Perl 6 - the future is here, just unevenly distributed

IRC log for #metacpan, 2016-08-31

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

All times shown according to UTC.

Time Nick Message
03:40 osfameron joined #metacpan
03:40 mst joined #metacpan
05:49 [Tux] Date: Tue, 30 Aug 2016 22:52:06 +0200 (CEST)
05:49 [Tux] [Cxn] ** [https://fastapi.metacpan.org:443]-[599] Could not connect to 'fastapi.metacpan.org:443': No route to host, called from sub Search::Elasticsearch::Role::Client::Direct::Main::scroll_helper at /pro/lib/perl5/site_perl/5.22.0/MetaCPAN/Client/Request.pm line 147. With vars: {'request' => {'method' => 'POST','mime_type' => 'application/json','serialize' => 'std','ignore' => [],'qs' => {'scroll' => '5m','size' => 1000,'search_type' => 'scan'},'path'
05:49 [Tux] => '/v1/favorite/_search','body' => {'query' => {'term' => {'distribution' => 'Text-CSV_XS'}}}},'status_code' => 599}
06:24 neilb joined #metacpan
06:58 oiami joined #metacpan
07:08 Lee joined #metacpan
07:59 osfabibisi joined #metacpan
08:21 neilb joined #metacpan
09:01 uree joined #metacpan
10:01 metacpan joined #metacpan
10:01 metacpan [metacpan-web] mickeyn created mickey/fix_file_categories (+1 new commit): https://git.io/viTLw
10:01 metacpan metacpan-web/mickey/fix_file_categories 09e2895 Mickey Nasriachi: controller/release: correct categories logic...
10:01 metacpan left #metacpan
10:11 ranguard I got 2 fails overnight - one a timeout of fastly talking to us, not sure about the other - so have replicated at least some error :)
10:15 dolmen joined #metacpan
10:49 mickey ranguard: https://github.com/metacpan/metacpan-web/pull/1768
10:49 mickey will solve the release page categories issue
12:51 ranguard mickey: stuck in middle of adventure playground so can't review atm :)
12:51 ranguard will look tonight
13:02 mickey cool, enjoy
13:21 * neilb had image of quicksand in the middle of a playground, with ranguard stuck in it. Fireswamp For Kids™
13:39 dolmen http://search.cpan.org/~autolife/ vs https://metacpan.org/author/AUTOLIFE : AUTOLIFE/Mojo-UserAgent_t2.tar.gz (which has a serious packaging problem) does not appear on MetaCPAN
13:51 edward joined #metacpan
14:02 mickey dolmen: that tarball is missing any meta file
14:03 mickey so i'm not surprised it's not indexed properl
14:03 mickey y
14:04 jberger pro(perl)y!
14:04 jberger zomg!
14:29 mickey :)
14:37 ranguard mickey: did you merge anything in the last 2 hours?
14:37 ranguard or play on lw-03 ?
14:38 ranguard we got a string of timeouts from fastly
14:40 mickey ranguard: no... just tried to run script release for one file
14:40 mickey don't think that's related
14:41 ranguard mickey: ok, cheers, I'll go look at some logs :)
14:53 mickey nice, let me know if you need anything
14:54 ranguard happen to have something I can stream logs to to anaylsys :) ?
14:55 mickey hmm... reyjrar has something that uses ES :)
14:55 mickey reyjrar: ^^
14:56 ranguard probably logstash->kibana, I guess I need to do that at some point in anycase
14:56 ranguard was just trying to avoid it :)
15:21 ranguard I've added basic pingdom (using my work account) monitoring of lw-mc-03's nginx - to help diagnose issues
15:23 ranguard any objections to me converting BM-mc-03 so it's NOT on legacy but on master branch (e.g. the new and shiney) :) ?
15:27 ranguard mickey: generally I like lots of comments, especially when there aren't any tests for a bug that you are fixing
15:28 ranguard but I'm happy to merge #1768 if you like?
15:40 oalders mickey: not sure what's going on with that test folder
15:55 haarg dolmen: metacpan is interpreting it as the dist "Mojo", version "UserAgent_t2" (which converts to 0)
15:56 mickey oalders: it seems like the github issues config for the tests is just missing
