Log for #openttd on 12th September 2020:
Times are UTC Toggle Colours
01:40:32  *** Progman_ has quit IRC
02:23:05  *** tokai|noir has joined #openttd
02:23:05  *** ChanServ sets mode: +v tokai|noir
02:29:53  *** tokai has quit IRC
02:48:49  *** Flygon has joined #openttd
03:02:59  *** snail_UES_ has quit IRC
03:07:09  <DorpsGek_III> [OpenTTD/OpenTTD] HaloFire1985 left a comment on commit: Update: Changelog for 1.9.2 and prepare for release
03:10:42  <DorpsGek_III> [OpenTTD/OpenTTD] HaloFire1985 opened pull request #8306: Keep Sake
03:11:06  <DorpsGek_III> [OpenTTD/OpenTTD] HaloFire1985 closed pull request #8306: Keep Sake
03:11:14  <DorpsGek_III> [OpenTTD/OpenTTD] HaloFire1985 reopened pull request #8306: Keep Sake
03:48:33  *** supermop_Home_ has quit IRC
05:43:32  <DorpsGek_III> [OpenTTD/OpenTTD] TrueBrain closed pull request #8306: Keep Sake
05:49:17  <TrueBrain> dwfreed: yeah, I remember the days that you could scan, and find, for SOCKS proxies :P And good point :)
05:50:03  <TrueBrain> Luckily, we use AWS, with sane Security Groups, that basically only allow traffic that should be allowed .. no public interface has "all open" rule, only "open port N on protocol X" :D
05:50:18  <TrueBrain> and the SOCKS port is not going to be part of the "allow" :P
05:50:24  <TrueBrain> (only on the private interface)
06:24:17  *** nielsm has joined #openttd
06:42:25  *** Wolf01 has joined #openttd
07:01:48  *** sla_ro|master has joined #openttd
07:06:42  *** HerzogDeXtEr has joined #openttd
07:23:50  *** Eddi|zuHause has quit IRC
07:24:58  *** Eddi|zuHause has joined #openttd
07:28:31  *** guru3_ has joined #openttd
07:29:28  *** labs[m] has quit IRC
07:29:28  *** jact[m] has quit IRC
07:30:28  *** nolep[m] has quit IRC
07:31:29  <LordAro> TrueBrain: i thought we moved off digitalocean because of the terrible ipv6 support? :p
07:31:49  <TrueBrain> DigitalOcean had -no- IPv6 support
07:32:04  <TrueBrain> but yeah, I did not expect this shit tbh ..
07:32:28  *** twom[m] has quit IRC
07:32:43  *** dave[m] has quit IRC
07:32:46  <LordAro> well as long as the proxy will not occasionally fall over and need manually restarting, it's still an improvement
07:32:57  <TrueBrain> exactly :)
07:32:58  *** johanna[m] has quit IRC
07:33:13  *** freu[m] has quit IRC
07:33:33  *** guru3 has quit IRC
07:35:18  *** linda[m] has quit IRC
07:36:43  *** dekeract[m] has quit IRC
07:37:33  *** ciet[m] has quit IRC
07:37:38  *** gretel[m] has quit IRC
07:38:43  *** udo[m] has quit IRC
07:39:13  *** philip[m] has quit IRC
07:41:56  *** labs[m] has joined #openttd
07:42:18  *** albert[m] has quit IRC
07:42:23  *** Heiki[m] has quit IRC
07:43:48  *** julie[m] has quit IRC
07:49:22  *** jact[m] has joined #openttd
07:53:45  *** nolep[m] has joined #openttd
07:54:11  *** Progman has joined #openttd
08:11:22  *** twom[m] has joined #openttd
08:13:41  *** freu[m] has joined #openttd
08:15:13  *** Gustavo6046 has quit IRC
08:17:38  <DorpsGek_III> [OpenTTD/master-server] TrueBrain opened pull request #7: Add: support SOCKS proxy for new outgoing connections
08:17:48  <TrueBrain> okay, that solves 1 of the two issues ... now lets see about the second ..
