(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
[11:01 AM PT] Andrew Hellershanks: Hello, everyone.
[11:01 AM PT] Ubit Umarov: Unauthorized CAPS client
[11:01 AM PT] Ubit Umarov: oops
[11:02 AM PT] Ubit Umarov: dam this will not self heal still
[11:02 AM PT] Lyr Lobo: Hello *smiles*
[11:02 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Ubrokeit, clearly ;)
[11:02 AM PT] Ubit Umarov: guess gavin did crash now has a stuck client
[11:02 AM PT] Lyr Lobo: oh no
[11:02 AM PT] Lyr Lobo: was he sitting here? I can move
[11:02 AM PT] Ubit Umarov: yeha BIll would had said that :P
[11:03 AM PT] Ubit Umarov: nopes
[11:03 AM PT] Ubit Umarov: he did crash bc getting in
[11:03 AM PT] Lyr Lobo: very sorry to hear that
[11:03 AM PT] Vincent.Sylvester @hg.zetaworlds.com: That's code assuming something without checking, few places that still do that
[11:04 AM PT] Lyr Lobo: online again
[11:04 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Well even by design if you believe protocol specs
[11:04 AM PT] Andrew Hellershanks: Hello, Gavin.
[11:04 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Gavin!
[11:04 AM PT] Ubit Umarov: ahh he managed to get in
[11:04 AM PT] Gavin.Hird @grid.xmir.org:8002: Hi, yes
[11:05 AM PT] Gavin.Hird @grid.xmir.org:8002: the problem was I had not restarted robust after my public IP had changed
[11:05 AM PT] Gavin.Hird @grid.xmir.org:8002: regions survive that fine, but not robust
[11:05 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Gotta love the caches heh
[11:05 AM PT] Ubit Umarov: well and got a stuck child agent that did not want to auto clear
[11:06 AM PT] Gavin.Hird @grid.xmir.org:8002: here?
[11:06 AM PT] Ubit Umarov: well it did clear so you got in
[11:06 AM PT] Gavin.Hird @grid.xmir.org:8002: great
[11:07 AM PT] Gavin.Hird @grid.xmir.org:8002: I got here once but only the terrain loaded and then it did not accept me back because robust used the old public IP
[11:07 AM PT] Ubit Umarov: yeack
[11:08 AM PT] Vincent.Sylvester @hg.zetaworlds.com: dyndns folks get to play that game once a day, imagine their fun
[11:08 AM PT] Gavin.Hird @grid.xmir.org:8002: anyways,
[11:08 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Ubit been busy poking mysql
[11:08 AM PT] Gavin.Hird @grid.xmir.org:8002: hopefully mine will only change every blue moon
[11:08 AM PT] Gavin.Hird @grid.xmir.org:8002: right, but are we supposed to used dotnet 8 now?
[11:09 AM PT] Ubit Umarov: with fiber mine holds longer also
[11:09 AM PT] Andrew Hellershanks: Hello, Joe
[11:09 AM PT] Ubit Umarov: used to change everyday
[11:09 AM PT] Joe Magarac: Hello all.
[11:09 AM PT] Ubit Umarov: hi
[11:09 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Yeah just install it, prebuild and compile and it should work. Will automatically fall back on 6 for what it needs to... for now
[11:10 AM PT] Ubit Umarov: it is not exactly fallback to 6
[11:10 AM PT] Gavin.Hird @grid.xmir.org:8002: I was under the impression dotnet 8 would be a total fail?
[11:10 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Well it's certainly not an improvement from what I can see
[11:10 AM PT] Ubit Umarov: andrew you may want to open the meeting?
[11:10 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Log already on :)
[11:10 AM PT] Ubit Umarov: ooops
[11:11 AM PT] Ubit Umarov: well yes.. i did move the minimal dotnet runtime version to 8.x
[11:12 AM PT] Gavin.Hird @grid.xmir.org:8002: any implications?
[11:12 AM PT] Ubit Umarov: the main issue about it was its worse use of memory
[11:12 AM PT] Gavin.Hird @grid.xmir.org:8002: how big a change is that?
[11:12 AM PT] Ubit Umarov: GC is again slower to release memory to the operating system
[11:13 AM PT] Andrew Hellershanks: Ubit, Seems like it has already started as the discussions are on topic. :)
[11:13 AM PT] Ubit Umarov: but testing on a few regions like osgrid LBSA, shown that it staid within reason
[11:13 AM PT] Andrew Hellershanks: dotnet 8 is now needed? Hm... I will need to track down a newer version of dotnet that I can install on my machine.
