NOT post that "completed" notice at the time it states...I looked earlier today (BST) and despite several refreshes it simply was NOT there. Also I now see that this week's RC rolls actually happened last week: "[Posted 9:40pm PDT, 26 June 2012] We will be performing rolling restarts for regions on the Magnum, BlueSteel, and LeTigre RC server channels Wednesday, June 20th starting at 07:00am PDT. Please save all builds and refrain from rezzing no copy objects or making inworld L$ transactions at that time." I can no longer trust this outfit to get even their dates right so what chance is there that the complex software is right, eh? Link to comment Share on other sites More sharing options...
Lindens Maestro Linden Posted June 27, 2012 Lindens Share Posted June 27, 2012 Hi Qie, I imported https://jira.secondlife.com/browse/SVC-8023 after confirming that the crossing failure from PF into non-PF applies to non-mesh objects as well - this is a bug. Link to comment Share on other sites More sharing options...
Oskar Linden Posted June 27, 2012 Author Share Posted June 27, 2012 Maestro Linden wrote: Hi Qie, I imported https://jira.secondlife.com/browse/SVC-8023 after confirming that the crossing failure from PF into non-PF applies to non-mesh objects as well - this is a bug. Because of this issue we are rolling Pathfinding off of Magnum. We had not yet rolled LeTigre. ALl channels will have the AvTextures project that was on BlueSteel. __Oskar Link to comment Share on other sites More sharing options...
Qie Niangao Posted June 27, 2012 Share Posted June 27, 2012 Thanks. That's a big relief. Link to comment Share on other sites More sharing options...
Miranda Umino Posted June 28, 2012 Share Posted June 28, 2012 We do some rollbacks on the test servers now ? As people are so scared about a JIRA , why not to do rollback on on the aditi servers tomorow ? And after tomorrow , why not to do rollbacks on developpers servers ? And why not to ask to Havok to rollback their product too ? ... And this , only because this incompatibilty is an incompatibilty who can happen while the deployement but who could n t happen after a complete deployement ? This situation is absurd . There are too many whiners in second life Link to comment Share on other sites More sharing options...
WolfBaginski Bearsfoot Posted June 28, 2012 Share Posted June 28, 2012 What is really worrying me is that these roll-outs are so incredibly slow, compred to the recent past. I can see reasons why rolling out the PF code might be slow, with a need to create the "Navmesh" for each region, but that doesn't explain the Main Channel roll-out that took 12 hours, when it apparently took 3.5 hours last week. If this slower roll-out process is going to be usual, with some regions down for over an hour, from my timezone I might as well forget using SL on two days a week. If this is a side effect of the PF system, if it is something we are going to be stuck with, I have to wonder if the person who signed off on this is playing with a full deck. Link to comment Share on other sites More sharing options...
Ayesha Askham Posted June 28, 2012 Share Posted June 28, 2012 @Miranda I will try to explain to you simply, since it seems that English is not your first language. The reason for the rollback is because the new version of Havok is not behaving as expected on the sims that it is deployed. It is affecting non-mesh objects in a way that was not expected. It is not safe to continue Secondlife with those errors ocurring and the Linden Lab Deploy Managers took that view. Incompatibility between versions is not unusual but should not occur as it does in this case. Changes in physics is a big change in any simulation and clearly needs more preliminary study. The Beta Grid, Aditi, is just that: a Beta Grid to be used for testing. New code, involving many things, is tested there prior to deployment on the Main Agni grid. Decisions taken there will, I have no doubt, be discussed in the Beta Server Group meetings: https://wiki.secondlife.com/wiki/Server_Beta_User_Group Those reporting the problems with Havok incompatibility are not whiners, they are doing what every user of Secondlife should do, and that is reporting unexpected results. That is the function of the JIRA. Perhaps in your view unfortunately, SecondLife is not simply a software development tool, it is a working Virtual Environment. Therefore disruptions in service are not to be tolerated. I must point out that views expressed in this post are purely mine and that I have no connection with Linden Lab other than that I use their Virtual Environment. As a PS: I also suffer from the same loss of service that Wolf describes...these long Rolls are severely impacting on my use of SL for two days per week. Even in the "Good Old Days" it was only one day! Link to comment Share on other sites More sharing options...
Oskar Linden Posted June 28, 2012 Author Share Posted June 28, 2012 Miranda Umino wrote: We do some rollbacks on the test servers now ? As people are so scared about a JIRA , why not to do rollback on on the aditi servers tomorow ? And after tomorrow , why not to do rollbacks on developpers servers ? And why not to ask to Havok to rollback their product too ? ... And this , only because this incompatibilty is an incompatibilty who can happen while the deployement but who could n t happen after a complete deployement ? This situation is absurd . There are too many whiners in second life I'm not sure I understand this post. __Oskar Link to comment Share on other sites More sharing options...
Oskar Linden Posted June 28, 2012 Author Share Posted June 28, 2012 WolfBaginski Bearsfoot wrote: What is really worrying me is that these roll-outs are so incredibly slow, compred to the recent past. I can see reasons why rolling out the PF code might be slow, with a need to create the "Navmesh" for each region, but that doesn't explain the Main Channel roll-out that took 12 hours, when it apparently took 3.5 hours last week. If this slower roll-out process is going to be usual, with some regions down for over an hour, from my timezone I might as well forget using SL on two days a week. If this is a side effect of the PF system, if it is something we are going to be stuck with, I have to wonder if the person who signed off on this is playing with a full deck. The longer rolling restart is a side effect of some of our infrastructure changes. It will get shorter. We're still tuning the process. We appreciate your patience. __Oskar Link to comment Share on other sites More sharing options...
Facebook
Instagram
Twitter
YouTube
Flickr
Recommended Posts
Please take a moment to consider if this thread is worth bumping.
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now