08:19:46  <DorpsGek_III> [OpenTTD/master-server] TrueBrain updated pull request #7: Add: support SOCKS proxy for new outgoing connections
08:22:31  *** ciet[m] has joined #openttd
08:23:48  *** dave[m] has joined #openttd
08:25:55  *** linda[m] has joined #openttd
08:26:11  *** udo[m] has joined #openttd
08:27:09  *** gretel[m] has joined #openttd
08:34:08  <TrueBrain> ugh, the other bug is ... an odd thing in PynamoDB ..
08:34:12  *** dekeract[m] has joined #openttd
08:34:25  <TrueBrain> an empty string is changed to None, which means the field is not written to the database
08:34:37  <TrueBrain> on reading the record, PynamoDB goes: THIS FIELD IS NOT THERE, PANIC
08:34:49  <TrueBrain> ... so why did you change an empty field to None in the first place, if you don't allow reading it back?
08:34:51  <TrueBrain> ugh ..
08:38:06  *** albert[m] has joined #openttd
08:38:11  *** johanna[m] has joined #openttd
08:38:22  *** Heiki[m] has joined #openttd
08:43:12  *** julie[m] has joined #openttd
08:58:15  <DorpsGek_III> [OpenTTD/master-server] TrueBrain opened pull request #8: Several smaller bugs and improvements found while running in production
08:58:53  *** philip[m] has joined #openttd
08:59:13  <TrueBrain> and I forgot to run black AGAIN
08:59:45  <DorpsGek_III> [OpenTTD/master-server] TrueBrain updated pull request #8: Several smaller bugs and improvements found while running in production
08:59:59  <TrueBrain> you would think I would learn, but alas, I do not :)
09:01:23  <LordAro> set up your editor to run it on save :p
09:01:37  <TrueBrain> pre-commit hook would be better tbh :P
09:01:51  <TrueBrain> but okay, with these two PRs, and some changes in the infra, we should be able to go live again :)
09:02:10  <TrueBrain> at least, all my pre-checks indicate that :P
09:05:03  <dwfreed> TrueBrain: check out :)
09:05:23  <TrueBrain> cool :)
09:11:21  <DorpsGek_III> [OpenTTD/master-server] LordAro approved pull request #8: Several smaller bugs and improvements found while running in production
09:13:22  <DorpsGek_III> [OpenTTD/master-server] TrueBrain merged pull request #8: Several smaller bugs and improvements found while running in production
09:13:36  <DorpsGek_III> [OpenTTD/master-server] TrueBrain closed issue #5: Allow settings a table name prefix
09:13:54  <DorpsGek_III> [OpenTTD/master-server] LordAro approved pull request #7: Add: support SOCKS proxy for new outgoing connections
09:14:10  <DorpsGek_III> [OpenTTD/master-server] TrueBrain dismissed a review for pull request #7: Add: support SOCKS proxy for new outgoing connections
09:14:10  <DorpsGek_III> [OpenTTD/master-server] TrueBrain updated pull request #7: Add: support SOCKS proxy for new outgoing connections
09:14:13  <TrueBrain> sorry, just forced pushed an rebase :P
09:14:43  <LordAro> bleh
09:14:55  <DorpsGek_III> [OpenTTD/master-server] LordAro approved pull request #7: Add: support SOCKS proxy for new outgoing connections
09:15:00  <TrueBrain> tnx LordAro :)
09:15:11  <TrueBrain> now I need to figure out how to run a socks proxy on AWS in a proper way :)
09:17:16  <TrueBrain> read: learn how to make a service in systemd :P
09:19:45  <dwfreed> systemd service units are dead simple for basic uses (one of few positives); just make sure you have Restart=on-failure
09:19:49  <dwfreed> and you'll be fine
09:24:33  <TrueBrain> and now I indeed know how to do this ... :P I appreciate it, but I am still none-the-wiser :)
09:25:25  <dwfreed> copy an existing one :D
09:25:54  <TrueBrain> but WHICH one, that is the trick :)
09:28:00  <dwfreed> if you have /lib/systemd/system/rsync.service, that's a good one
09:28:04  <dwfreed> it's very simple
09:28:22  <dwfreed> Restart=on-failure goes under the [Service] section
09:28:32  <TrueBrain> tnx :)
09:28:45  <dwfreed> ssh.service is a bit more complex, but demonstrates some other common options
09:29:04  <TrueBrain> it just needs to run a Python app :)
09:29:09  <TrueBrain> and reboot on any failure
09:30:01  <dwfreed> manpage
09:31:06  <TrueBrain> <- guess that should do
09:31:31  <dwfreed> yes
09:32:02  <dwfreed> drop that in /etc/systemd/system, then systemctl enable pproxy.service
09:32:39  <TrueBrain> tyvm :)
09:32:59  <dwfreed> note that will run as root by default
09:33:16  <dwfreed> add User=foo to run as foo user (also under [Service])
09:34:40  <TrueBrain> not a bad idea
09:35:30  <TrueBrain> does it need a systemctl reload, or what is it called?
