[11:03] Andrew Hellershanks: Hello, everyone.
[11:03] Selby.Evans @grid.kitely.com:8002: hi everyone
[11:03] Cuga.Rajal @rajal.org:9000: Hi
[11:03] Orbert.Tatham @hg.zetaworlds.com:80: Andrew
[11:04] Andrew Hellershanks: I'm here before Ubit today.
[11:04] Cuga.Rajal @rajal.org:9000: That is way north of me, thankfully
[11:04] Cuga.Rajal @rajal.org:9000: We weren;t sure if there was going to be a meeting
[11:04] Andrew Hellershanks: Hm... I don't see Ubit in IRC.
[11:05] Cuga.Rajal @rajal.org:9000: I missed last week and thought that perhaps there was a plan not to meet?
[11:05] Cuga.Rajal @rajal.org:9000: but you are here :)
[11:05] Kayaker Magic: Plan? There are plans?
[11:05] Vincent.Sylvester @hg.zetaworlds.com:8002: Come hell or high water we usually meet, though it might be high water that has taken Ubit out, he reported flooding yesterday.
[11:06] Cuga.Rajal @rajal.org:9000: oh dear
[11:06] Cuga.Rajal @rajal.org:9000: "save the laptop!"
[11:07] Andrew Hellershanks: I have skipped these meetings when they were the day of, or the day before, Christmas and New Years.
[11:07] Cuga.Rajal @rajal.org:9000: I'm glad I stuck around :)
[11:07] Andrew Hellershanks: Oh, Ubit is experience floods? I hope it doesn't get too bad where he is.
[11:08] Vincent.Sylvester @hg.zetaworlds.com:8002: Dry ground and lots of rain don't mix, quite literally
[11:08] Cuga.Rajal @rajal.org:9000: I have some news to present on Bullet at some point
[11:08] Andrew Hellershanks: ok. You can kick things off today, Cuga.
[11:08] Cuga.Rajal @rajal.org:9000: ok
[11:09] Cuga.Rajal @rajal.org:9000: I'm waiting for a contractor so there is a chance I might have to stop abruptly
[11:09] Cuga.Rajal @rajal.org:9000: ok
[11:09] Andrew Hellershanks: Vincent, frozen ground and lots of melting snow is also a bad combo.
[11:09] Cuga.Rajal @rajal.org:9000: So, there is a lot of activity now behind the scenes on updating Bullet
[11:09] Andrew Hellershanks: ok, Cuga.
[11:10] Kayaker Magic: Is Mr Blue involved?
[11:10] Cuga.Rajal @rajal.org:9000: Misterblue and I have been meeting and working out a plan
[11:10] Kayaker Magic: Cool!
[11:10] Cuga.Rajal @rajal.org:9000: trying to boil it down
[11:10] Cuga.Rajal @rajal.org:9000: We're planning to create a new set of Bullet libraries for all platforms, bring it to the latest version, and add bug fixes
[11:11] Cuga.Rajal @rajal.org:9000: new one will be Bullet 3.25
[11:11] Cuga.Rajal @rajal.org:9000: currently we are at 2.86
[11:11] Cuga.Rajal @rajal.org:9000: there are also some major bug fixes for Bullet that need to be included
[11:12] Cuga.Rajal @rajal.org:9000: those should really be submitted to the Bullet team also
[11:12] Cuga.Rajal @rajal.org:9000: I have a preliminary version to test for Mac and MrBlue os working on Windows and Linux versions
[11:13] Cuga.Rajal @rajal.org:9000: once he has those, we plan to set up some test regions and do some testing
[11:13] Cuga.Rajal @rajal.org:9000: also will need other people to test on their regions
[11:14] Cuga.Rajal @rajal.org:9000: then after testing and working anything out, they will replace the current Bullet libs in trunk
[11:14] Vincent.Sylvester @hg.zetaworlds.com:8002: Mac as in arm M1, are the linux versions going to work for those still with Intel Macs?
[11:14] Cuga.Rajal @rajal.org:9000: Mac libs are going to be universal x86_64 and arm64
[11:14] Cuga.Rajal @rajal.org:9000: probably going to drop 32 bit on Mac
[11:15] Cuga.Rajal @rajal.org:9000: Mac libs wont work on Linux even if same processor
[11:15] Kayaker Magic: Drop? As it stop support?