[11:14 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Andrew, they did move things to proper packages so package managers should have it in their mirrors, if not you can add the official one
[11:14 AM PT] Andrew Hellershanks: nm. I do have that version.
[11:15 AM PT] Ubit Umarov: and considering the claimed overall improvements and that 6 is already 2 years old on a lts of 3.. .. i decided to move :)
[11:15 AM PT] Andrew Hellershanks: Vincent, true in theory. Linux Mint can be a bit out of date.
[11:15 AM PT] Andrew Hellershanks: Hello, Cuga.
[11:15 AM PT] Ubit Umarov: in fact already did commit one or 2 lines fo code that depend on c#12 :)
[11:15 AM PT] Cuga.Rajal @rajal.org:9000: Hi all, sorry late
[11:15 AM PT] Andrew Hellershanks: np, Cuga.
[11:15 AM PT] Gavin.Hird @grid.xmir.org:8002: will region crossings be snappier?
[11:16 AM PT] Ubit Umarov: did update a few support dlls, including oracle mysql and that did not went so well
[11:16 AM PT] Gavin.Hird @grid.xmir.org:8002: what happened?
[11:16 AM PT] Joe Magarac: Region crossings on the SL side are getting improvements from Monty Linden. That's in test now.
[11:16 AM PT] Ubit Umarov: it did not work on soem arm64 like PIs or orange tiny boards
[11:17 AM PT] Ubit Umarov: so i did revert to previus version
[11:17 AM PT] Gavin.Hird @grid.xmir.org:8002: k
[11:17 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Course that might also be issue on kernel level rather than arch level
[11:17 AM PT] Ubit Umarov: about that "fallback to 6" that is not how it works...
[11:17 AM PT] Vincent.Sylvester @hg.zetaworlds.com: I know, was simplifying
[11:18 AM PT] Ubit Umarov: the runtime version to use is decided by the start program..
[11:18 AM PT] Ubit Umarov: opensim now ( bc of dotnet) will refuse to start on runtimes < 8
[11:18 AM PT] Ubit Umarov: then that runtime will run code done for older versions, itself
[11:19 AM PT] Ubit Umarov: except if it does hit a "braking change etc"
[11:19 AM PT] Ubit Umarov: if is compatible it will run inside the updated VM
[11:19 AM PT] Ubit Umarov: well nothing that new.. all versions do that..
[11:20 AM PT] Ubit Umarov: only for some reason .net 3.5 had to be installed...
[11:20 AM PT] Ubit Umarov: i hope i did remove dependencies on it .. :)
[11:20 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Well some code hasn't been touched in a decade so no surprises there
[11:21 AM PT] Gavin.Hird @grid.xmir.org:8002: so you have to have 3.5 installed too?
[11:21 AM PT] Ubit Umarov: yeah but as i said, some of that old code will be compiled into native code using 8.0 jit compiler.. so may be better..
[11:21 AM PT] Ubit Umarov: as i said i hope i did remove all dependendics on 3.5 gavin.Hird
[11:22 AM PT] Gavin.Hird @grid.xmir.org:8002: is that a new requirement over dotnet 6 version?
[11:22 AM PT] Ubit Umarov: but ws a expection. on the rule that a runtime supports all older ones
[11:22 AM PT] Ubit Umarov: no...
[11:22 AM PT] Gavin.Hird @grid.xmir.org:8002: hmmm - will be interesting to see how that turns out on macOS
[11:23 AM PT] Ubit Umarov: well we needed to wait and see
[11:23 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Explosions are fun :)
[11:23 AM PT] Ubit Umarov: the list requirements of dotnnet 8 and 6 seem identical
[11:24 AM PT] Ubit Umarov: so.. moved on into brave new unknown bugs.. :)
[11:24 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Living on the edge~
[11:24 AM PT] Gavin.Hird @grid.xmir.org:8002: 8 needs macOS 12 or higher. 6 needs 10.5 or higher
[11:24 AM PT] Kayaker Magic: New bugs for Old!
[11:25 AM PT] Ubit Umarov: as you know.. this region is running on 8 for a few weeks also
[11:25 AM PT] Gavin.Hird @grid.xmir.org:8002: 10.15
[11:25 AM PT] Ubit Umarov: ofc on top of that i also made my own bugs.. err changes..
