Chat log from the meeting on 2024-10-22

From OpenSimulator

Revision as of 11:21, 22 October 2024 by Tampa (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search
[11:01 AM PT]  Andrew Hellershanks: Hello, everyone.
[11:01 AM PT]  Gavin.Hird @grid.xmir.org:8002: Hi Andrew
[11:02 AM PT]  Ubit Umarov: hi
[11:03 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Let's see what's new. So dotnet is not offered via MS repos on any newer things as they expect distros to provide them and then also maintain them, which is quite funny given the available version on ubuntu is already out of date by weeks. So either compile it or use one from other distro version, which probably will work fine
[11:03 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: I setup some virtual machines to test the latest release or beta versions of mysql, mariadb, postgres and well sqlite not that that has really changed
[11:03 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Expecting explosions, but hoping for none
[11:04 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: mysql and mariadb going far apart now to the point they now have their own clients too
[11:04 AM PT]  Gavin.Hird @grid.xmir.org:8002: I'm still lingering on mysql 8.0.x
[11:04 AM PT]  Kayaker Magic: I've switched back and forth between MariaDB and MySQL with no issues.
[11:04 AM PT]  Ubit Umarov: https://dotnet.microsoft.com/en-us/download   is still there
[11:05 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: There are no actual packages in the noble, just for jammy
[11:05 AM PT]  Ubit Umarov: well lik emany, htey may drop linux
[11:05 AM PT]  Kayaker Magic: I'm sticking with MySQL 8 because it has logical operations on blobs, and I neededed that.
[11:05 AM PT]  Lyr Lobo smiles in greeting
[11:05 AM PT]  Gavin.Hird @grid.xmir.org:8002: yes, that is where a make a hackyish install of dotnet on Raspbian from
[11:05 AM PT]  Cuga.Rajal @rajal.org:9000: https://dotnet.microsoft.com/en-us/download/dotnet/8.0 has security updates
[11:06 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Basically if a distro supplies it MS won't anymore so you now have to bug your distro to keep things up to date
[11:06 AM PT]  Ubit Umarov: https://dotnet.microsoft.com/en-us/download/dotnet/8.0
[11:06 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Not sure how it is with debian, but likely similar
[11:07 AM PT]  Ubit Umarov: where had you seen those news?
[11:07 AM PT]  Gavin.Hird @grid.xmir.org:8002: IDK - debian only had a SDL update a couple weeks ago
[11:07 AM PT]  Gavin.Hird @grid.xmir.org:8002: SDK
[11:08 AM PT]  Ubit Umarov: well if ms stops doing releases, distros will do do them offen also..  C# is not very popular on linux devs
[11:09 AM PT]  Andrew Hellershanks: Hello, Lyr and Web.
[11:09 AM PT]  Ubit Umarov: well ofc they do not support linux on several things already
[11:09 AM PT]  Web Rain: Hi there :) hi andrew :)
[11:09 AM PT]  Ubit Umarov: like gui things
[11:10 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Canonical has it on their repos now, but it's not the latest version. If you want that you either have to compile yourself or grab it from another distro version. On Ubuntu noble you can grab jammy version and it'll likely work. MS said on their release info they wouldn't ship a release for a distro version if that had it in its own repos, which noble now does
[11:11 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: All it means is that each time a security thing is patched now have to bark at Canonical to update it in their repo
[11:11 AM PT]  Kayaker Magic: When I let apt-get install dotnet, it gave me dotnet 8.0.200 which was messed up, i had to remove it and follow MicroSoft's instructions and that ended up as dotnet 8.0.108 which is working fine.
[11:12 AM PT]  Ubit Umarov: nothing new on that.. distros mono was broken a lot of time
[11:12 AM PT]  Gavin.Hird @grid.xmir.org:8002: yep
[11:13 AM PT]  Gavin.Hird @grid.xmir.org:8002: uncertain how many actually care about dotnet on linux
[11:13 AM PT]  Kayaker Magic: I care!