09:35:36  <TrueBrain> or will enable just work after dropping in the file?
09:36:42  <dwfreed> 'systemctl daemon-reload' may be needed before enable or start will work
09:36:51  <dwfreed> note that enable does not start the unit
09:37:02  <TrueBrain> I assumed as much :)
09:37:10  <TrueBrain> cool .. I will test this in a sec .. now a private DNS ..
09:39:38  <DorpsGek_III> [OpenTTD/master-server] TrueBrain merged pull request #7: Add: support SOCKS proxy for new outgoing connections
09:46:36  <TrueBrain> Failed to start pproxy.service: Unit is not loaded properly: Invalid argument.
09:46:38  <TrueBrain> boooeeeee
09:47:18  <TrueBrain> pproxy.service lacks both ExecStart= and ExecStop= setting. Refusing.
09:47:27  <TrueBrain> I think something else went wrong :D
09:50:21  <TrueBrain> okay, service works, but application does not .. hmm
09:52:16  <TrueBrain> right, logs don't really show why it failed .. hmmmmmm
09:54:04  <TrueBrain> there we go!
09:54:14  <TrueBrain> tnx dwfreed :) That was relative painless :D
10:00:57  <TrueBrain> right .. now the scary part: redeploy some major parts of the AWS infra :D
10:10:59  <TrueBrain> well, first lunch I guess, because .. brrr :P
10:21:01  *** frosch123 has joined #openttd
10:31:02  <TrueBrain> frosch123: <- good enough, you think?
10:41:11  <frosch123> do you have redirects installed?
10:41:18  <frosch123> the urls are linked from the official CVE entries
10:41:57  <frosch123> oi, you replaced the svn revs with git hashes :)
10:42:29  <TrueBrain> yeah ... I made it nice and clean :)
10:42:34  <TrueBrain> and no, redirects I still have to do
10:42:38  <TrueBrain> but that will be easy enough :)
10:43:11  <frosch123> where does the "login" at the bottom come from?
10:43:34  <frosch123> oh, main website still has that :p
10:45:20  <TrueBrain> lolz .. well, if you can PR that, it will be gone soon ;)
10:45:35  <TrueBrain> the "patch" files were the most interesting
10:45:38  <TrueBrain> they are generated
10:45:44  <TrueBrain> the on-disk patch is just the diff
10:45:51  <TrueBrain> the rest is added based on the information of the CVE
10:49:37  <frosch123> i like how the code changes over time in the diffs
10:49:51  <frosch123> from crappy old C declaring all vars at the top
10:50:05  *** gelignite has joined #openttd
10:50:11  <frosch123> to fixing signedness stuff
10:51:07  <TrueBrain> :D
10:52:15  <TrueBrain> okay, SOCKS proxy is running .. now some minor tweaks to the infra to get it "just right" ..
10:58:01  <DorpsGek_III> [OpenTTD/website] somini commented on pull request #157: Include final URL
10:59:30  <DorpsGek_III> [OpenTTD/website] TrueBrain commented on pull request #157: Include final URL
11:03:01  <DorpsGek_III> [OpenTTD/website] somini commented on pull request #157: Include final URL
11:04:05  <TrueBrain> okay .. who has an IPv6 connection and can run a modified server? :D
11:08:06  <FLHerne> IPv6 on the server?