[11:25 AM PT] Orbert.Tatham @hg.zetaworlds.com: Kayaker, they never fix the old ones, just add new
[11:25 AM PT] Gavin.Hird @grid.xmir.org:8002: I'll have to test it on a 12.0 config
[11:25 AM PT] Ubit Umarov: the only god use of memory on dotnet is no use at all :P
[11:25 AM PT] Ubit Umarov: good not god :)
[11:26 AM PT] Cuga.Rajal @rajal.org:9000: Is the plan to make OS compatible with both dotnet 6 and 8?
[11:26 AM PT] Ubit Umarov: why would that even be a plan ??
[11:26 AM PT] Cuga.Rajal @rajal.org:9000: Is the desire to....
[11:26 AM PT] Ubit Umarov: luck it is compatible with 8 now .)
[11:26 AM PT] Ubit Umarov: nopes that is not how it works
[11:26 AM PT] Ubit Umarov: sadly
[11:27 AM PT] Ubit Umarov: versions like time only move one way...
[11:27 AM PT] Vincent.Sylvester @hg.zetaworlds.com: You could backport, but that time is better spent fixing bugs
[11:27 AM PT] Ubit Umarov: 8 wlll run most all that did run
[11:27 AM PT] Cuga.Rajal @rajal.org:9000: so some day in the future we will say we are officially switching to 8
[11:27 AM PT] Cuga.Rajal @rajal.org:9000: and just testing now
[11:27 AM PT] Ubit Umarov: that 6 did run, i mean
[11:28 AM PT] Gavin.Hird @grid.xmir.org:8002: almost?
[11:28 AM PT] Ubit Umarov: there are a few breaking changes ( also added on 7 )
[11:28 AM PT] Ubit Umarov: but hope covered
[11:29 AM PT] Cuga.Rajal @rajal.org:9000: Are there advantages of running on 8 instead of 6?
[11:29 AM PT] Ubit Umarov: the big breaking changes on 6 versus 4 ,, you remember.. xengine gone etc
[11:29 AM PT] Cuga.Rajal @rajal.org:9000: other than future-proofing
[11:29 AM PT] Ubit Umarov: well on the .dotnet evangelists blogs.. more than tons
[11:29 AM PT] Ubit Umarov: :)
[11:29 AM PT] Vincent.Sylvester @hg.zetaworlds.com: MS claims it to be faster... duh cause they removed stuff out of it
[11:29 AM PT] Gavin.Hird @grid.xmir.org:8002: have they added AI?
[11:29 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Car go faster if you remove the seats
[11:29 AM PT] Ubit Umarov: they made more use of SIMD for example
[11:30 AM PT] Orbert.Tatham @hg.zetaworlds.com: So more bugs and breaking changes than even the usual version change
[11:30 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Gavin, god I hope not
[11:30 AM PT] Ubit Umarov: even on strings compares :)
[11:30 AM PT] Gavin.Hird @grid.xmir.org:8002: :-)
[11:30 AM PT] Vincent.Sylvester @hg.zetaworlds.com: jit already makes enough assumptions that turn out to be worse machine code
[11:31 AM PT] Ubit Umarov: more and deeper massive use of Unsafe crap to bypass they own nosense rules like array bound checks
[11:31 AM PT] Ubit Umarov: looking to the frameworks sources it is a entire diferent ( and ugly) c#
[11:32 AM PT] Ubit Umarov: more massive use of ref... etc
[11:33 AM PT] Ubit Umarov: so some nice good moves.. just doing it the worng way... messing up the language as i said..
[11:33 AM PT] Ubit Umarov: but well that is details...
[11:33 AM PT] Vincent.Sylvester @hg.zetaworlds.com: ref is one thing, but the straight pointers are more concerning, point of memory safe design and they stuff that in there on levels you cannot control
[11:33 AM PT] Ubit Umarov: we can't control anything this days
[11:33 AM PT] Gavin.Hird @grid.xmir.org:8002: hopefully we won't get messages on the console nagging to back up to OneDrive
[11:33 AM PT] Vincent.Sylvester @hg.zetaworlds.com: XD
[11:33 AM PT] Ubit Umarov: everytime you start a region you will get diferent native code
[11:34 AM PT] Orbert.Tatham @hg.zetaworlds.com: Stability is not something that Micro$haft wants available
[11:34 AM PT] Cuga.Rajal @rajal.org:9000: you talking .net8?
[11:34 AM PT] Ubit Umarov: this region max physical memory use when it is opens logins ranges from 600MB to 800MB
[11:34 AM PT] Ubit Umarov: same code,,, similar conditions..