[11:15] Cuga.Rajal @rajal.org:9000: well unless somebody tells me there are still 32-bit Mac users out there
[11:16] Cuga.Rajal @rajal.org:9000: dotnet6 isnt available for any 32 bit platform
[11:16] Cuga.Rajal @rajal.org:9000: Its possible the old libs will be retained for cases that can't upgrade for some reason
[11:16] Cuga.Rajal @rajal.org:9000: or a platform we no longer have dev capabilities with
[11:16] Cuga.Rajal @rajal.org:9000: but those will be the old versions of Bullet
[11:17] Cuga.Rajal @rajal.org:9000: I'm sure Ubit would chime in on this
[11:17] Orbert.Tatham @hg.zetaworlds.com:80: Dotnet6 isn't "really" available except on Windows
[11:17] Orbert.Tatham @hg.zetaworlds.com:80: Everybody else gets a version on crutches
[11:17] Cuga.Rajal @rajal.org:9000: Dotnet6 works great on Linux and Mac
[11:17] Cuga.Rajal @rajal.org:9000: arm64 and x86_64
[11:17] Andrew Hellershanks: That sounds like a lot of work you and Misterblue have planned, Cuga. Any ideas on how long it will take to get all that work done?
[11:17] Cuga.Rajal @rajal.org:9000: Hard to say
[11:18] Cuga.Rajal @rajal.org:9000: right now MrBlue is having trouble getting the latest Bullet to build, on his system
[11:18] Andrew Hellershanks: You have to the end of the year (if you are on Christmas break) to get it done. ;)
[11:18] Cuga.Rajal @rajal.org:9000: we have been comparing notes
[11:18] Kayaker Magic: lol
[11:18] Cuga.Rajal @rajal.org:9000: but he's only been working at that for about a week and busy with RL stuff
[11:18] Andrew Hellershanks nods
[11:19] Cuga.Rajal @rajal.org:9000: MrBlue can probably talk more about that next time he comes
[11:19] Andrew Hellershanks: Getting code to compile is the first step. I have another project I want to take one for a program no longer supported. I haven't been able to compile it yet either.
[11:19] Cuga.Rajal @rajal.org:9000: Often its just one little obscure change to make a compiler happy
[11:19] Cuga.Rajal @rajal.org:9000: taking many hours to discover solution
[11:19] Andrew Hellershanks: ok. Thank you for the information. That is exciting news for everyone who is using Bullet physics.
[11:20] Cuga.Rajal @rajal.org:9000: There is a Mac candidiate available to test on my github page
[11:20] Cuga.Rajal @rajal.org:9000: although I have the impression I'm ther only one put there lol
[11:20] Cuga.Rajal @rajal.org:9000: let me post the link...
[11:20] Vincent.Sylvester @hg.zetaworlds.com:8002: The bullet integration code in OpenSim has not been touched since I cleaned up indentation over a year ago. So that might need a review as well.
[11:20] Andrew Hellershanks: Cuga, true. In my case the program has dependencies which are also no longer supported and/or considered obsolete which also makes it "interesting".
[11:20] Cuga.Rajal @rajal.org:9000: https://github.com/cuga-rajal/opensim_apple_arm64
[11:21] Cuga.Rajal @rajal.org:9000: I also discovered that the Mac library for BulletSim is missing a critical bug fix that was included in the other playforms
[11:21] Cuga.Rajal @rajal.org:9000: so I need to get Ubit to update the Mac library for Bullet once again
[11:21] Cuga.Rajal @rajal.org:9000: and we need to discuss using universal binaries for Mac going forward instead of separate ones
[11:22] Cuga.Rajal @rajal.org:9000: Ubit doesnt want to use them but we were already using those from Gavin
[11:22] Cuga.Rajal @rajal.org:9000: so thats a discussion next time Ubit is here
[11:22] Andrew Hellershanks: Cuga, if you can't email Ubit directly you can file a bug report as a reminder.
[11:22] Cuga.Rajal @rajal.org:9000: ok
[11:23] Cuga.Rajal @rajal.org:9000: thats everything I had on Bullet and Mac platform for now
[11:23] Cuga.Rajal @rajal.org:9000: unkess there are Q's
[11:23] Andrew Hellershanks: Any questions from anyone?