11:08:19  <FLHerne> I have that, but not at home :-(
11:11:44  <TrueBrain> I think I enabled it on my own server ... we will see :D
11:15:15  <_dp_> I probably can, at least it does ping6
11:16:25  <TrueBrain> hmm .. SOCKS proxy doesn't work in full chain ... this will be fun to find the issue :P
11:19:07  <TrueBrain> okay, that was the firewall
11:19:09  <TrueBrain> fine
11:19:20  <TrueBrain> [2020-09-12 13:18:59] dbg: [net] [udp] advertising on master server successful (IPv6)
11:22:15  <TrueBrain> right, time to finetune the firewall first ..
11:32:52  *** emeraldsnorlax[m] has joined #openttd
11:35:15  <TrueBrain> right, there we go :) Okay ... this should work now ... famous last words? :P
11:36:33  <DorpsGek_III> [OpenTTD/website] TrueBrain opened pull request #166: Fix 38fd927: don't end "url" with a slash; it results in double slash
11:40:51  <TrueBrain> [OpenTTD/master-server] TrueBrain created new tag: 1.0.2
11:46:03  <DorpsGek_III> [OpenTTD/website] LordAro approved pull request #166: Fix 38fd927: don't end "url" with a slash; it results in double slash
11:46:17  <DorpsGek_III> [OpenTTD/website] TrueBrain merged pull request #166: Fix 38fd927: don't end "url" with a slash; it results in double slash
12:03:23  <TrueBrain> okay, time to flip the switch again, and see if the new master-server holds this time :P
12:06:28  <frosch123> looks like you slipped and revealed your true bot nature
12:07:41  <TrueBrain> yeah, I need to fix .dorpsgek.yml for that repo :P
12:07:42  <TrueBrain> too lazy :)
12:07:59  <TrueBrain> hmm .. there are servers advertising, but they don't have their firewall configured correctly :P
12:08:58  <_dp_> yeah, I probably have some of those
12:09:07  <_dp_> doesn't matter, they're just for testing
12:10:08  <TrueBrain> they are spamming my logs! :P
12:10:24  <TrueBrain> but most servers are getting through :)
12:10:41  <TrueBrain> 40 IPv4 and 8 IPv6 so far, will takes ~15 minutes for the rest to arrive
12:11:11  <_dp_> what api url should I use btw, ?
12:11:18  <TrueBrain> so those that fail querying, are most likely their local firewall issues .. I hope at least :) The IPs that fail so far also never made it to the old MSU, so I guess
12:11:21  <TrueBrain> yes
12:11:55  <TrueBrain> that page caches for 5 minutes I believe, but a server only updates every 15
12:14:14  <TrueBrain> owh, the amount of used memory is also a lot lower than I expected ... <50MB
12:14:23  *** gelignite has quit IRC
12:15:46  <TrueBrain> lol, seems could have used a redirect from /en/ to / :P
12:15:55  <TrueBrain> we didn't do that for BaNaNaS either, so meh :P
12:19:11  <TrueBrain> sentry remains empty .. seems this version is a lot more stable :D
12:21:13  <_dp_> oh, I restarted my pc today so spammy servers aren't mine :p
12:22:01  <TrueBrain> the IPs are also all over the place, so I didn't really expect them to be :D
12:22:12  <TrueBrain> there is 1 IPv6 with many ports, of which none of them reply :p
12:23:24  <TrueBrain> the amount of traffic on the NLB did increase drastically .. seems master-server uses more traffic than content-server (excluding downloads, ofc)
12:24:33  <TrueBrain> "There are 105 clients, 270 IPv4 servers and 58 IPv6 servers in this list."