15:57 mickey ranguard: what are we talking about? :)
15:57 * mickey came back to the desk and got confused by too many messages :)
15:58 mickey ranguard: please do merge it if you have no concerns :)
15:59 * mickey is heading home
15:59 ranguard mickey: ok :)
15:59 ranguard merged
15:59 metacpan joined #metacpan
15:59 metacpan [metacpan-web] ranguard pushed 1 new commit to master: https://git.io/vikJU
15:59 metacpan metacpan-web/master 0f45d96 Leo Lapworth: Merge pull request #1768 from metacpan/mickey/fix_file_categories...
15:59 metacpan left #metacpan
15:59 metacpan joined #metacpan
15:59 metacpan [metacpan-web] ranguard deleted mickey/fix_file_categories at 09e2895: https://git.io/vikJT
15:59 metacpan left #metacpan
16:48 reyjrar ranguard: I have a perl syslog => es thing I'm using
16:55 mickey ranguard++
17:07 jberger reyjrar: we just set one of those up too
17:53 batman joined #metacpan
18:12 ranguard reyjrar / jberger - this is from an external source that can send rsyslog
18:12 ranguard work has fastly -> rsyslog  -> logstash -> es
18:13 ranguard if anyone has something simpler I can just copy and go with I'd love to use it :)
18:13 bpmedley joined #metacpan
18:14 ranguard this isn't for 100% reliability or anything
18:14 Grinnz_ i was wondering for a while why you're logging to español in particular
18:15 ranguard Grinnz_: slack gets confused when you see foo::es::search...
18:15 ranguard you end up with foo: <spanish flag> :search  :)
18:16 Grinnz_ hahah
18:16 ranguard totally confused use at work the other day for a moment :)
19:01 metacpan joined #metacpan
19:01 metacpan [metacpan-api] ranguard deleted leo/download_url at b67d768: https://git.io/vikwB
19:01 metacpan left #metacpan
19:02 neilb joined #metacpan
20:00 neilb joined #metacpan
20:03 y1mmm joined #metacpan
21:02 reyjrar ranguard: I haven't put my stuff on GitHub yet, but it'll be there soon.. Moose + POE.. Rather easily pluggable to take a feed from rsyslog in a number of ways..
21:03 reyjrar I'm using it on a TCP port with an rsyslog destination set to queue to disk if it goes down
21:03 mst kicking it old school
21:03 neilb joined #metacpan
21:03 reyjrar mst: indeed, get off my lawn!
21:03 reyjrar ;)
21:03 * mst mostly does Moo+IO::Async for this now, but OTOH POE stuff totally still works so *cool*
21:04 mst (I can still debug POE fine, I just can't be bothered writing it from scratch anymore ;)
21:05 reyjrar mst: good recommendation.. you reminded me I was supposed to look at IO::Async a long time ago and it fell off my todo list
21:05 reyjrar I skipped AnyEvent for uhm, personal reasons? :)
21:05 * mst also
21:06 mst I like Future and Future::Utils too much to give up
21:06 mst must release my MojoX::FutureDo thing at some point
21:06 mst since IO::Async and Mojo co-operate great
21:06 mst having futures for both makes me happy
21:06 reyjrar I think I know what I'm doing tonight! :)
21:08 reyjrar ranguard: This is an example of the raw document generated by parsing: https://gist.github.com/reyjrar/9b01e599332718332566
21:12 reyjrar actually, that's old, this is the new version: https://gist.github.com/reyjrar/9b01e599332718332566#file-updated
21:13 reyjrar The idea is to have a predefined dictionary of field names to be indexed/analyzed by elasticsearch.. then the rest of the document can be stored, but violating the dictionary won't kill elasticsearch
21:14 reyjrar In this case the "context" subsection is the predefined dictionary and the rest would be stored as type:object/enabled:false
23:07 atta joined #metacpan
23:08 neilb oalders: I just did a first rough draft of the prospectus. Have a look later and hack away / add comments. I’ll do another iteration tomorrow night.

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