[11:13 AM PT]  Ubit Umarov: guess not many
[11:13 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: At least on ubuntu the version of libgdiplus is mostly up to date and not the memory leak version
[11:13 AM PT]  Cuga.Rajal @rajal.org:9000: we do, but likely not many others
[11:13 AM PT]  Gavin.Hird @grid.xmir.org:8002: exactly Cuga
[11:14 AM PT]  Kayaker Magic: Several grids run on Linux, Wolf Terratories, Discovery Grid (shutting down soon)
[11:14 AM PT]  Ubit Umarov: guess most grids do run on linux
[11:14 AM PT]  Gavin.Hird @grid.xmir.org:8002: I can't say I have seen any updates to libgdiplus for ages
[11:14 AM PT]  Ubit Umarov: a few on win :)
[11:14 AM PT]  Cuga.Rajal @rajal.org:9000: Macos works!
[11:14 AM PT]  Cuga.Rajal @rajal.org:9000: lol
[11:14 AM PT]  Kayaker Magic: Utopia Skye is on WinCows servers
[11:15 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Anyways, also had some commits this past week :)
[11:15 AM PT]  Ubit Umarov: yeah but minro bugs fixes
[11:15 AM PT]  Ubit Umarov: minor also
[11:17 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Fun little typos I somehow keep stumbling across
[11:18 AM PT]  Ubit Umarov: :)
[11:19 AM PT]  Andrew Hellershanks: libgdiplus is a bit of an issue. Programs built using it some times crash with messages related to gdi.
[11:20 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Eventually has to be replaced anyways seeing as System.Drawing is gone
[11:21 AM PT]  Ubit Umarov: well excuse why sys.d is gone is actually libgdiplus
[11:21 AM PT]  Ubit Umarov: and not gone in windows...
[11:22 AM PT]  Ubit Umarov: ofc there is uses the real thing..
[11:23 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Supposedly there is a library that can be used that even offers better performance, but integrating it still means slightly rewriting things and it's one more dependency handled by an outside force
[11:23 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: ImageSharp or whatever it's called again
[11:24 AM PT]  Ubit Umarov: yeah a big "slightly"
[11:24 AM PT]  Andrew Hellershanks: That's available for multiple operating systems?
[11:24 AM PT]  Ubit Umarov: one is just a c# wrapper over a google thing
[11:25 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Says its multi platform so I'd assume so
[11:26 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Has a split license attached to it, so no idea how that'll work, but beyond that there isn't really one I could find that wouldn't require completely rebuilding the entire texture rendering things from the ground up
[11:27 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: ImageSharp is supposedly not that different to only require small changes to drop in. Still means changing stuff up
[11:27 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: So far no drop-in replacement for it
[11:29 AM PT]  Gavin.Hird @grid.xmir.org:8002: SL is adding a new cap RegionSchedule where you can scedule a time to restart a region
[11:29 AM PT]  Ubit Umarov: just set your phone alarm clock... duhh
[11:29 AM PT]  Gavin.Hird @grid.xmir.org:8002: :-))
[11:29 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: I also saw they added something to bring up teleport destinations and such through a script
[11:30 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: That seems like something that might require some new viewer handling too unless it's another pseudo instant message blob thing
[11:30 AM PT]  Gavin.Hird @grid.xmir.org:8002: There is also a new floater, of course, to shcedule the region retsart
[11:30 AM PT]  Gavin.Hird @grid.xmir.org:8002: restart
[11:31 AM PT]  Web Rain: i think we have something in opensim for this if i remember... scheduled regions restart using script commands
[11:31 AM PT]  Cuga.Rajal @rajal.org:9000: console commands or lsl?
[11:32 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: You can only shut a simulator down, after that you need something external to start it back up
[11:32 AM PT]  Gavin.Hird @grid.xmir.org:8002: true
[11:32 AM PT]  Web Rain: no no in in the config... is possible to set and run the execution of a script command file
[11:32 AM PT]  Gavin.Hird @grid.xmir.org:8002: so for now the floater will be marked onlyInSL it sounds
[11:33 AM PT]  Web Rain: in the script commands file you can change region > restart
[11:33 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: I tried at one point to look into restart markers that can be used for that, but they are a pain to implement and if you start the simulator with config options it gets even more complicated
[11:34 AM PT]  Ubit Umarov: code should only do shutdown
[11:34 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: The "restart" function was removed a while back, because it never worked, so that does nothing now. Only thing it can do is call a shutdown
[11:34 AM PT]  Ubit Umarov: soft restart in dotnet is..  well  not good
[11:35 AM PT]  Cuga.Rajal @rajal.org:9000: shell scripts at the Operating system level are how I do it
[11:35 AM PT]  Ubit Umarov: and it is not just dotnet/mono.. opensim code is also not good
[11:35 AM PT]  Ubit Umarov: some just use a while loop on .sh
[11:35 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: I found it easier to implement a "redirect" for the estate debug restart to instead just call a url for a local script to perform the restart
[11:36 AM PT]  Ubit Umarov: if region dies  it auto restarts
[11:36 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Or cronjob a routine to start if pid is missing :)
[11:36 AM PT]  Ubit Umarov: of you need to kill that script if you realy want to kill the region
[11:36 AM PT]  Ubit Umarov: ofc..