[11:24] Andrew Hellershanks: I see Vincent is typing something.
[11:24] Cuga.Rajal @rajal.org:9000: the buiold instruction on my github page *might* work for Unix too
[11:24] Cuga.Rajal @rajal.org:9000: I ,ight give it a shot for Linux x86_64
[11:24] Andrew Hellershanks: ok, good to know.
[11:24] Cuga.Rajal @rajal.org:9000: oops I meant Linux (same thing lol)
[11:25] Andrew Hellershanks: :)
[11:25] Andrew Hellershanks: oh. I finally twigged to why the ground around us seems all white. It snowed since the last time we where here. :)
[11:26] Kayaker Magic: ?Here in-world? The sand looks sandy colored to me still.
[11:26] Cuga.Rajal @rajal.org:9000: yellow sand to me
[11:26] Vincent.Sylvester @hg.zetaworlds.com:8002: Commits this week were mostly some security stuff on inventory, broke a test in the process, but eventually did find a fix for that. Not entirely sure what was wrong. Security fixes are always a good thing though.
[11:26] Andrew Hellershanks: ok. It is white to me and most of the other textures of this building were all white/gray at first.
[11:27] Andrew Hellershanks: Vincent, yes they are.
[11:27] Vincent.Sylvester @hg.zetaworlds.com:8002: I spent some time with search again after finding whenever I looked for sandbox I got a list of various other regions with that in the name rather than the exact match
[11:27] Vincent.Sylvester @hg.zetaworlds.com:8002: It does a fuzzy search on the regionlist, so it does find it, I guess the viewer sorts it by alphabet instead of exact matches first
[11:28] Cuga.Rajal @rajal.org:9000: are bug fixes and patches being done now double on both branches dotnet6 and master?
[11:28] Vincent.Sylvester @hg.zetaworlds.com:8002: In some instances where they are critical, but there is some stuff on dotnet6 that has not been backported
[11:28] Vincent.Sylvester @hg.zetaworlds.com:8002: I did make some patches for doing so, but of course huge time sink to do
[11:28] Andrew Hellershanks: Vincent, that is by design. There was a time when what you typed in was always assumed to be the start of the name of a region. If you wanted to find a sandbox and searched using that box you would not have found a region called, for example, Public Sandbox
[11:29] Vincent.Sylvester @hg.zetaworlds.com:8002: Yep it is meant to do a like search, which will find exact and fuzzy matches, my main concern is that exact matches, which one would prefer to see first, are not automatically selected
[11:30] Ubit Umarov: hi
[11:30] Ubit Umarov: (sorry im late)
[11:30] Orbert.Tatham @hg.zetaworlds.com:80: Welcome, UBit
[11:30] Vincent.Sylvester @hg.zetaworlds.com:8002: so if I search for Sandbox knowing I want to go to that region the one it selects is one that alphabetically comes before that
[11:30] Vincent.Sylvester @hg.zetaworlds.com:8002: It's a viewer thing though
[11:30] Andrew Hellershanks: Hello, Ubit.
[11:31] Vincent.Sylvester @hg.zetaworlds.com:8002: Needs to sort exact match first and then fuzzy ones imo, that's what I would expect it to do at least, maybe that's just me though
[11:31] Andrew Hellershanks: Ubit, Cuga had some news related to work he and Misterblue are doing on BulletSim physics.
[11:31] Andrew Hellershanks: Cuga, You can ask Ubit about updating the libraries.
[11:31] Cuga.Rajal @rajal.org:9000: Shall I recap?
[11:31] Andrew Hellershanks: Sure.