[11:35 AM PT] Ubit Umarov: just dotnet internal "random noise"
[11:35 AM PT] Ubit Umarov: and i disabled features like tired and PG compilation crap
[11:35 AM PT] Vincent.Sylvester @hg.zetaworlds.com: I recall the whole uuid compare being 4 times the size in machine code instructions than it needs to be
[11:36 AM PT] Vincent.Sylvester @hg.zetaworlds.com: jit sometimes just derps out
[11:36 AM PT] Ubit Umarov: wel they claim to had improved some things
[11:36 AM PT] Cuga.Rajal @rajal.org:9000: So if things don't work right in your dotnet app, just restart it and cross your fingers... Just like Windows
[11:36 AM PT] Ubit Umarov: some improvements will only work if we change the code..
[11:36 AM PT] Ubit Umarov: ofc
[11:36 AM PT] Orbert.Tatham @hg.zetaworlds.com: They improved their profit margins
[11:36 AM PT] Andrew Hellershanks: Vincent, hehe... I remember the discussions about that
[11:37 AM PT] Ubit Umarov: well that was the major code change.. :)
[11:37 AM PT] Ubit Umarov: lets see how it goes..
[11:37 AM PT] Ubit Umarov: minor ones.. hmm
[11:37 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Restarting fixes half the IT problems in the world as is, suppose dotnet just wanting to fit in maybe
[11:38 AM PT] Vincent.Sylvester @hg.zetaworlds.com: If something works first try you get suspicious heh
[11:38 AM PT] Ubit Umarov: removed soem lost code for llCould, long gone
[11:38 AM PT] Orbert.Tatham @hg.zetaworlds.com: Well, IT is mostly windoze, what did you expect
[11:38 AM PT] Ubit Umarov: did some fixes to our mysql code
[11:38 AM PT] Ubit Umarov: fun we had bad names on some columns and no one even runtime did compain
[11:39 AM PT] Ubit Umarov: like covenant last time change store . was broken
[11:40 AM PT] Gavin.Hird @grid.xmir.org:8002: odd
[11:40 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Typos.. in code.. no way
[11:40 AM PT] Ubit Umarov: made soem more usless changes on caps, changing its llsd encoding to simple fastfoward only
[11:40 AM PT] Ubit Umarov: that saces a lot of time and reduces creation of temp things on heap, so saving gc time also..
[11:41 AM PT] Ubit Umarov: well and that was it mostly :)
[11:42 AM PT] Ubit Umarov: we are still alive here on dotnet 8.. hmm are we?
[11:42 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Haven't yet tried libomv compile on 8
[11:42 AM PT] Ubit Umarov: wel i did change that also to 8
[11:43 AM PT] Ubit Umarov: and made a simpler core prebuild.xml that only includes the dlls needed for opensim
[11:43 AM PT] Ubit Umarov: that does compile fine on linux also
[11:43 AM PT] Ubit Umarov: so i did update it also
[11:43 AM PT] Ubit Umarov: as you know now also with a copy at github
[11:44 AM PT] Vincent.Sylvester @hg.zetaworlds.com: 158 warnings, but yeah does compile
[11:44 AM PT] Ubit Umarov: i did cleanup opensim-libs
[11:44 AM PT] Ubit Umarov: removing things we no longer use
[11:44 AM PT] Ubit Umarov: bitmap on openjpeg code alone creates a ton :P
[11:45 AM PT] Lyr Lobo grins at Vincent
[11:45 AM PT] Ubit Umarov: so you can see our repos at bitbucket
[11:46 AM PT] Ubit Umarov: https://bitbucket.org/opensimulator/workspace/overview/
[11:46 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Some of those warnings look likely relatively simple fixes
[11:46 AM PT] Ubit Umarov: github https://github.com/opensim
[11:46 AM PT] Ubit Umarov: and on our own git thingy
[11:47 AM PT] Cuga.Rajal @rajal.org:9000: I just tried building current trunk on macOS/dotnet6 and I get an error. it's telling me I need dotnet 8
[11:47 AM PT] Ubit Umarov: http://opensimulator.org/viewgit/
[11:47 AM PT] Ubit Umarov: that does not include our libomv fork
[11:47 AM PT] Gavin.Hird @grid.xmir.org:8002: as the man said, you need dotnet 8
[11:48 AM PT] Ubit Umarov: as i said i moved libomv master also to dotnet 8
[11:48 AM PT] Cuga.Rajal @rajal.org:9000: ok, so we need to update now, got it
[11:48 AM PT] Ubit Umarov: 6 is gone long live 8 ( 2 more years )
[11:48 AM PT] Ubit Umarov: :p
[11:49 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Huh compile on 8 are 25% slower compared to 6
[11:49 AM PT] Cuga.Rajal @rajal.org:9000: thanks, M$
[11:50 AM PT] Andrew Hellershanks: Seeing the reference to github reminds me of something.