[11:36 AM PT]  Cuga.Rajal @rajal.org:9000: sometimes threads hang when shutting down
[11:36 AM PT]  Web Rain: this need a bit of love :)
[11:36 AM PT]  Web Rain: http://opensimulator.org/wiki/Startup_script_linux
[11:36 AM PT]  Cuga.Rajal @rajal.org:9000: so have to kill the tty
[11:38 AM PT]  Ubit Umarov: in linux better run opensim under screen or tmux (??)
[11:38 AM PT]  Gavin.Hird @grid.xmir.org:8002: I run under screen both on macOS and Linux
[11:38 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: On Windows one can setup scheduled tasks and services for it
[11:38 AM PT]  Kayaker Magic loves screen!
[11:38 AM PT]  Web Rain: is still using mono/screen
[11:39 AM PT]  Ubit Umarov: **path not found
[11:39 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Running dotnet directly in terminal will mess with your terminal config as I keep finding out anew every time I test things
[11:39 AM PT]  Kayaker Magic: IS THAT WHY MY TERMINAL GETS MESSED UP!
[11:39 AM PT]  Gavin.Hird @grid.xmir.org:8002: runs fin in terminal on macOS, but I guess MS pays more attention to it
[11:40 AM PT]  Ubit Umarov: nahh you do mess it up kayaker ...
[11:40 AM PT]  Kayaker Magic: LOL
[11:40 AM PT]  Gavin.Hird @grid.xmir.org:8002: not fin but fine
[11:40 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: It keeps messing with the numpad for some reason
[11:40 AM PT]  Ubit Umarov: ;)
[11:40 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Got no answer when I complained on their discord about that
[11:40 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Both tmux and screen will log crashes as well, so useful to have
[11:43 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: I'm still looking for those mimetypes if anyone feels generous enough to look for them. Might be possible to fetch them from SL CDN response, but me no have SL anymore
[11:47 AM PT]  Andrew Hellershanks: Vincent, which mimetypes?
[11:47 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: For the asset types, settings and materials specifically
[11:48 AM PT]  Web Rain: i left SL the same month they banned L$ exchange outside their system... an era ago...
[11:48 AM PT]  Cuga.Rajal @rajal.org:9000: and now they sold that to a 3rd party
[11:51 AM PT]  Andrew Hellershanks: I haven't been in SL for some time now. Been even long since I wanted to cash out any L$
[11:53 AM PT]  Andrew Hellershanks: Just under 10 minutes left in the hour. Any other questions/comments from anyone re: OS?
[11:53 AM PT]  Cuga.Rajal @rajal.org:9000: Just curious, is anyone somewhat familiar with Bullet wrapper and internals bedides MisterBlue?
[11:55 AM PT]  Cuga.Rajal @rajal.org:9000: guess not, ok thanks
[11:55 AM PT]  Andrew Hellershanks: :)
[11:55 AM PT]  Ubit Umarov: :)
[11:55 AM PT]  Web Rain: talking about economy... druskus the owner of Zone Nations have published in the jOpenSim forum a prototype of a modified version  of JOpenSimPayPal with BTCPay support :) happy to see that my modest amateur in action work serve to someone/thing :)
[11:56 AM PT]  Web Rain: his first work is for jOpenSim 5 for Joomla 5 
[11:56 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: I have dug around in bullet years ago to find more typos and stuff, but never really to the point of understanding what it does. Physics are a lot of trig calc that goes over my small brain, but if you have time I'll happily throw my brain in there and help you work things out
[11:57 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Still trying to fix the bullet bugs I guess?