[11:32] Cuga.Rajal @rajal.org:9000: MrBlue and I have been busy meeting and making a plan for Bullet
[11:32] Cuga.Rajal @rajal.org:9000: We're going to work on a new version that is updated to the latest 3.25 and contain bug fixes
[11:33] Cuga.Rajal @rajal.org:9000: we discussed which patches to use and then we are now hard at work trying to build them
[11:33] Cuga.Rajal @rajal.org:9000: oce built, we will set up some testing regions, and ask others to test
[11:33] Cuga.Rajal @rajal.org:9000: after testing, they will replace libs in trunk, so everybody has parity
[11:34] Cuga.Rajal @rajal.org:9000: I also discovered that the Mac versions of Bullet currently in trunk are missing some bug fixes that were already applied to Linux and Windows by MrBlue
[11:34] Cuga.Rajal @rajal.org:9000: Gavin didn;t include them
[11:35] Cuga.Rajal @rajal.org:9000: So I have a new version of Bullet 2.86 that includes that patch
[11:35] Cuga.Rajal @rajal.org:9000: Also, learned that the Mac Bullet libs from Gavin ere 2.86 since 2017
[11:35] Cuga.Rajal @rajal.org:9000: So the same version as recent arm64 libs, by accident
[11:36] Ubit Umarov: ok Cuga.Rajal
[11:36] Cuga.Rajal @rajal.org:9000: it;s documented in an old Mantis http://opensimulator.org/mantis/view.php?id=8232
[11:36] Cuga.Rajal @rajal.org:9000: But
[11:36] Cuga.Rajal @rajal.org:9000: Ubit you and I need to talk about using universal binaries
[11:36] Cuga.Rajal @rajal.org:9000: maybe not today but sometime soon
[11:37] Ubit Umarov: universal?
[11:37] Cuga.Rajal @rajal.org:9000: I am making universal binaries now with x86_64 and arm64 combined, and testing them on my production x86_64
[11:37] Ubit Umarov: guess you mean universal on mac world
[11:37] Ubit Umarov: ;)
[11:37] Cuga.Rajal @rajal.org:9000: There are no issues, and it's the common/preferred way on the Mac for other sw
[11:37] Ubit Umarov: and older macs?
[11:38] Cuga.Rajal @rajal.org:9000: correct, thats the nomenclature that Apple uses
[11:38] Cuga.Rajal @rajal.org:9000: they have been doing this since 2005 with "universal binaroes" for Power{C and Intel back then
[11:39] Cuga.Rajal @rajal.org:9000: So, Apple dropped support for 32 bit in their dev environment since 2019, in their hardware since 2011
[11:39] Cuga.Rajal @rajal.org:9000: I honestly dont think there are any 32 bit Mac users out there,
[11:39] Cuga.Rajal @rajal.org:9000: It's my opinion that we dont need to build for 32 bit Mac
[11:39] Cuga.Rajal @rajal.org:9000: Linux might be another story
[11:39] Vincent.Sylvester @hg.zetaworlds.com:8002: Most of the small arm dev boards are 64bit now too
[11:40] Cuga.Rajal @rajal.org:9000: even Raspberry Pi is 64 bit lol
[11:40] Orbert.Tatham @hg.zetaworlds.com:80: If you can get it
[11:40] Ubit Umarov: ok mayber when you have them done, make a mantis with this reasoning and the libs?
[11:40] Cuga.Rajal @rajal.org:9000: ok
[11:41] Cuga.Rajal @rajal.org:9000: The Mac version has a special problem
[11:41] Ubit Umarov: ill them change the .config things and commit them
[11:41] MichaelOmega ChristopherGeode: cuga - what problem?
[11:41] Cuga.Rajal @rajal.org:9000: bc Gavin didnt include the "floating prim" patch, so all Mac users are seeing a major bug
[11:41] Cuga.Rajal @rajal.org:9000: Afdter talking to MrBlue this week I know how to fix it
[11:41] Ubit Umarov: htat is a low level bullet fix?
[11:42] Cuga.Rajal @rajal.org:9000: he made a patch, but it didn;t get communicated out well
[11:42] Cuga.Rajal @rajal.org:9000: so I re-built the Mac Bullet with that patch a couple days ago and that;s ready to swap into trunk
[11:42] Cuga.Rajal @rajal.org:9000: thats the old version with just that patch added
[11:43] Cuga.Rajal @rajal.org:9000: yes it is a low lever Bullet fix, that really should get submitted to the Bullet team
[11:43] MichaelOmega ChristopherGeode: ok ty
[11:43] Cuga.Rajal @rajal.org:9000: so the new mac lib brings parity to the Msc version finally with othewr platforms
[11:43] Andrew Hellershanks: Hello, Michael.
