00:03:09 Unstable branch on CRAWL.XTAHUA.COM updated to: 0.24-a0-447-ge8da9206dc (34) 00:04:39 Stable (0.23) branch on CRAWL.XTAHUA.COM updated to: 0.23.1-79-g0e89c35649 00:13:41 -!- Tiobot is now known as Guest42755 00:55:31 this might be related to the des cache corruption cpo ran into in the past 03:11:47 Unstable branch on crawl.beRotato.org updated to: 0.24-a0-447-ge8da920 (34) 03:52:12 Fork (stoatsoup) on crawl.kelbi.org updated to: 0.21.2-992-g8c730f8d2a 03:57:19 -!- amalloy is now known as amalloy_ 09:58:59 Could someone update Stoat Soup on CKO, please? I've written another Webtiles-only bug and failed to test Webtiles. :-( 10:05:23 03advil02 07* 0.24-a0-448-g8f2d45b: Sort tags_string() again 10(11 minutes ago, 1 file, 2+ 1-) 13https://github.com/crawl/crawl/commit/8f2d45b691dc 10:05:23 03advil02 07* 0.24-a0-449-gc0fab78: Fix a des cache random tag issue 10(5 minutes ago, 1 file, 3+ 1-) 13https://github.com/crawl/crawl/commit/c0fab7842f11 10:06:59 03advil02 07[stone_soup-0.23] * 0.23.2-31-g48d7f60: Fix a des cache random tag issue 10(7 minutes ago, 1 file, 3+ 1-) 13https://github.com/crawl/crawl/commit/48d7f609f902 10:08:00 !tell gammafunk I'm 65% certain that c0fab7842f11 fixes what was causing your gotm seed issue 10:08:01 advil: OK, I'll let gammafunk know. 10:08:41 Pinkbeast I triggered a rebuild but the rebuild interface seems to have frozen partway through, not sure what's up with that 10:08:51 we'll see if it actually updates I guess 10:28:11 Fork (stoatsoup) on crawl.kelbi.org updated to: 0.21.2-993-gf1a063321b 10:29:37 03advil02 07* 0.24-a0-450-g1dd0858: More des cache stability fixes 10(42 seconds ago, 2 files, 5+ 5-) 13https://github.com/crawl/crawl/commit/1dd085888ca3 10:30:46 03advil02 07[stone_soup-0.23] * 0.23.2-32-g37ab1d4: More des cache stability fixes 10(2 minutes ago, 2 files, 5+ 5-) 13https://github.com/crawl/crawl/commit/37ab1d464986 10:32:51 Unstable branch on crawl.kelbi.org updated to: 0.24-a0-450-g1dd085888c (34) 10:41:36 advil: Thanks; just tested and the fix has definitely been deployed 11:22:24 Stable (0.23) branch on underhound.eu updated to: 0.23.1-81-g37ab1d4649 13:02:23 Stable (0.23) branch on crawl.akrasiac.org updated to: 0.23.1-81-g37ab1d4 13:07:51 Unstable branch on crawl.akrasiac.org updated to: 0.24-a0-450-g1dd0858 (34) 13:43:41 -!- amalloy_ is now known as amalloy 14:30:57 advil: there is a bug in nginx that won't let me turn off gzip compression for fcgi scripts. i don't know how to fix it other than wait for buster to come out 14:31:17 ah 14:31:20 it worked on the second try 14:31:33 it should only be visual, the build will still run 14:31:53 hm it didn't seem like it did, in that the bot didn't announce 14:32:37 well, it worked eventually so not a big deal 14:32:38 maybe i'm mistaken then, that's even more reason to get it fixed 14:33:11 i tried upgrading just the relevant package to its buster version and i ran into problems, i'm afraid to mess up the system so i'm just waiting for buster to come out in a few months 15:21:32 %git c0fab7842f11 15:21:33 07advil02 * 0.24-a0-449-gc0fab78: Fix a des cache random tag issue 10(5 hours ago, 1 file, 3+ 1-) 13https://github.com/crawl/crawl/commit/c0fab7842f11 16:17:35 advil: I don't suppose there's really a way to verify that it fixes things? 16:17:43 perhaps I should have backed up my des cache before clearing it 16:58:39 it might be directly testable 17:00:07 one of the affected vaults (beast_lair_carwin_2) is present on the floor where things diverge, at least 17:17:01 gammafunk: now 100% certain. I tested it by forcing an extra tag on that vault during validation, and it generates the dungeon you saw 17:18:58 incidentally this would only be likely to happen for dev installs because we end up regenerating lots of des files in random order; des generation is done with a fixed seed and most users only do it once on first startup for all files at once 17:19:21 possibly for mac users who switch around versions (since they use a user-wide des cache for some reason) 17:19:48 though I guess I have no idea where the des cache is for linux installs 17:19:54 windows it's per version 17:20:12 -!