12:24:34  <TrueBrain> w00p
12:24:56  <TrueBrain> hmm .. sorting of servers is not as it should :P
12:36:19  <DorpsGek_III> [OpenTTD/master-server-web] TrueBrain opened pull request #6: Small improvements notices after going live
12:36:59  <DorpsGek_III> [OpenTTD/master-server-web] TrueBrain updated pull request #6: Small improvements notices after going live
12:37:05  <TrueBrain> every ... fucking ... single ... time .... I need a pre-commit hook :P
12:40:20  <DorpsGek_III> [OpenTTD/master-server] TrueBrain opened pull request #9: Fix: don't log every time a server is not responding
12:40:27  <TrueBrain> there ... as this is too spammy in the logs :P
12:40:45  <Eddi|zuHause> but once you go through the trouble of making it into a pre-commit-hook, you'll also remember it so it won't ever trigger :p
12:43:03  <TrueBrain> ~30 servers are announcing with a wrongly configured firewall ...
12:43:19  <DorpsGek_III> [OpenTTD/master-server-web] frosch123 approved pull request #6: Small improvements notices after going live
12:43:46  <DorpsGek_III> [OpenTTD/master-server-web] TrueBrain merged pull request #6: Small improvements notices after going live
12:44:11  <DorpsGek_III> [OpenTTD/master-server-web] TrueBrain created new tag: 1.0.1
12:44:15  <DorpsGek_III> [OpenTTD/master-server] frosch123 approved pull request #9: Fix: don't log every time a server is not responding
12:44:25  <DorpsGek_III> [OpenTTD/master-server] TrueBrain merged pull request #9: Fix: don't log every time a server is not responding
12:44:58  <TrueBrain> [OpenTTD/master-server] TrueBrain created new tag: 1.0.3
12:45:00  <TrueBrain> :P
12:45:02  <TrueBrain> tnx frosch123 !
12:45:44  <TrueBrain> owh, and frosch123  .. I forgot we old accounts system was still running on the Django .. so I cannot shut down the old VM yet :P
12:46:42  <TrueBrain> we = the, what-ever :P
12:47:46  <frosch123> so the oldest part stays the longest?
12:47:54  <frosch123> i guess you have to peel the onion from the outside
12:48:00  <TrueBrain> sadly, yup ...
12:48:05  <TrueBrain> well, I can shutdown another VM now, so that is a bonus
12:48:09  <TrueBrain> and I can restrict even more traffic
12:48:28  <TrueBrain> I think tomorrow I am going through the old nginx web forwarder, to see which subdomains it replies to .. and move them all away
12:48:33  <TrueBrain> that would be a good day too :D
12:49:11  <TrueBrain> cool, all the servers on master-server list I can query ingame too, except for 2 .. guess that server just went offline :D
12:52:09  <frosch123> <- sometimes software suprises me
12:52:38  <TrueBrain> this new master-server should be fully UTF-8 compatible
12:52:44  <TrueBrain> mostly because I did absolutely nothing with it :D
12:52:50  <TrueBrain> and in Python3 .. that is mostly enough :P
12:52:57  <TrueBrain> I liked my server-name .. I put trains in it :D
12:53:05  <TrueBrain> I am shocked nobody does that ..
12:53:39  *** Eddi|zuHause has quit IRC
12:53:56  <frosch123> Eddi|zuHause should file a bugreport that python should support proper uppercase ẞ
12:54:09  <frosch123> oh, the force is strong in him
12:54:15  <TrueBrain> bah, a sentry entry ..
12:54:16  <frosch123> he must have anticipated that
12:54:22  <TrueBrain> and one very odd one ..
12:54:23  *** Eddi|zuHause has joined #openttd
12:55:30  <frosch123> [14:53] <TrueBrain> I am shocked nobody does that .. <- probably because the in-game serverlist is not as advanced as your browser
12:55:54  <TrueBrain> owh, the sentry entry happened because of the roll-over
12:56:01  <frosch123> and i think ottd will even strip the private-range chars that we use for the sprite train symbol
12:56:35  <TrueBrain> it showed them as a weird icon
12:56:40  <TrueBrain> I assumed my font ingame didn't support it
12:59:57  <frosch123> "2607:5300:60:1bde::1:4030" <- is that how you print ipv6 addresses? i thought it was "[2607:5300:60:1bde::1]:4030
13:00:19  <TrueBrain> make a pull request! :D
13:00:54  <frosch123> is that an answer?