[11:43] MichaelOmega ChristopherGeode: hello :|)
[11:43] Cuga.Rajal @rajal.org:9000: I'll open a Mantis for the new bug fix Bullet lib fro Mac
[11:43] MichaelOmega ChristopherGeode: cuga - is that for intel mac or m1?
[11:44] Cuga.Rajal @rajal.org:9000: Well, the bug exists for all Mac Bullet libs, and has been there since the current one was built in 2019
[11:45] Cuga.Rajal @rajal.org:9000: My bug fox version, I built iyt as a universal library, x86_64 and arm64, so it will work on both
[11:45] Cuga.Rajal @rajal.org:9000: But that would require a change in the configs that Ubit has been working on
[11:45] Cuga.Rajal @rajal.org:9000: and the question of universal libs pans out to Openjpeg and ubODE too
[11:46] Cuga.Rajal @rajal.org:9000: What we have now works fine, so no need to change it
[11:46] Cuga.Rajal @rajal.org:9000: but going forward, as we update Mac libs, I suggest replacing with arm64/x86_64 universal binary
[11:47] Cuga.Rajal @rajal.org:9000: Ubit has some concerns because the architectures are so different, and some concern how the loader in dotnet handles the universal lib loading
[11:48] Cuga.Rajal @rajal.org:9000: But this is an old trick in the mac world, also, I've done testing on the same file on both architectures, and it works
[11:49] Cuga.Rajal @rajal.org:9000: My feeling is if there is some issue with library loading using the universal binary, I could build separarte files, but I bet Ubit $5 that will never happen
[11:50] Andrew Hellershanks: :)
[11:50] Cuga.Rajal @rajal.org:9000: Ubit, whats your feeling about using the universal libs going forward?
[11:51] Cuga.Rajal @rajal.org:9000: for Bullet, I can;t build 32 bit anyway, so the choice is 2 separate libs for x86_64 and arm64, or one universal lib
[11:52] Cuga.Rajal @rajal.org:9000: I sort of need to know this no or soon-ish for submitting the Bullet fix
[11:52] Ubit Umarov: if they work.. ok
[11:52] Cuga.Rajal @rajal.org:9000: it's a little easier
[11:53] Ubit Umarov: only that :)
[11:53] Ubit Umarov: bc those are basicly like a tar file with all :)
[11:53] Cuga.Rajal @rajal.org:9000: simpler configs too
[11:53] Ubit Umarov: yeha the os does the select at load time, we hope
[11:54] Cuga.Rajal @rajal.org:9000: so far its working
[11:54] Cuga.Rajal @rajal.org:9000: I did put the new universal lib, that I built on my arm64 machine, onto my x86_64 production machine, and it loaded fine
[11:55] Cuga.Rajal @rajal.org:9000: OK so I'll open a mantis and get you a new Mac universal binary for Bullet with the bug fix
[11:56] Cuga.Rajal @rajal.org:9000: I also have a universal binary for Openjpeg if you want, although nothing is currently broken there, but maybe for consistency?
[11:56] Cuga.Rajal @rajal.org:9000: up to you
[11:57] Ubit Umarov: if not broken why break it?
[11:57] Ubit Umarov: ;)
[11:57] Cuga.Rajal @rajal.org:9000: consuistency
[11:58] Cuga.Rajal @rajal.org:9000 whispers: up to you :)
[11:58] Cuga.Rajal @rajal.org:9000: If ubODE needs an update, that will fall into the same issue/solution
[11:59] Cuga.Rajal @rajal.org:9000: but no need to change until an update is needed
[11:59] Cuga.Rajal @rajal.org:9000: unless you want to change them all now for consistency
[11:59] Cuga.Rajal @rajal.org:9000: this would be a Mac-only change
[11:59] Andrew Hellershanks: Cuga, I think you just answered the question about whether to update the other library.
[11:59] Andrew Hellershanks: "no need to change until an update is needed"
[12:00] Cuga.Rajal @rajal.org:9000: uless you care about consistency, inwhich case, you might want to change all now
[12:01] Cuga.Rajal @rajal.org:9000: the configs that point different archs and OS's won;t be consistent across all the libs for Mac
[12:01] Andrew Hellershanks: It is a bit of a toss up. Consistency is fine but as long as it isn't likely to break something in which case it can be better to leave well enough alone.
