01:04:44 Stable branch on crawl.develz.org updated to: 0.26.0-20-g3ca415f5b8 (34) 01:34:31 Unstable branch on crawl.develz.org updated to: 0.27-a0-109-g416527ed24 (34) 01:55:24 Windows builds of master branch on crawl.develz.org updated to: 0.27-a0-109-g416527ed24 02:47:57 Unstable branch on cbro.berotato.org updated to: 0.27-a0-109-g416527ed24 (34) 02:53:56 Monster database of master branch on crawl.develz.org updated to: 0.27-a0-109-g416527ed24 03:45:36 -!- amalloy is now known as amalloy_ 03:47:05 Stable (0.26) branch on cbro.berotato.org updated to: 0.26.0-20-g3ca415f5b8 03:47:49 Stable (0.26) branch on crawl.kelbi.org updated to: 0.26.0-20-g3ca415f5b8 03:50:35 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-3859-ge9853cb0d6 03:51:36 Fork (bcadrencrawl) on crawl.kelbi.org updated to: 0.03-98-gc08451f611 04:37:28 <03w​heals> that interaction seems perfectly reasonable to me, to be honest 04:37:31 <03w​heals> i guess the question is whether it's consistent with what rampaging does with teleport traps 05:54:23 Rjs (L14 MiMo) Crash caused by signal #6: Aborted (Elf:1) 05:56:17 That crash ^^ is http://crawl.develz.org/saves/dumps/Rjs-c5c842f807-210124-1054.cs 05:58:10 and this time I was running ssh inside a script command, so I got the error messages: 05:58:15 Generating dungeon...o\\/o/ building a lair branch \o\ /o/ \o\ /o/building another lair branch \o\ /o/ \o\ /o/ building the Vaults \o\ /o/ \o\ /o/ \o\ building the Crypt /o/ \o\ /o/building the Depths \o\ /o/ \o\ /o/ \o\building the Vestibule of Hell 05:58:15 05:58:18 /o/ building the Elven Halls \o\terminate called after throwing an instance of 'std::bad_alloc' 05:58:18 what(): std::bad_alloc 05:58:43 (I was entering Elf when it crashed) 05:59:32 <03w​heals> !crashlog 05:59:34 <04C​erebot> 23345. Rjs, XL14 MiMo, T:19275 (milestone): http://crawl.develz.org/morgues/0.26/Rjs/crash-Rjs-20210124-105422.txt 06:07:41 Has anyone else reported level generation crashes recently? Since I've now had three in about a week on CDO, and not a lot of people seem to be playing on CDO, I wonder if it could be related to how CDO is set up... 06:09:40 <03w​heals> it could be, a bad_alloc gets thrown when the program is out of memory 06:09:40 <03w​heals> though you'd expect that to happen more on busier servers 06:10:03 I'm not sure how the process getting out of memory would easily happen on Linux - since usually the oomkiller would kill the process and log in syslog, not fail the allocation? Maybe there is an ulimit on memory usage or something? 06:10:49 (or it is a crawl bug with an infinite loop that allocates memory and fills up the address space, but in that case you would expect it to happen on all servers) 06:14:35 I now tried again and it didn't reproduce, just like the earlier times - so there is also some randomness involved, either in the level generation (should it always give the same result nowadays?) or in the available memory. 06:30:45 Rjs: not saying there's no configuration issue, but CDO doesn't get a lot of players because it doesn't have WebTiles, and 95% of crawl's players use WebTiles 06:32:42 I communicatee with the admin Napkin regularly and advil and I both have logins there for website/tournament stuff, so can potentially take a look and forward info to Napkin 06:32:42 *communicate 06:42:39 Ok. I've been meaning to try out and maybe switch to WebTiles as well, but I returned to crawl after a long break and thought I'd start with what I'm used to... 06:44:32 So in theory this could also be a memory leak in the non-Tiles code, which gets triggered on level gen because that uses a lot of memory at once - but still you'd expect some other non-tiles player to have seen it (and I suppose the non-tiles code is a lot simpler), so maybe this isn't likely. 