13:01:05  <TrueBrain> that is me agreeing with you :)
13:01:07  <frosch123> are the [ ] optional? or required?
13:01:17  <TrueBrain> it is better readable with them
13:01:24  <LordAro> i don't think it's technically necessary
13:01:42  <TrueBrain> and in the example your give, the current format is even wrong
13:01:46  <frosch123> how do you distinguish ip and port then?
13:01:50  <TrueBrain> as :: resolves to "as many 0's as needed"
13:01:55  <LordAro> hmm yes
13:01:56  <TrueBrain> so it is unclear there is a port there :P
13:02:15  * LordAro runs away again
13:04:01  <TrueBrain> <- but frosch123 , just look at that .. it is so pretty!
13:04:26  <TrueBrain> <- I really like this :) It is a nice addition :)
13:04:31  <frosch123> ah, those are trains... i was wondering the other day what they were...
13:04:43  <frosch123> i think i settled for aliens
13:04:54  <TrueBrain> use a better font!
13:06:56  <frosch123> <- you have hard competition though
13:09:36  <_dp_> Current date: 	19353-06-30 xD
13:11:11  <DorpsGek_III> [OpenTTD/master-server-web] frosch123 opened issue #7: Display of ipv6 addresses
13:13:03  <TrueBrain> frosch123: pretty sure that the fix took more time than the issue :P
13:13:45  <frosch123> no, i would have to clone and clone and branch and commit and push and pr
13:16:42  <frosch123> oh, i can use the gh editor for that
13:18:20  <frosch123> hmm, i think that would make a branch on the official repo though
13:20:41  <DorpsGek_III> [OpenTTD/master-server-web] frosch123 opened pull request #8: Fix #7: distinguish ipv6 address and port.
13:21:20  <frosch123> ok, looks like i could skip 3 of the 6 steps that way
13:21:43  <frosch123> though it is completely untested now :p
13:21:52  <DorpsGek_III> [OpenTTD/master-server-web] TrueBrain approved pull request #8: Fix #7: distinguish ipv6 address and port.
13:22:03  <TrueBrain> just those dots at the end of commit messages
13:22:07  <TrueBrain> that makes no sense .... :P
13:22:40  <frosch123> hey, i already continued with lowercase after :
13:22:41  <frosch123> just for you
13:23:28  <DorpsGek_III> [OpenTTD/actions] dependabot[bot] opened pull request #5: Build(deps): Bump node-fetch from 2.6.0 to 2.6.1 in /deploy-aws
13:23:42  <DorpsGek_III> [OpenTTD/actions] dependabot[bot] opened pull request #6: Build(deps): Bump node-fetch from 2.6.0 to 2.6.1 in /deployments-create
13:23:57  <DorpsGek_III> [OpenTTD/actions] dependabot[bot] opened pull request #7: Build(deps): Bump node-fetch from 2.6.0 to 2.6.1 in /deployments-update
13:24:09  <frosch123> how many will follow?
13:24:18  <DorpsGek_III> [OpenTTD/actions] dependabot[bot] opened pull request #8: Build(deps): Bump node-fetch from 2.6.0 to 2.6.1 in /checkout-dispatch
13:25:18  <DorpsGek_III> [OpenTTD/master-server-web] frosch123 merged pull request #8: Fix #7: distinguish ipv6 address and port.
13:25:20  <DorpsGek_III> [OpenTTD/master-server-web] frosch123 closed issue #7: Display of ipv6 addresses
13:26:33  <LordAro> uh, why are there 3 PRs apparently for the same thing?
13:26:34  <frosch123> does the list on staging magically fill with the real servers? or does it stay empty unless you patch your server?
13:26:50  <frosch123> LordAro: 4
13:29:20  <_dp_> meanwhile...