[11:50 AM PT] Ubit Umarov: if you look the the blue tex at my right, mem use is similar to older use
[11:50 AM PT] Ubit Umarov: well with 100MB margin :p
[11:50 AM PT] Ubit Umarov: within...
[11:51 AM PT] Ubit Umarov: a lot better than we had with old mono anyways
[11:51 AM PT] Andrew Hellershanks: I'm moving all of my OpenSim related projects from github to gitlab. The copies on github will be marked as archived. This is due to the broken 2FA requirement I now have to deal with at github. It makes it difficult to login.
[11:51 AM PT] Ubit Umarov: on that i added a obscure setting
[11:51 AM PT] Ubit Umarov: well added use of..
[11:52 AM PT] Ubit Umarov: on a file called OpenSim.runtimeconfig.json
[11:52 AM PT] Gavin.Hird @grid.xmir.org:8002: MS making a mess of github too
[11:52 AM PT] Ubit Umarov: ( lost on every compile ..)
[11:52 AM PT] Ubit Umarov: by default GC things a application runs alone on a machine
[11:52 AM PT] Vincent.Sylvester @hg.zetaworlds.com: So gitlab.com not your own gitlab instance I take it?
[11:53 AM PT] Ubit Umarov: it will only panic and start releasing memory harder with the app is using 95% of the entire machine physical ram
[11:53 AM PT] Andrew Hellershanks: Vincent, yes. I have imported all my projects that were at github.com in to my account at gitlab.com
[11:54 AM PT] Vincent.Sylvester @hg.zetaworlds.com: That's good, better than sourceforge
[11:54 AM PT] Vincent.Sylvester @hg.zetaworlds.com: I'll make my pull requests there then :)
[11:54 AM PT] Gavin.Hird @grid.xmir.org:8002: how does that work when you run multiple instances of a simulator on the same machine?
[11:54 AM PT] Ubit Umarov: i added a line "System.GC.HighMemoryPercent": 50,
[11:55 AM PT] Gavin.Hird @grid.xmir.org:8002: added where?
[11:55 AM PT] Ubit Umarov: by default telling it to do it at half of memory use :P
[11:55 AM PT] Ubit Umarov: on this region i changed it to
[11:55 AM PT] Gavin.Hird @grid.xmir.org:8002: in code or config
[11:55 AM PT] Ubit Umarov: "System.GC.HighMemoryPercent": 1, so to do it at 1% memory usage
[11:56 AM PT] Ubit Umarov: that is 638MB
[11:56 AM PT] Ubit Umarov: ( gavin file OpenSim.runtimeconfig.json, as i said remade on evey compile)
[11:57 AM PT] Ubit Umarov: the console stats comand shows it
[11:57 AM PT] Ubit Umarov: GCTotalCommited: 121MB GCTotalAvaiable 63849MB GCHMthreshold 638MB [11:57 AM PT] Ubit Umarov: machines running many small regions may need to change this also
[11:58 AM PT] Gavin.Hird @grid.xmir.org:8002: ok
[11:58 AM PT] Ubit Umarov: the made improvements on gc server mode.. but worstation seems the best for use still
[11:59 AM PT] Ubit Umarov: the made it try to guess a app memory needs..
[12:00 PM PT] Ubit Umarov: warp3d map at start totaly kills that
[12:00 PM PT] Ubit Umarov: bc they did not consider that a app may have a high mem use at start then use a lot less.. for example
[12:01 PM PT] Ubit Umarov: well and about you? what news do you have ?
[12:01 PM PT] Ubit Umarov: err abotu opensim first please :)
[12:02 PM PT] Vincent.Sylvester @hg.zetaworlds.com: Still pulling my hair out over sqlite heh
[12:02 PM PT] Ubit Umarov: vincent.Sylvester you moved.. mem usage increased
[12:02 PM PT] Vincent.Sylvester @hg.zetaworlds.com: Migrations coming along though
[12:02 PM PT] Andrew Hellershanks: Vincent, what is the problem with SQLite?