[11:57 AM PT]  Cuga.Rajal @rajal.org:9000: Thanks Vincent
[11:57 AM PT]  Cuga.Rajal @rajal.org:9000: thewres one that seems should be able to fix
[11:57 AM PT]  Andrew Hellershanks: Cuga, Is there some particular issue you are having with Bullet?
[11:57 AM PT]  Cuga.Rajal @rajal.org:9000: No
[11:57 AM PT]  Cuga.Rajal @rajal.org:9000: I'll message you offline
[11:57 AM PT]  Cuga.Rajal @rajal.org:9000: possibly fix an ld bug
[11:57 AM PT]  Cuga.Rajal @rajal.org:9000: old*
[11:57 AM PT]  Andrew Hellershanks: ok. I haven't looked at the internals of any of the physics engines either.
[11:58 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: You can poke me on IRC or discord if you have any questions, can't make promises though :)
[11:58 AM PT]  Cuga.Rajal @rajal.org:9000: it's that llDetectedLinkNumber(integer x) works when the linkset os not physical
[11:58 AM PT]  Cuga.Rajal @rajal.org:9000: changes when goes physical
[11:58 AM PT]  Cuga.Rajal @rajal.org:9000: shouldnt
[11:58 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Isn't there a mantis on that?
[11:58 AM PT]  Cuga.Rajal @rajal.org:9000: yes
[11:59 AM PT]  Cuga.Rajal @rajal.org:9000: but not looked at
[11:59 AM PT]  Cuga.Rajal @rajal.org:9000: I fogot to include that is does work when not physical
[11:59 AM PT]  Cuga.Rajal @rajal.org:9000: so maybe that is a clue
[11:59 AM PT]  Ubit Umarov: guess bullet can't make that work, bc loses information about the prim
[12:00 PM PT]  Cuga.Rajal @rajal.org:9000: or its a code bug
[12:00 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: Probably when physical all calc goes through root prim
[12:00 PM PT]  Ubit Umarov: for colisions
[12:00 PM PT]  Cuga.Rajal @rajal.org:9000: but collisions on non-physical are ok
[12:00 PM PT]  Ubit Umarov: well bullet things :p
[12:00 PM PT]  Cuga.Rajal @rajal.org:9000: :p
[12:01 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: Context switching. As you know on bullet when things go physical they sort of become something else
[12:01 PM PT]  Cuga.Rajal @rajal.org:9000: well thats what needfs to be looked at
[12:01 PM PT]  Ubit Umarov: if i remember bullet remakes some meshes for physics
[12:01 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: Or even just another copy paste issue, wouldn't be the first heh
[12:01 PM PT]  Ubit Umarov: mergim then as convex things
[12:01 PM PT]  Ubit Umarov: with o memory of the prim
[12:02 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: Prim inside bounding box might be skipped on detect then
[12:02 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: Plausible
[12:02 PM PT]  Ubit Umarov: but wel im guessing :)
[12:02 PM PT]  Cuga.Rajal @rajal.org:9000: yeah I think has to do with using root only for movement
[12:02 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: I think I asked this before, but what is missing from ubode that you elected to use bullet?
[12:03 PM PT]  Cuga.Rajal @rajal.org:9000: good question
[12:03 PM PT]  Cuga.Rajal @rajal.org:9000: Initially most of my scripting was vehicles, and Bullet seems more life-like for that
[12:03 PM PT]  Cuga.Rajal @rajal.org:9000: partuclarly land vehicles
[12:03 PM PT]  Cuga.Rajal @rajal.org:9000: ubODE you cant go over 180mph
[12:03 PM PT]  Cuga.Rajal @rajal.org:9000: Bullet 400 
[12:04 PM PT]  Andrew Hellershanks: Are you trying to simulate jet aircraft?
[12:04 PM PT]  Ubit Umarov: nahh not realy
[12:04 PM PT]  Gavin.Hird @grid.xmir.org:8002: scripting a bullet train?