13:30:36  <frosch123> boo, c&p
13:30:46  <_dp_> :p
13:34:35  <_dp_> ppl keep doing events with breakdowns and I got too annoyed building stupid depots everywhere xD
13:36:19  <frosch123> just cheat with overflow depots before the loading station
13:36:57  <_dp_> it's bad if loading time is too long
13:37:10  <_dp_> so I build overflow before and forced service after
13:40:08  <_dp_>
13:40:12  <_dp_> not overflow but whatever
13:41:31  <_dp_> also it super annoying whet people in same company are fighting for service interval :p
13:44:32  *** gelignite has joined #openttd
13:44:33  <_dp_> why are some ppl even setting it to 5%, doesn't it just screw all the pathfinding?
13:44:57  <TrueBrain> frosch123: someone would have to add a server for it to show up on staging :P It is no longer shared
13:46:16  <TrueBrain> Invalid file descriptor: -1 <- wtf is asyncio doing? :P
13:47:25  <TrueBrain> "socket.accept() out of system resource"
13:47:28  <TrueBrain> owh, this is bad ..
13:47:40  <TrueBrain> something is not closing sockets ...
13:51:17  *** Progman has quit IRC
13:57:44  <DorpsGek_III> [OpenTTD/master-server] TrueBrain opened pull request #10: Fix: close the UDP socket when we are done with it
13:57:54  <TrueBrain> I think this solves it, but very hard to validate ..
14:02:50  <frosch123> try..finally :) poor with...-fanboys
14:03:01  <DorpsGek_III> [OpenTTD/master-server] frosch123 approved pull request #10: Fix: close the UDP socket when we are done with it
14:03:26  <DorpsGek_III> [OpenTTD/master-server] TrueBrain merged pull request #10: Fix: close the UDP socket when we are done with it
14:03:46  <TrueBrain> what I didn't know, that if a task is cancelled, an exception is thrown inside the task
14:03:50  <TrueBrain> the reason finally-statements do hit
14:03:52  <TrueBrain> pretty nice
14:04:16  <TrueBrain> [OpenTTD/master-server] TrueBrain created new tag: 1.0.4
14:04:24  <TrueBrain> we will see in an hour if this fixed anything :D
14:06:26  <TrueBrain> okay, I can confirm that fixes leaking fds :D
14:06:33  <TrueBrain> I forgot in linux you can just see how many fds are open
14:07:23  <frosch123> i know how to use /proc, but you can see that inside an aws container?
14:07:38  <TrueBrain> no, I was reproducing this locally
14:07:41  <TrueBrain> and that was easy enough :D
14:07:46  <Eddi|zuHause> there's "lsof"
14:07:57  <TrueBrain> or just "ls /proc/<pid>/fd | wc -l"
14:08:25  <frosch123> know your tools :)
14:09:37  <TrueBrain> either way, fd-leak should be gone now :)
14:09:43  <TrueBrain> how silly of me .. leaking fds :P
14:10:12  <TrueBrain> and I absolutely love sentry :) Shows really quickly what is going wrong where :)
14:10:57  <TrueBrain> I also like the infra btw ... the pods crashed, but there was a new one running before anyone would really notice :)
14:12:46  <DorpsGek_III> [OpenTTD/master-server-web] TrueBrain created new tag: 1.0.2
14:12:52  <TrueBrain> let's bring your [] fix online :)
14:14:12  <DorpsGek_III> [OpenTTD/website] TrueBrain created new tag: 1.4.0
14:14:20  <TrueBrain> and let's bring the new security pages to production :)
14:14:38  <frosch123> now you are just flipping switches :p
14:14:42  <TrueBrain> yes :D
14:14:51  <TrueBrain> it is sad, the moment you realise the infra is more active than the game :P
14:14:53  <TrueBrain> just saying :D
14:15:21  <TrueBrain> and yes, I really do enjoy that I can just punch a few buttons and in a moment we have a brand new version online .. I absolutely love it
14:24:44  *** gelignite has quit IRC
14:25:03  <TrueBrain> frosch123: much better :)
14:25:57  <frosch123> i was scary to edit via gh interface and not test it :)
14:38:00  <TrueBrain>  Pfft, have a bit of faith :p
14:38:59  <frosch123> i am a c++ programmer. [] have very different meaning in a lot of places :p
14:39:43  *** arikover has joined #openttd
15:05:36  *** sla_ro|master has quit IRC
15:24:25  *** arikover has quit IRC
15:42:43  *** sla_ro|master has joined #openttd
15:57:21  *** debdog has joined #openttd
16:13:58  *** Gustavo6046 has joined #openttd
16:16:23  *** Wormnest has joined #openttd
16:16:39  *** Progman has joined #openttd
16:17:45  *** Flygon has quit IRC
17:08:41  <Wolf01>
17:10:00  <Eddi|zuHause> Wolf01: that will be a great AI when fully trained :)
17:10:46  <frosch123> the bugs are the two functions who have english names
17:23:40  <Wolf01> Added by the developer after decompiling?