[12:02 PM PT] Vincent.Sylvester @hg.zetaworlds.com: It's not mysql xD
[12:03 PM PT] Vincent.Sylvester @hg.zetaworlds.com: Have to make different routines for it since it doesn't have table comments
[12:03 PM PT] Vincent.Sylvester @hg.zetaworlds.com: Such stuff, just differences that need individual handling
[12:03 PM PT] Vincent.Sylvester @hg.zetaworlds.com: Not complicated, just time consuming
[12:03 PM PT] Ubit Umarov: well our use of sqlite is .. well a thing :)
[12:04 PM PT] Ubit Umarov: hate that crap of tables thingies
[12:04 PM PT] Ubit Umarov: who made original mysql code also used that ms thing.. then got clever... not on sqlite
[12:04 PM PT] Andrew Hellershanks: IIRC, it doesn't support ALTER TABLES so that can complicate DB migrations.
[12:05 PM PT] Vincent.Sylvester @hg.zetaworlds.com: In older versions, I think it does these days
[12:05 PM PT] Vincent.Sylvester @hg.zetaworlds.com: There is a few things it does different
[12:05 PM PT] Vincent.Sylvester @hg.zetaworlds.com: floats and doubles
[12:05 PM PT] Vincent.Sylvester @hg.zetaworlds.com: indexes
[12:05 PM PT] Ubit Umarov: you did notive the 'lite' on its name right? ;)
[12:05 PM PT] Ubit Umarov: notice..
[12:05 PM PT] Andrew Hellershanks: :)
[12:06 PM PT] Andrew Hellershanks: We are a few minutes past the hour. Does anyone have a question or comment to ask before the meeting for today comes to a close?
[12:07 PM PT] Ubit Umarov: ( or the region crashes.. )
[12:07 PM PT] Luisillo Contepomi: I have big problems with the new compilation from today
[12:07 PM PT] Luisillo Contepomi: when import oar or iar....
[12:07 PM PT] Luisillo Contepomi: mariadb
[12:07 PM PT] Luisillo Contepomi: all new
[12:07 PM PT] Ubit Umarov: oh?
[12:07 PM PT] Luisillo Contepomi: database new region new
[12:07 PM PT] Andrew Hellershanks: What sort of problem?
[12:07 PM PT] Luisillo Contepomi: 19:56:32 - [LOGHTTP]: Slow handling of 1082 POST /xinventory from 192.168.1.9:38252 took 4272ms
19:56:36 - [LOGHTTP]: Slow handling of 1083 POST /xinventory from 192.168.1.9:38252 took 4424ms
19:56:40 - [LOGHTTP]: Slow handling of 1084 POST /xinventory from 192.168.1.9:38252 took 4268ms
19:57:16 - [LOGHTTP]: Slow handling of 1177 POST /xinventory from 192.168.1.9:51054 took 10964ms
19:57:42 - [LOGHTTP]: Slow handling of 1236 POST /xinventory from 192.168.1.9:59390 took 6516ms
19:57:47 - [LOGHTTP]: Slow handling of 1237 POST /xinventory from 192.168.1.9:59390 took 5308ms
19:58:02 - [LOGHTTP]: Slow handling of 1258 POST /xinventory from 192.168.1.9:57296 took 4256ms
19:58:08 - [LOGHTTP]: Slow handling of 1264 POST /xinventory from 192.168.1.9:51644 took 6676ms
19:58:28 - [LOGHTTP]: Slow handling of 1326 POST /xinventory from 192.168.1.9:34836 took 10001ms
19:58:31 - [LOGHTTP]: Slow handling of 1327 POST /xinventory from 192.168.1.9:34836 took 3748ms
19:58:44 - [LOGHTTP]: Slow handling of 1369 POST /xinventory
[12:07 PM PT] Luisillo Contepomi: from 192.168.1.9:34536 took 3136ms
19:58:53 - [LOGHTTP]: Slow handling of 1370 POST /xinventory from 192.168.1.9:34536 took 9448ms
19:58:58 - [LOGHTTP]: Slow handling of 1371 POST /xinventory from 192.168.1.9:34536 took 4449ms
19:59:03 - [LOGHTTP]: Slow handling of 1375 POST /xinventory from 192.168.1.9:34536 took 4852ms
19:59:20 - [LOGHTTP]: Slow handling of 1413 POST /xinventory from 192.168.1.9:51496 took 3616ms
19:59:30 - [LOGHTTP]: Slow handling of 1414 POST /xinventory from 192.168.1.9:51496 took 9957ms
19:59:33 - [LOGHTTP]: Slow handling of 1415 POST /xinventory from 192.168.1.9:51496 took 3312ms
19:59:38 - [LOGHTTP]: Slow handling of 1416 POST /xinventory from 192.168.1.9:51496 took 4688ms
19:59:53 - [LOGHTTP]: Slow handling of 1457 POST /xinventory from 192.168.1.9:54578 took 6948ms