[12:04 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: heh
[12:04 PM PT]  Ubit Umarov: protocol has issues with speed
[12:04 PM PT]  Cuga.Rajal @rajal.org:9000: also some physics objects didnt work well under ubODE = the ball and socket atuculating thing that I use a lot
[12:04 PM PT]  Ubit Umarov: there are "hidden" limits
[12:04 PM PT]  Cuga.Rajal @rajal.org:9000: such as hanging hool
[12:04 PM PT]  Cuga.Rajal @rajal.org:9000: hoop
[12:05 PM PT]  Cuga.Rajal @rajal.org:9000: invisible ball and socket at top gives illusion of hanging object swinging free
[12:05 PM PT]  Ubit Umarov: in same cases you can see that as a bug at sl
[12:05 PM PT]  Cuga.Rajal @rajal.org:9000: works quiter well with Bullet
[12:05 PM PT]  Cuga.Rajal @rajal.org:9000: and closer behavior to SL
[12:05 PM PT]  Cuga.Rajal @rajal.org:9000: However in recent months I am learning how to re-fractor scripts to work with ubODE
[12:06 PM PT]  Cuga.Rajal @rajal.org:9000: many forces/impilses ahve to be refactored
[12:06 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: You should catalog those differences and whether you consider either a bug or just a difference. Then perhaps that can be looked at and ubode improved a bit. I'm sure we can bully Ubit into making changes ;)
[12:06 PM PT]  Cuga.Rajal @rajal.org:9000: in some cases, I have to change wihich prim in a linkset should be root
[12:06 PM PT]  Ubit Umarov: hmm it is a nice blue sky here today
[12:06 PM PT]  Cuga.Rajal @rajal.org:9000: as a rult of switch from Bullet to ubODE
[12:06 PM PT]  Cuga.Rajal @rajal.org:9000: but I am working through it
[12:06 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: Nah but for real be nice to have a list of the caveats somewhere in case someone runs into them
[12:07 PM PT]  Cuga.Rajal @rajal.org:9000: Some of my no-priority projects only run on ubODE because of theis issue
[12:07 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: Nothings perfect. Knowing what might need a bit of love is the first step
[12:07 PM PT]  Cuga.Rajal @rajal.org:9000: so I am in the process of popssib ly permanently switchinh to ubODE
[12:07 PM PT]  Cuga.Rajal @rajal.org:9000: because of this issue
[12:08 PM PT]  Ubit Umarov: oops rl time
[12:08 PM PT]  Ubit Umarov: jezzz
[12:08 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: The vehicle thing strikes me though. I have not played around with that a lot on land though
[12:08 PM PT]  Gavin.Hird @grid.xmir.org:8002: yeah, I should go too
[12:08 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: Most vehicles I have are so old they behave weird due to that mostly
[12:08 PM PT]  Ubit Umarov: oh EU will change time next sunday
[12:08 PM PT]  Cuga.Rajal @rajal.org:9000: so really bith physics engines have pros/cons
[12:08 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: Thinking if I wrote something from scratch might be able to get them working nicely, but time is not my friend lately
[12:08 PM PT]  Andrew Hellershanks: Hm... That means we are about due for a time change in North America.
[12:08 PM PT]  Ubit Umarov: ofc i will forget abotu it :p
[12:09 PM PT]  Ubit Umarov: usa is Nov 3
[12:09 PM PT]  Gavin.Hird @grid.xmir.org:8002: wuld be more fun if they did it on election day
[12:09 PM PT]  Ubit Umarov: :)
[12:09 PM PT]  Kayaker Magic: Daylight wastings time, always a waste of time.
[12:09 PM PT]  Andrew Hellershanks: Oh. EU changing before NA. Interesting. I tend to think of it as usually the other way around.
[12:09 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: If I recall correctly next week EU folks get here an hour earlier than normal
[12:10 PM PT]  Andrew Hellershanks: The US is still talking about stopping the clock changing business.
[12:10 PM PT]  Ubit Umarov: we chane on last sundays of march and oct
[12:10 PM PT]  Cuga.Rajal @rajal.org:9000: but eaving it on Daylight time forever!
[12:10 PM PT]  Cuga.Rajal @rajal.org:9000: aaarrrrggghhhhh
[12:10 PM PT]  Web Rain: there is a security problems related to the binary thing in ubODE... OpenSim/Region/PhysicsModules/ubOdeMeshing/Meshmerizer.cs
[12:10 PM PT]  Ubit Umarov: hmm our rule is to follow pacific time no?