07:05:42 Another one in the same game when entering Slime: http://crawl.develz.org/saves/dumps/Rjs-c5c842f807-210124-1204.cs 07:06:06 Generating dungeon...o\\/o/ building the Elven Halls \o\ /o/building the Realm of Zot \o\ /o/ \o\terminate called after throwing an instance of 'std::bad_alloc' 07:06:06 what(): std::bad_alloc 07:06:15 Crash caused by signal #6: Abortedterminate called recursively 07:06:15 Recursive crash. 07:06:15 terminate called recursively 09:51:28 Rjs I think it's only you; I've tried to replicate those crashes with your saves and they unfortunately don't replicate off CDO, and I don't have the rights to try on cdo 09:51:40 I do suspect it's an out-of-memory issue 09:52:03 I doubt it's a leak per se 09:52:37 I just found an ancient comment in the chat logs "14:11:30 on CDO for example there is a hard cap on 100MB" 09:53:00 if something like that is still true, it definitely seems plausible for modern crawl to hit 100MB 12:06:43 Unstable branch on crawl.akrasiac.org updated to: 0.27-a0-109-g416527e (34) 13:43:52 stickyfingers (L19 HOGl) Crash caused by signal #6: Aborted (Crypt:1) 14:00:13 <03w​heals> wonder if it's the same crash again 14:00:16 <03w​heals> !crashlog 14:00:18 <04C​erebot> 23347. stickyfingers, XL19 HOGl, T:34839 (milestone): http://crawl.develz.org/morgues/0.26/stickyfingers/crash-stickyfingers-20210124-184350.txt 14:01:49 <03w​heals> yup 14:01:59 <03w​heals> potentially relevant that err_trace is on the dlua stack? 14:38:29 <10P​leasingFungus> @advil do you know how much it would cost to store all of the ancient ttyrecs somewhere, maybe on s3 glacier or something 15:52:36 !kw nchoice 15:52:36 Keyword: nchoice => HaEn|FoSu|PaWn|GhDe|BaAE|MfDe|DEAr|PaGl|MuAM|VSCj|TeAK|VpVM|HaEE|BaNe|KoTm|HuEn|SpIE|GnDe|DgAM|DDAE|DEWr|MfWz|DETm|FoFE|VpFE|HaNe|TeEE|GhWz|BaBr|DDSu|PaHu|TeIE|FeDe|GhFE|VSAM|VpAr|BaWr|DDAK|FoCj|NaDe|KoAE|GhVM|HaTm|MfEE|DDEn|FoIE|DrDe|DEHu|MfCj|BaWz|TeAM|KoFE|VpAK|PaNe|HaSu|GhAE|DgDe|PaIE|DEAK|PaEE 15:52:48 =nemelex TeEn 15:52:50 Defined keyword: nchoice => HaEn|FoSu|PaWn|GhDe|BaAE|MfDe|DEAr|PaGl|MuAM|VSCj|TeAK|VpVM|HaEE|BaNe|KoTm|HuEn|SpIE|GnDe|DgAM|DDAE|DEWr|MfWz|DETm|FoFE|VpFE|HaNe|TeEE|GhWz|BaBr|DDSu|PaHu|TeIE|FeDe|GhFE|VSAM|VpAr|BaWr|DDAK|FoCj|NaDe|KoAE|GhVM|HaTm|MfEE|DDEn|FoIE|DrDe|DEHu|MfCj|BaWz|TeAM|KoFE|VpAK|PaNe|HaSu|GhAE|DgDe|PaIE|DEAK|PaEE|TeEn 16:34:22 03PleasingFungus02 07* 0.27-a0-110-ge2d8f9c: Rewrite LRD setup 10(4 weeks ago, 1 file, 211+ 197-) 13https://github.com/crawl/crawl/commit/e2d8f9ccf320 16:34:22 03PleasingFungus02 07* 0.27-a0-111-gb6a8d6b: Refactor further 10(4 weeks ago, 1 file, 60+ 45-) 13https://github.com/crawl/crawl/commit/b6a8d6b5b39c 16:37:27 03PleasingFungus02 07* 0.27-a0-112-gf16e6d0: Remove Summon Greater Demon 10(4 weeks ago, 6 files, 4+ 42-) 13https://github.com/crawl/crawl/commit/f16e6d0322db 16:38:16 03PleasingFungus02 07* 0.27-a0-113-gd412354: Reduce net trap trigger chance 10(8 days ago, 1 file, 3+ 0-) 13https://github.com/crawl/crawl/commit/d412354acc90 16:39:07 03PleasingFungus02 07* 0.27-a0-114-g3644e03: Display accuracy on ranged targeters 10(8 days ago, 4 files, 80+ 73-) 13https://github.com/crawl/crawl/commit/3644e03777a4 16:40:41 03PleasingFungus02 07* 0.27-a0-115-gee1497a: Show cloud damage for player spells 10(28 hours ago, 3 files, 37+ 3-) 13https://github.com/crawl/crawl/commit/ee1497a6375b 16:41:46 03PleasingFungus02 07* 0.27-a0-116-g19610a9: Un-de-enhance Scarf of Shadows spells 10(28 hours ago, 2 files, 2+ 5-) 13https://github.com/crawl/crawl/commit/19610a9a0ceb 16:43:50 03PleasingFungus02 