19:59:59 - [LOGHTTP]: Slow handling of 1459 POST /xinventory from 192.168.1.9:54578 took 5341ms
20:00:09 - [LOGHTTP]: Slow handling of 1460 POST /xinventory from 192.168.1.9:54578 took 9312ms
20:00:27 - [LOGHTTP]: Slow
[12:07 PM PT] Luisillo Contepomi: handling of 1501 POST /xinventory from 192.168.1.9:49346 took 4408ms
20:00:49 - [LOGHTTP]: Slow handling of 1539 POST /xinventory from 192.168.1.9:35214 took 9280ms
20:01:25 - [LOGHTTP]: Slow handling of 1604 POST /xinventory from 192.168.1.9:59582 took 4120ms
20:01:41 - [LOGHTTP]: Slow handling of 1605 POST /xinventory from 192.168.1.9:59582 took 16729ms
20:01:52 - [LOGHTTP]: Slow handling of 1606 POST /xinventory from 192.168.1.9:59582 took 10520ms
[12:08 PM PT] Luisillo Contepomi: one hour and half after....
[12:08 PM PT] Luisillo Contepomi: 20:53:50 - [LOGHTTP]: Slow handling of 4794 POST /xinventory from 192.168.1.9:52258 took 4700ms
20:54:21 - [LOGHTTP]: Slow handling of 4797 POST /xinventory from 192.168.1.9:52258 took 30417ms
REGION
/////////////////////////////////////////////////////////////////
region when have 20:54:21 - [LOGHTTP]: Slow handling of 4797 POST /xinventory from 192.168.1.9:52258 took 30417ms
20:53:50 - [XInventory]: POST http://192.168.1.9:8003/xinventory took 4700ms 537/80bytes
20:54:20 - [XInventory]: Error receiving response from http://192.168.1.9:8003/xinventory: The request was canceled due to the configured HttpClient.Timeout of 30 seconds elapsing.
20:54:20 - [INVENTORY ARCHIVER]: Archive load for Luisillo Contepomi failed - The request was canceled due to the configured HttpClient.Timeout of 30 seconds elapsing.
Region (Continente) #
[12:08 PM PT] Luisillo Contepomi: canceled without end....due configured httpclient...
[12:08 PM PT] Andrew Hellershanks: Luisillo, You should have posted a link to a pastebin with the messages instead of pasting the messages here.
[12:08 PM PT] Luisillo Contepomi: now i have half iar in the server
[12:09 PM PT] Luisillo Contepomi: oh sorry
[12:09 PM PT] Ubit Umarov: wlel that is about you sql server performance
[12:09 PM PT] Luisillo Contepomi: yes from today the performance is this
[12:09 PM PT] Ubit Umarov: using the previus mysql lib there should be no reason
[12:09 PM PT] Luisillo Contepomi: yesterday not have this problem
[12:09 PM PT] Luisillo Contepomi: for this i come urgently for say you
[12:09 PM PT] Luisillo Contepomi: :-)
[12:09 PM PT] Ubit Umarov: :)
[12:10 PM PT] Vincent.Sylvester @hg.zetaworlds.com: Could be a number of things, worst case your disk has started to die
[12:10 PM PT] Luisillo Contepomi: are monitored and have not load
[12:10 PM PT] Vincent.Sylvester @hg.zetaworlds.com: In mysql you can up the memory limit, packet size limit and the connection limit to reduce bottlenecks
[12:10 PM PT] Luisillo Contepomi: or problems
[12:10 PM PT] Ubit Umarov: those are timing issues.. mysql server is very slow on reply
[12:11 PM PT] Luisillo Contepomi: all configurations are the same one week ago in mysql
[12:11 PM PT] Ubit Umarov: ( the 30s limit should be more than enough )
[12:11 PM PT] Luisillo Contepomi: but ill test more
[12:11 PM PT] Ubit Umarov: well the change of the mysql.data.dll could have impact on that
[12:12 PM PT] Luisillo Contepomi: smart service say disk are ok
[12:12 PM PT] Cuga.Rajal @rajal.org:9000: I just tried creating a large IAR with the latest trunk/dotnet8/macOS/mySQL8 and it worked OK
[12:12 PM PT] Ubit Umarov: we where using 8.0.31 for oracle,, then changed to 8.3.0 then back to 8.0.31 :)
[12:12 PM PT] Luisillo Contepomi: i test with compilation from 3 days ago and have not thiss problem
[12:12 PM PT] Ubit Umarov: bc 8.3.0 fails on arm64
[12:12 PM PT] Luisillo Contepomi: and yesterday have not this problems
[12:13 PM PT] Ubit Umarov: you did runprebuild on each change right?