[12:11 PM PT]  Andrew Hellershanks: I prefer to have it light later in the day. Never remember which is which of the two clock times. I'd prefer the clocks ahead an hour thing.
[12:11 PM PT]  Web Rain: reported by the github AI
[12:11 PM PT]  Gavin.Hird @grid.xmir.org:8002: ak, guys. Have a great week
[12:11 PM PT]  Andrew Hellershanks: yes, The meetings are based on gridtime (ie. Pacific time).
[12:11 PM PT]  Ubit Umarov: ok
[12:11 PM PT]  Andrew Hellershanks: ok, Gavin. See you again next week.
[12:11 PM PT]  Cuga.Rajal @rajal.org:9000: I need to head out too.. see you next week
[12:12 PM PT]  Andrew Hellershanks: Web, was that filed in a bug report?
[12:12 PM PT]  Andrew Hellershanks: ok, Cuga. See you again next week.
[12:13 PM PT]  Web Rain: not in mantis
[12:13 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: Depending on criticality file as private mantis to not disclose anything potentially dangerous
[12:13 PM PT]  Andrew Hellershanks: true
[12:14 PM PT]  Web Rain: cp from github lvl critical
[12:14 PM PT]  Web Rain: Unvalidated local pointer arithmetic
[12:14 PM PT]  Andrew Hellershanks: That would be something for Ubit/
[12:14 PM PT]  Web Rain: It is dangerous to use the result of a virtual method call in pointer arithmetic without validation if external users can provide their own implementation of the virtual method. For example, if the analyzed project is distributed as a library or framework, then the end-user could provide a new implementation that returns any value.
[12:14 PM PT]  Joe Magarac: Gotta go.
[12:14 PM PT]  Ubit Umarov: Unvalidated local pointer arithmetic  yeah...  AI things :p
[12:15 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: So it'll just null ref or null pointer at worse crash the simulator
[12:15 PM PT]  Andrew Hellershanks: ok, Joe. Thanks for dropping by.
[12:15 PM PT]  Web Rain: knowing that handle the mesh uploads...
[12:15 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: Bad mesh can already crash and softlock a region
[12:16 PM PT]  Ubit Umarov: yeah all those c   programs with pointers  crahs all the time
[12:16 PM PT]  Ubit Umarov: bahh
[12:16 PM PT]  Andrew Hellershanks: :)
[12:16 PM PT]  Ubit Umarov: but sure file it and set private :)
[12:16 PM PT]  Web Rain: may be a vector for malicious code injection using mesh files
[12:16 PM PT]  Lyr Lobo: have a great day!  deliveries are here *waves*
[12:17 PM PT]  Andrew Hellershanks: ok, Lyr.
[12:17 PM PT]  Ubit Umarov: other than thje meshes themselfs? :p
[12:17 PM PT]  Andrew Hellershanks: hehe
[12:17 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: Doesn't sound like something critical, more in the annoying category if you somehow manage to run into it
[12:17 PM PT]  Web Rain: i don't know i'm not good in that part of opensim :(
[12:17 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: Critical is more when user data or system security is compromised
[12:18 PM PT]  Andrew Hellershanks: Should still be looked at if it can cause a crash.
[12:18 PM PT]  Ubit Umarov: wel file the message you got with references to code..  and set private mantis issue
[12:19 PM PT]  Ubit Umarov: ppl did love to set code vitual also for no good reason
[12:19 PM PT]  Ubit Umarov: i may pasted it from the other meshmerizer :)
[12:19 PM PT]  Ubit Umarov: ok rl calls
[12:20 PM PT]  Ubit Umarov: have fun :)
[12:20 PM PT]  Web Rain: is better to enable CodeQL  scanner in your github settings... will point you to the right files
[12:20 PM PT]  Andrew Hellershanks: I should be going as well.
[12:20 PM PT]  Ubit Umarov: thoxe things count to the free cpu time we habe
[12:20 PM PT]  Ubit Umarov: have..
[12:20 PM PT]  Vincent.Sylvester @hg.zetaworlds.com: Only pointers in that code is for the bounding box stuff
[12:20 PM PT]  Andrew Hellershanks: Remember to check the grid time vs local time next week. Thank you all for coming. See you again next week.
[12:20 PM PT]  Web Rain: oki thanks all :)
Personal tools
General
About This Wiki