07* 0.27-a0-117-gab96107: Reduce gnolls attributes to human 10(28 hours ago, 1 file, 3+ 3-) 13https://github.com/crawl/crawl/commit/ab96107f46c2 16:45:44 03PleasingFungus02 07* 0.27-a0-118-gb3a650c: Increase Stone Arrow damage by 2 10(4 weeks ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/b3a650cb6aa7 16:50:24 03PleasingFungus02 07* 0.27-a0-119-g68f98bd: Retheme /disintegrate -> /mindburst 10(4 weeks ago, 32 files, 65+ 58-) 13https://github.com/crawl/crawl/commit/68f98bdfd779 16:54:35 03PleasingFungus02 07* 0.27-a0-120-gd7dd4bc: Give monsters infinite ammo 10(3 weeks ago, 3 files, 5+ 40-) 13https://github.com/crawl/crawl/commit/d7dd4bcd5325 16:54:35 03PleasingFungus02 07* 0.27-a0-121-ged86a5f: Make nets finite again 10(8 days ago, 3 files, 9+ 5-) 13https://github.com/crawl/crawl/commit/ed86a5fd0030 16:58:17 04Build failed for 08master @ f16e6d03 06https://github.com/crawl/crawl/actions/runs/507945455 16:58:22 04Build failed for 08master @ b6a8d6b5 06https://github.com/crawl/crawl/actions/runs/507941965 16:58:49 New branch created: armour_spirit (2 commits) 13https://github.com/crawl/crawl/tree/armour_spirit 16:58:49 03PleasingFungus02 07[armour_spirit] * 0.27-a0-94-ge42f7de: WIP summon armour spirit 10(8 days ago, 12 files, 68+ 0-) 13https://github.com/crawl/crawl/commit/e42f7dea1ae2 16:58:49 03PleasingFungus02 07[armour_spirit] * 0.27-a0-95-g9390861: New spell: Summon Panzergeist 10(8 days ago, 15 files, 127+ 61-) 13https://github.com/crawl/crawl/commit/9390861aeca0 16:59:49 03PleasingFungus02 07[armour_spirit] * 0.27-a0-122-g6683436: New spell: Summon Panzergeist 10(8 days ago, 16 files, 167+ 33-) 13https://github.com/crawl/crawl/commit/6683436658e5 17:01:06 03PleasingFungus02 07* 0.27-a0-122-g58b3cc3: Checkwhite 10(23 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/58b3cc31320a 17:01:26 New branch created: pull/1696 (1 commit) 13https://github.com/crawl/crawl/pull/1696 17:01:26 Branch pull/1696 updated to be equal with armour_spirit: 13https://github.com/crawl/crawl/pull/1696 17:01:42 Unstable branch on crawl.kelbi.org updated to: 0.27-a0-117-gab96107f46 (34) 17:04:14 @PleasingFungus were you planning to remove regular summon demon as well? i'd been meaning to do that one (as well as SGD) but i thought i remembered seeing a commit for it already in a branch maybe 17:04:49 New branch created: projected_weapon (1 commit) 13https://github.com/crawl/crawl/tree/projected_weapon 17:04:49 03PleasingFungus02 07[projected_weapon] * 0.27-a0-123-gdbd6142: New spell: Projected Weapon (L5 Translocations) 10(7 days ago, 10 files, 98+ 7-) 13https://github.com/crawl/crawl/commit/dbd6142137b5 17:06:30 <10P​leasingFungus> kate-: i have a commit, didn't push it 17:06:38 <10P​leasingFungus> can push it soonish; was undecided 17:09:17 aha yeah, i see the branch now. and yeah, feels like it's in pretty much the same place as SGD where not only are there lots of other summons around that level, the other options are mostly more interesting and fun designs already 17:10:52 New branch created: pull/1697 (1 commit) 13https://github.com/crawl/crawl/pull/1697 17:10:52 Branch pull/1697 updated to be equal with projected_weapon: 13https://github.com/crawl/crawl/pull/1697 17:14:46 -!