[12:01] Cuga.Rajal @rajal.org:9000: Bullet will be different
[12:02] Cuga.Rajal @rajal.org:9000: So I just wanted to bring that up and confirm thats what Ubit wants :)
[12:02] Andrew Hellershanks nod
[12:02] Vincent.Sylvester @hg.zetaworlds.com:8002: We really need to find someone with the same enthusiasm to update the database connectors and OpenSim.Data projects
[12:03] Cuga.Rajal @rajal.org:9000: maybe when I get caught up I can look, tho Im not a C# programmer...
[12:03] Andrew Hellershanks: We are just past the hour. Before we talk any more about physics, are there any other questions for today?
[12:03] MichaelOmega ChristopherGeode: not today
[12:04] Cuga.Rajal @rajal.org:9000: contractor here. gotta go ttyl
[12:04] Andrew Hellershanks: ok. That was partly a question for you, Michael, as you came in late.
[12:04] Vincent.Sylvester @hg.zetaworlds.com:8002: Anyone lately seen prims they deleted return after a region restart?
[12:04] Andrew Hellershanks: ok, Cuga. Thank you for the update.
[12:04] Orbert.Tatham @hg.zetaworlds.com:80: I have, a few times
[12:04] MichaelOmega ChristopherGeode: im talking t lisa abouthelping with sdcenegate
[12:04] Andrew Hellershanks: Vincent, I haven't seen that problem in quite a while.
[12:04] MichaelOmega ChristopherGeode: so once i get into it i may have more ??s
[12:05] Vincent.Sylvester @hg.zetaworlds.com:8002: I have only seen that maybe 5 times in the last couple years, but I did stumble across it again yesterday
[12:05] Andrew Hellershanks: Michael, Good luck with that. I'm sure they will be happy for the help. It is a big project.
[12:05] Vincent.Sylvester @hg.zetaworlds.com:8002: It usually is with things that were linked and unlinked, some weird thing internally most likely
[12:06] Vincent.Sylvester @hg.zetaworlds.com:8002: Bad lock or timing would not surprise me
[12:06] Vincent.Sylvester @hg.zetaworlds.com:8002: True voodoo heh
[12:06] Orbert.Tatham @hg.zetaworlds.com:80: Vincent, I have randomly seen prims appear that didn't belong where I saw them. I wasn't sure if it was ones I previously deleted, or what, so I haven't submitted anything
[12:06] Andrew Hellershanks: Vincent, I don't think I've seen that issue since I've been using 0.9.x but I haven't been building that much lately.
[12:07] Vincent.Sylvester @hg.zetaworlds.com:8002: With deletions it could be inventory related given that's what delete actually does, send to inventory
[12:07] Vincent.Sylvester @hg.zetaworlds.com:8002: But I did see it with a prim that I just changed and didn't delete
[12:07] Andrew Hellershanks: Vincent, what database are you using and what version of OS?
[12:07] Vincent.Sylvester @hg.zetaworlds.com:8002: mariadb with master dev on mono
[12:08] Orbert.Tatham @hg.zetaworlds.com:80: I have also had persistence issues, when a deleted prim didn't disappear, but with two different opeople looking ones saw it, one didn't
[12:08] Andrew Hellershanks: Clearing viewer cache makes no difference to the issue?
[12:09] Vincent.Sylvester @hg.zetaworlds.com:8002: I wrote a command a while back to check on delete queue and other operations related to this, but even running that to clear those before shutting the region down it didn't report anything. So I have no clue where it gets stuck at
[12:09] Orbert.Tatham @hg.zetaworlds.com:80: At least once it was still there when one of us logged back in
[12:09] Vincent.Sylvester @hg.zetaworlds.com:8002: Just strange behavior and completely random it seems
[12:09] Orbert.Tatham @hg.zetaworlds.com:80: Not very frequent
[12:09] Andrew Hellershanks nods
[12:10] Vincent.Sylvester @hg.zetaworlds.com:8002: When you build a lot and have a lot of people go through a region it does show more often, but that's not guarantee either
[12:11] Andrew Hellershanks: In the past I have had the issue in regions with little traffic.