- neunon_ is now known as neunon 17:23:18 advil: well I'm using a local build, and it's just in source/saves/des 17:23:37 yeah I mean for however regular linux users do it 17:23:43 with a .deb or whatever 17:23:59 I think possibly ~/.crawl 17:24:11 yes, ~/.crawl/saves 17:24:19 with same subdirs 17:24:28 ah, so per-user like mac then 17:24:31 right 17:29:57 03advil02 07* 0.24-a0-451-g263dff2: Tweak the beast_lair_carwin fix 10(25 seconds ago, 1 file, 3+ 4-) 13https://github.com/crawl/crawl/commit/263dff2034ed 17:30:42 03advil02 07[stone_soup-0.23] * 0.23.2-33-g5129a83: Tweak the beast_lair_carwin fix 10(70 seconds ago, 1 file, 3+ 4-) 13https://github.com/crawl/crawl/commit/5129a8318406 17:32:15 Unstable branch on crawl.kelbi.org updated to: 0.24-a0-451-g263dff2034 (34) 18:22:38 Unstable branch on underhound.eu updated to: 0.24-a0-451-g263dff2034 (34) 18:53:38 if we aren't getting close to tourney time, I'm going to be doing some server maintenance stuff... 18:54:05 I hope I don't botch anything. I will try to minimize disruption 19:12:02 -!- amalloy is now known as amalloy_ 19:28:27 johnstein: good luck! 19:58:42 floraline: whats this about a nginx bug causing issues? 20:02:07 advil: looking at your fixes, it seems like vault lua should not have randomness during validation? It might be tough to communicate that to future vault makers. Could the random functions be mocked during validation for maps instead? or something... I just worry in 12 months someone will submit a new vault with this issue and it won't be caught 20:33:28 I looked into just making random calls fully non-random during validation at one point, but decided that was too involved 20:34:17 it's not that they can't have randomization, but tags in particular can't be random at validation time 20:34:39 I did change the border wall in one map because that shows up in the .idx files, but that's not very important, it just makes the diff clean 20:34:51 one set of maps 20:50:41 03gammafunk02 07* 0.24-a0-452-g26b8ecb: Sif Rework Part 1: Piety on kills 10(10 days ago, 2 files, 6+ 46-) 13https://github.com/crawl/crawl/commit/26b8ecb3bfd1 20:50:41 03gammafunk02 07* 0.24-a0-453-g3db0654: Sif Rework Part 2: Grant some abilities earlier 10(10 days ago, 2 files, 6+ 6-) 13https://github.com/crawl/crawl/commit/3db0654b3e36 20:50:41 03gammafunk02 07* 0.24-a0-454-g0c6bcda: Sif Rework Part 3: Remove Divine Energy 10(10 days ago, 15 files, 21+ 89-) 13https://github.com/crawl/crawl/commit/0c6bcda03798 20:50:41 03gammafunk02 07* 0.24-a0-455-ge12f10f: Sif Rework Part 4: Add a Divine Exegesis ability 10(10 days ago, 14 files, 152+ 39-) 13https://github.com/crawl/crawl/commit/e12f10fa2d7f 20:50:41 03gammafunk02 07* 0.24-a0-456-g09db799: Sif Rework Part 5: Remove miscast protection 10(10 days ago, 4 files, 2+ 36-) 13https://github.com/crawl/crawl/commit/09db7996fae9 20:50:41 03gammafunk02 07* 0.24-a0-457-gaec7895: Clean up a condition 10(3 hours ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/aec789570c1d 20:51:59 alexjurkiewicz: https://metadata.ftp-master.debian.org/changelogs//main/f/fcgiwrap/fcgiwrap_1.1.0-12_changelog 20:52:12 patch 11 adds an option to fully disable output buffering in debian package fcgiwrap 20:52:28 i'm using this package and for whatever reason stretch only has up to patch 9 20:53:26 so not exactly an nginx bug 21:04:47 03advil02 07* 0.24-a0-458-gfcb5e6d: More specific documentation about des stability and tags 10(79 seconds ago, 2 files, 27+ 1-) 13https://github.com/crawl/crawl/commit/fcb5e6da0e9c 21:23:00 Unstable branch on crawl.kelbi.org updated to: 0.24-a0-457-gaec789570c (34) 21:28:56 !kw newsif 21:28:57 Keyword: newsif => vlong>=0.19-a0-890-gff1bd09 sif 21:29:11 !kw newsif vlong>=0.24-a0-457-gaec7895 21:29:13 Defined keyword: newsif => vlong>=0.24-a0-457-gaec7895 21:30:53 hm interesting, thanks 21:32:19 Unstable branch on crawl.kelbi.org updated to: 0.24-a0-458-gfcb5e6da0e (34) 21:35:58 elliptic: did you have a verdict on new vampires? I see the venerable kiku zombie armour failed to protect yours from zot trap paralysis 21:39:05 -!- amalloy_ is now known as amalloy 23:38:01 -!- Krakhan|2 is now known as Krakhan 23:51:48 alien (L18 FoHu) ASSERT(attacker) in 'attack.cc' at line 354 failed. (Shoals:4)