- amalloy_ is now known as amalloy 17:22:39 04Build failed for 08master @ d412354a 06https://github.com/crawl/crawl/actions/runs/507946186 17:25:22 04Build failed for 08master @ 3644e037 06https://github.com/crawl/crawl/actions/runs/507946757 17:30:17 04Build failed for 08master @ ee1497a6 06https://github.com/crawl/crawl/actions/runs/507949219 17:30:17 04Build failed for 08master @ 19610a9a 06https://github.com/crawl/crawl/actions/runs/507950517 17:30:18 04Build failed for 08master @ b3a650cb 06https://github.com/crawl/crawl/actions/runs/507955350 17:30:24 04Build failed for 08master @ ab96107f 06https://github.com/crawl/crawl/actions/runs/507953092 17:31:53 04Build failed for 08master @ 68f98bdf 06https://github.com/crawl/crawl/actions/runs/507960920 17:35:00 Unstable branch on crawl.kelbi.org updated to: 0.27-a0-122-g58b3cc3132 (34) 17:35:18 <03w​heals> i think the build may have failed 17:36:23 04Build failed for 08master @ ed86a5fd 06https://github.com/crawl/crawl/actions/runs/507965564 17:44:29 04Build failed for 08armour_spirit @ 9390861a 06https://github.com/crawl/crawl/actions/runs/507970622 17:48:47 04Build failed for 08armour_spirit @ 66834366 06https://github.com/crawl/crawl/actions/runs/507971693 18:04:27 04Build failed for 08projected_weapon @ dbd61421 06https://github.com/crawl/crawl/actions/runs/507977211 18:11:36 <10P​leasingFungus> not sure what'd make you think that 18:22:59 Unstable branch on underhound.eu updated to: 0.27-a0-122-g58b3cc3132 (34) 18:49:20 PleasingFungus: it's not super expensive, and I started down that route, but it becomes extremely unmanageable because of the very large number of small files 18:50:26 not sure about glacier though, it's not very trivial to get anything out of it (and I suspect we never would) 19:11:12 -!- FutureShock is now known as abandonsapien 19:12:06 -!- abandonsapien is now known as AbandonSapien 20:13:04 <10P​leasingFungus> hrm 20:13:21 <10P​leasingFungus> there's also intermediate versions, long term storage classes for S3 20:21:14 <10P​leasingFungus> i’m not clear on the current technical challenges, though, so hard to advise 20:21:14 <10P​leasingFungus> would just be sad to lose years of history, esp if we wipe all the old ttyrecs instead of just some definition of ‘boring’ ones 21:03:50 03PleasingFungus02 07[projected_weapon] * 0.27-a0-124-g456c784: Rename and re-level Projected Weapon 10(8 minutes ago, 2 files, 8+ 8-) 13https://github.com/crawl/crawl/commit/456c784262e9 21:03:50 03PleasingFungus02 07[projected_weapon] * 0.27-a0-125-g985a322: Don't allow sleepstabbing with Proj Weapon 10(34 seconds ago, 1 file, 11+ 3-) 13https://github.com/crawl/crawl/commit/985a322822c6 21:04:07 Branch pull/1697 updated to be equal with projected_weapon: 13https://github.com/crawl/crawl/pull/1697 21:04:31 03PleasingFungus02 07[armour_spirit] * 0.27-a0-123-g4dac63e: Raise Panzergeist's level a bit 10(31 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/4dac63e651b0 21:04:42 New branch created: simplispectral (1 commit) 13https://github.com/crawl/crawl/tree/simplispectral 21:04:42 03PleasingFungus02 07[simplispectral] * 0.27-a0-109-gaf31ee7: Remove Evocation scaling from Spectral 10(32 hours ago, 3 files, 7+ 17-) 13https://github.com/crawl/crawl/commit/af31ee78d938 21:04:56 Branch pull/1696 updated to be equal with armour_spirit: 13https://github.com/crawl/crawl/pull/1696 21:12:43 03PleasingFungus02 07[simplispectral] * 0.27-a0-123-g9ea0f8b: Remove Evocation scaling from Spectral 10(32 hours ago, 3 files, 7+ 17-) 13https://github.com/crawl/crawl/commit/9ea0f8b6df3a 21:12:58 New branch created: pull/1698 (1 commit) 13https://github.com/crawl/crawl/pull/1698 21:12:58 Branch pull/1698 updated to be equal with simplispectral: 13https://github.com/crawl/crawl/pull/1698 21:21:19 03PleasingFungus02 07* 0.27-a0-123-geb620a3: Remove Summon Demon 10(4 weeks ago, 7 files, 9+ 17-) 13https://github.com/crawl/crawl/commit/eb620a33452e 21:21:19 03PleasingFungus02 07* 0.27-a0-124-g19cb506: Tweak summoning titles 10(3 minutes ago, 1 file, 1+ 11-) 13https://github.com/crawl/crawl/commit/19cb506db8f7 21:27:41 