17:31:27  <frosch123> decompiled javascript?
17:46:33  <Wolf01> Unminified and prettified
18:22:06  <TrueBrain> No sentry errors anymore so far .. guess it is stable :D
18:23:26  <frosch123> no more mysql :)
18:25:05  <TrueBrain> Valid point :D
18:25:24  <TrueBrain> Another VM I can delete \o/
18:25:35  <frosch123> does the wiki have its own instance?
18:29:28  <TrueBrain> No .... I just ignore its existence.. it is easier that way
18:30:22  <TrueBrain> Tomorrow I am first going to check out your wiki work, and how rusty my Ruby is ...
18:36:18  *** gelignite has joined #openttd
18:57:45  <Eddi|zuHause> i'm assuming ruby gets more rusty when you learn rust? :p
18:59:59  <_dp_> everything gets
19:00:07  * _dp_ missing rust enums in c++ 
19:05:05  <Eddi|zuHause> i've never actually looked into what rust actually is
19:06:13  <frosch123> it has the stuff that modern c++ is about, but without the legacy stuff that c++ has to suffer from
19:10:39  *** frosch123 has quit IRC
19:40:33  <TrueBrain> the master-server database size is current a wopping 300 kB
19:45:54  <_dp_> TrueBrain, whoa, bigdata :p
19:46:52  *** supermop_Home_ has joined #openttd
20:21:24  <TrueBrain> and it uses ~2 RCU and 6 WCU (in case you know DynamoDB) .. which is ... 2 euro a month? :P
20:21:28  <TrueBrain> (well, it is within the free-tier
20:21:46  <TrueBrain> cheapest database evah
20:35:59  <_dp_> I did some dynamodb years ago but only thing I remember is that I had to do some stupid crap with data schema to reduce costs
20:41:48  *** WormnestAndroid has quit IRC
20:42:11  *** WormnestAndroid has joined #openttd
20:45:42  <TrueBrain> it is a very simple database
20:45:45  <TrueBrain> and very fast
20:45:56  <TrueBrain> but only a single index, with at most a single "filter" that is quick/cheap
20:46:34  <TrueBrain> so it took me a bit of time to model something for the master-server that worked :P
20:46:38  <TrueBrain> total of 2 tables and 6 indexes :D
21:02:52  *** virtualrandomnumber has joined #openttd
21:03:07  *** virtualrandomnumber has quit IRC
21:05:15  *** gelignite has quit IRC
21:17:16  *** sla_ro|master has quit IRC
21:43:24  *** WormnestAndroid has quit IRC
21:44:31  *** WormnestAndroid has joined #openttd
21:45:27  *** heffer has quit IRC
21:52:17  *** Smedles_ has joined #openttd
21:53:49  *** Smedles has quit IRC
21:56:27  *** heffer has joined #openttd
22:10:47  *** Wolf01 has quit IRC
22:27:48  *** tokai|noir has quit IRC
22:45:55  *** tokai has joined #openttd
22:45:55  *** ChanServ sets mode: +v tokai
22:58:39  *** tokai has quit IRC
23:11:50  *** Wormnest has quit IRC
23:59:14  *** Gustavo6046 has quit IRC

Powered by YARRSTE version: svn-trunk