[12:12] Orbert.Tatham @hg.zetaworlds.com:80: I never have traffic, but lots of build/change/move around stuff
[12:13] Vincent.Sylvester @hg.zetaworlds.com:8002: It restores them from the database since they are still in there so the delete there must have failed, but that train of code is pretty long before it gets to that. I have to find all those parts and add debug output I guess, see what it says
[12:13] Vincent.Sylvester @hg.zetaworlds.com:8002: Fun log spam that will be
[12:13] MichaelOmega ChristopherGeode: oh how fun that sounds
[12:13] Andrew Hellershanks: Vincent, another option would be to track SQL commands issued to the DB.
[12:14] Vincent.Sylvester @hg.zetaworlds.com:8002: I checked whether there were failed queries and I saw nothing reported
[12:15] Vincent.Sylvester @hg.zetaworlds.com:8002: This is like that timing lock all over again or the viewer objects cache, just voodoo
[12:16] Andrew Hellershanks: I think you have another "fun" project on your hand tracking down that issue.
[12:16] Orbert.Tatham @hg.zetaworlds.com:80: I sure haven't been able to find any ryme or reason when it has happened to me. Nothing in common except that most of the time there was a server restart
[12:17] Vincent.Sylvester @hg.zetaworlds.com:8002: Yeah no amount of running backup on the console will help with that it seems
[12:17] Orbert.Tatham @hg.zetaworlds.com:80: Nothing showed up there
[12:17] Andrew Hellershanks: Is it possible the DB delete request didn't get passed to the DB engine before the region restart?
[12:18] Andrew Hellershanks: Another thought is a possible cache related issue.
[12:18] Selby.Evans @grid.kitely.com:8002: bye all
[12:18] Andrew Hellershanks: ok Bye Selby.
[12:18] Orbert.Tatham @hg.zetaworlds.com:80: Peace, Selby
[12:18] Andrew Hellershanks: Merry Christmas to you
[12:18] MichaelOmega ChristopherGeode: bye!
[12:18] MichaelOmega ChristopherGeode whispers: merry chirstmas!|
[12:18] Vincent.Sylvester @hg.zetaworlds.com:8002: I had stopped building on that region a day before, by then cache should have cleared out
[12:19] Andrew Hellershanks: Vincent, one would hope so.
[12:19] Vincent.Sylvester @hg.zetaworlds.com:8002: I could try disabling it though and try to force it to occur to check
[12:19] Andrew Hellershanks: Any updates should have been flushed to the DB by then.
[12:19] Kayaker Magic: Gota run, Buy All!
[12:19] MichaelOmega ChristopherGeode: bye1
[12:19] Vincent.Sylvester @hg.zetaworlds.com:8002: It's either inventory delete fail or something with committing changes to db getting blocked or breaking
[12:19] Andrew Hellershanks: I didn't get a chance to say bye to Kayaker.
[12:20] Vincent.Sylvester @hg.zetaworlds.com:8002: Internal blacklisting or something too many minor changes trying to be committed
[12:20] Vincent.Sylvester @hg.zetaworlds.com:8002: There is so much code in that pipeline
[12:20] Vincent.Sylvester @hg.zetaworlds.com:8002: hay in a stack of needles
[12:20] Andrew Hellershanks: If anyone is able to reproduce the problem with any consistency pass your information on to Vincent.
[12:21] Vincent.Sylvester @hg.zetaworlds.com:8002: Yeah first step is reproduction conditions, then I can either try tracking or at least mantis it for others to help with
[12:21] Vincent.Sylvester @hg.zetaworlds.com:8002: Anyway that's all I had this week
[12:22] Andrew Hellershanks: Having experienced the problem in the past getting a set of steps to reproduce the issue with any consistency will be difficult.
[12:22] Andrew Hellershanks: ty, Vincent.
[12:22] MichaelOmega ChristopherGeode: i'd better get going to my next one too
[12:22] Andrew Hellershanks: As the time is getting late I will wrap up todays meeting.
[12:22] MichaelOmega ChristopherGeode: i was about to say see younext week - but will anyone be here lol
[12:23] Andrew Hellershanks: Merry Christmas, Happy holidays, Seasons greetings, etc. to all. I hope you enjoy sometime off to spend with friends and family. See you all again in a week from today.
[12:23] Andrew Hellershanks: Thank you all for coming.