04Build failed for 08projected_weapon @ 985a3228 06https://github.com/crawl/crawl/actions/runs/508261156 21:29:37 04Build failed for 08armour_spirit @ 4dac63e6 06https://github.com/crawl/crawl/actions/runs/508262134 21:34:21 PleasingFungus: genuine question, what would be sad about it? 21:35:20 Unstable branch on crawl.kelbi.org updated to: 0.27-a0-124-g19cb506db8 (34) 21:36:27 <10P​leasingFungus> being able to go back and look at really old games is very cool, and there's plenty of learndb entries that link to old funny games 21:36:32 I'm sorta in the camp where it's sad along the lines of, my parents were for quite a long time sad to throw away their 40yo unused camping gear in their basement and so it mouldered for most of my life and they didn't do it until the oil tank failed right next to that particular collection 21:36:37 <10P​leasingFungus> haha 21:36:54 I think it would be very reasonable to selectively archive things that are actually linked to and accessed 21:37:32 but sometime last year before sequell died and we lost those stats, we were able to look at how much stuff was accessed via footv, and something on the order of 2000 games had *ever* been watched 21:37:34 !lg 21:37:35 564. advil the Vandal (L1 HOHu), entered wizard mode on D:1 (eino_arrival_water_cross) on 2020-12-17 00:34:22, with 0 points after 0 turns and 0:00:07. 21:37:39 !lg * 21:37:41 12684141. PPCmaster the Chopper (L9 DsFi of Gozag), mangled by an orc knight (a +2 great mace) in Bailey (bailey_axe_2) on 2021-01-25 02:35:24, with 2464 points after 7152 turns and 0:12:46. 21:37:43 <10P​leasingFungus> dang 21:37:45 out of that many 21:38:13 <10P​leasingFungus> i'm actually quite surprised by that... i'd guess an order of magnitude higher watch count 21:39:14 I don't know exactly when the sequell stats were last reset before that 21:39:14 and that isn't watch count 21:39:14 just the number of ttyrecs that have been watched 21:39:14 <10P​leasingFungus> sorry, unique tvs watched count? 21:39:14 <10P​leasingFungus> though ofc compared to 12,684,141, even 20k would still be quite small 21:39:25 the main reason I've come up with to keep everything would be if some sort of AI system used it as training data, that's very speculative though 21:42:10 these queries were findable in the logs with the right search term, trying to reconstruct 21:43:10 they also don't include regular downloads 21:43:37 when I priced things out the best option for storage was backblaze B2 btw 21:43:55 I've actually moved the highest ttyrec produces on CAO to my B2 bucket already (both bots) 21:44:56 *producers 21:48:53 !lg * ntv>0 21:48:58 121. bairyn the Marksspriggan (L10 SpDe of Gozag), slain by an elephant on D:10 (david_lair_generic) on 2021-01-24 00:05:28, with 6061 points after 21155 turns and 0:30:47. 21:49:04 !lg * ntv>0 1 21:49:09 1/121. Iaido the Thanatomancer (L27 DsTm of Okawaru), mangled by a reaper (a scythe of speed) on Orc:3 on 2007-07-19 22:58:14, with 1090588 points after 183893 turns and 2d+6:19:33. 21:49:35 well, far from 0 21:50:39 it's too bad those stats did get reset because we could have saved every game that's ever been watched without breaking a sweat 21:52:13 <10P​leasingFungus> 😦 21:52:54 <06a​dvil> 11:47:26 <|amethyst> !lg * ntv>0 11:47:32 34178. nodergak the Skirmisher (L4 HaCK of Xom), mangled by an ice beast in Sewer (sewer_co_tree) on 2020-05-20 00:41:13, with 86 points after 560 turns and 0:05:10. (from ##crawl-dev-20200522.lg) 21:53:00 <06a​dvil> so your 20k estimate was more correct 21:53:06 <10P​leasingFungus> ooh, nice idea 21:53:41 <06a​dvil> just had to remember that the sequell keyword was ntv to find the right log