[12:13 PM PT] Luisillo Contepomi: yes
[12:13 PM PT] Luisillo Contepomi: i download from git when you do some change
[12:13 PM PT] Luisillo Contepomi: and then compile new
[12:13 PM PT] Luisillo Contepomi: all new
[12:13 PM PT] Luisillo Contepomi: and then test with one oar and one iar
[12:14 PM PT] Luisillo Contepomi: for give load to the region
[12:14 PM PT] Ubit Umarov: well last change was to revert to mysql 8.0.31
[12:14 PM PT] Cuga.Rajal @rajal.org:9000: no issue here with 8.0.27
[12:14 PM PT] Ubit Umarov: maybe you did overload you box? :)
[12:14 PM PT] Luisillo Contepomi: yes i am using mariadb
[12:14 PM PT] Luisillo Contepomi: mariadb Ver 15.1 Distrib 10.6.15-MariaDB, for Linux (x86_64) using EditLine wrapper
[12:15 PM PT] Luisillo Contepomi: from some month ago
[12:15 PM PT] Luisillo Contepomi: no problems
[12:15 PM PT] Luisillo Contepomi: ill repeat all process again
[12:15 PM PT] Ubit Umarov: well it is not a easy to debug issue..
[12:15 PM PT] Luisillo Contepomi: and open tomorrow a mantis if i obtain again this result
[12:16 PM PT] Luisillo Contepomi: thank you all for your work
[12:16 PM PT] Luisillo Contepomi: specially to ubit
[12:16 PM PT] Ubit Umarov: it is timing.. so can be a ton of things even ssd getting old as vicent tols
[12:16 PM PT] Luisillo Contepomi: ;-)
[12:16 PM PT] Lyr Lobo grins
[12:16 PM PT] Ubit Umarov: :)
[12:16 PM PT] Luisillo Contepomi: ill try all new again
[12:16 PM PT] Luisillo Contepomi: have good afternoon
[12:17 PM PT] Luisillo Contepomi: bye
[12:17 PM PT] Gavin.Hird @grid.xmir.org:8002: remember PM on your ssds
[12:17 PM PT] Ubit Umarov: you can try the other mysql.data.dll
[12:17 PM PT] Luisillo Contepomi: thank you again
[12:18 PM PT] Cuga.Rajal @rajal.org:9000: anyone here good with math and have a few mins? I'm having some challenges with a scripting a rotation
[12:18 PM PT] Cuga.Rajal @rajal.org:9000: over at my region
[12:18 PM PT] Gavin.Hird @grid.xmir.org:8002: I need to get going
[12:18 PM PT] Gavin.Hird @grid.xmir.org:8002: have a great week all
[12:18 PM PT] Andrew Hellershanks: ok, Gavin. Thanks dropping by
[12:18 PM PT] Ubit Umarov: you 2
[12:18 PM PT] Lyr Lobo: bye gavin
[12:19 PM PT] Cuga.Rajal @rajal.org:9000: tc Gavin
[12:19 PM PT] Ubit Umarov: see you all next week if not before :)
[12:19 PM PT] Cuga.Rajal @rajal.org:9000: I should head out too
[12:19 PM PT] Andrew Hellershanks: Cuga, I can have a look. If it involves quaternions it can get "interesting.
[12:19 PM PT] Cuga.Rajal @rajal.org:9000: it's exactly that :)
[12:19 PM PT] Kayaker Magic: Buy all!
[12:19 PM PT] Andrew Hellershanks: Sounds like time to call todays meeting to a close.
[12:20 PM PT] Andrew Hellershanks: Thank you all for coming. See you again next week.
[12:20 PM PT] Lyr Lobo: bye
[12:20 PM PT] Lyr Lobo: take care
