Perl 6 - the future is here, just unevenly distributed

IRC log for #metacpan, 2013-12-30

| Channels | #metacpan index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:18 priodev joined #metacpan
00:19 klapperl joined #metacpan
00:19 sivoais_ joined #metacpan
02:13 klapperl_ joined #metacpan
04:25 castaway joined #metacpan
04:45 preflex_ joined #metacpan
09:02 celogeek joined #metacpan
10:02 [Sno]_ joined #metacpan
11:06 dolmen joined #metacpan
13:57 ribasushi girls and boys - it's been a month: http://paste.scsys.co.uk/288365
13:57 dipsy [ magnet_web paste from Someone at 217.168.150.38... ]
13:57 ribasushi how hard is it to remove a DNS record?
13:58 ribasushi fwiw this makes cpanm --dev useless on an IPV6-enabld box
14:11 leont http://ip6.nl/#!metacpan.org
14:11 dipsy [ IPv6 domain readiness tester ]
14:11 ribasushi leont: not metacpan
14:11 ribasushi api.metacpan (different box)
14:15 haarg ribasushi: i don't get any slowdown using ipv6
14:16 ribasushi haarg: with `cpanm --dev` ?
14:20 ribasushi haarg: let me know, it may be something wrong with the host I tested from
14:20 haarg --dev works find, and the nmap just you showed gives the same times for ipv4 and ipv6
14:21 ribasushi this is so odd... let me try from somewhere else
14:23 ribasushi haarg: you are telling me that `nmap -p 80 -P0 -T5 -6 2001:41c8:10:1044:5054:ff:fe77:1b9d` gives you 'open' ?
14:23 ribasushi it gives me filtered from both the uk and au
14:23 ribasushi (and from travis at least 2 weeks ago)
14:24 haarg ribasushi: oh yeah sorry.  i get filtered.
14:24 ribasushi haarg: right which in the case of `cpanm --dev ...` results in systemic timeouts
14:24 rwstauner rafl: ^^
14:25 ribasushi furthermore metacpan.org and api.metacpan.org return the same A record
14:25 ribasushi which is even more bizarre
14:25 ribasushi and api. is a cname to somewhere "else", whereas metacpan is a record on its own
14:26 rwstauner i know ranguard has been moving some things to fastly
14:27 rwstauner and something in there was goofy, but i don't recall the details
14:28 ribasushi rwstauner: http://paste.scsys.co.uk/288368 as easy as this
14:28 dipsy [ magnet_web paste from Someone at 217.168.150.38... ]
14:29 ribasushi this cname pointing to a random box can't be right (and the A of bm-n2.metacpan.org actually matches the A of metacpan.org)
14:30 rwstauner hmm
14:32 rwstauner i haven't done anything with the dns for metacpan.  oalders?
15:00 ranguard bm-n2 is THE server (we only have one)
15:01 ranguard previously api worked becasue we have a wildcard record for both ipv4 and ipv6
15:02 ranguard in trying to switch to a CDN we added a record for api - which we then cnamed to bm-n2 when that didn't work (Es client not RFC compliant)
15:02 ranguard at the moment only cpan.metacpan.org is going via fastly - which is ipv4 only (at the moment, they are working on ipv6)
15:04 ribasushi ranguard: but the ipv6 port 80 on bm-n2 is *firewalled away*
15:05 ranguard there was also an issue rafl was looking at sorting (bm-n2 is a VM running on bm-n1 with port forwarding or something)
15:06 ranguard sorry, have to run, will try chase this evening
15:06 ribasushi ranguard: yes, and this issue has been going on for over a month, so temporarily killing the AAAA record seems like a sane solution
15:06 ranguard ribasushi: which ticket it is again? - so I can replicate
15:07 ranguard back in about 4 hours
15:07 ribasushi ranguard: I didn't add a ticket, assumed it's faster to fix it than to close said ticket
15:07 ribasushi filing...
15:13 ribasushi ranguard: https://github.com/CPAN-API/metacpan-web/issues/1029
15:13 dipsy [ [IMPORTANT] api.metacpan.com broken over IPV6 · Issue #1029 · CPAN-API/metacpan-web · GitHub ]
15:19 oalders is somebody currently logged in to the DNS management? i'm getting a message about an existing  session. ranguard?
15:19 Topic for #metacpan is now Have you installed your MetaCPAN VM? https://github.com/CPAN-API/metacpan-developer | Chat logs available at http://irclog.perlgeek.de/metacpan/ | Can't find your module on MetaCPAN? https://metacpan.org/about/missing_modules | Feature suggestions now belong on the wish list: https://github.com/CPAN-API/cpan-api/wiki/Wishlist
15:22 oalders i guess rafl hasn't had a chance to look at the DNS, so we can just kill the AAAA record in the meantime
15:22 ribasushi oalders++
15:23 oalders i just need to be able to log in to the console to do that :)
15:24 ribasushi .oO( some APNIC users ought to be pissed by this but... THE WESTERN 1ST WORLD PROBLEMS ARE MOAR IMPORTANT :D
15:24 ribasushi actually I take this back - it doesn't currently work for them *anyway*
15:33 oalders :)
15:39 oalders ribasushi: record deleted
15:40 ribasushi oalders++ # sorry to be a pain about this but... yeah ;)
15:40 oalders ribasushi++ # following up :)
16:00 ranguard well, ok, so that wasn't 4 hours - $child watching postman pat for a bit :)
16:00 ranguard thanks for sorting out - I'll close the ticket
16:00 ranguard (useful for furture reference)

| Channels | #metacpan index | Today | | Search | Google Search | Plain-Text | summary