Perl 6 - the future is here, just unevenly distributed

IRC log for #metacpan, 2016-07-07

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

All times shown according to UTC.

Time Nick Message
03:28 oiami joined #metacpan
06:13 [Tux] mickey, does this mean it is *NOT* my fault? (and I will not be blocked by ranguard)
06:43 * jberger blames [Tux] anyway (no idea what for :p)
06:44 jberger [Tux]: o/
06:44 [Tux] and you are probably right
06:44 * [Tux] feels like the canary in the coalmine here, just like in the perl6 world :P
06:45 jberger How're you breathing?
06:45 [Tux] not
06:45 jberger :/
06:46 jberger Dare I ask?
06:46 [Tux] ask away
06:46 jberger What is troubling the canary at the moment?
06:46 [Tux] backlog: <ranguard> it's 80.100.130.165 that's calling the DELETE
06:47 [Tux] https://gist.github.com/ranguard/9edfe206c6e54964b451b016e4e36884
06:47 [Tux] I was dared to use MetaCPAN::Client to the max
06:47 [Tux] and I have already encountered several problems that were picked up here
06:47 neilb joined #metacpan
06:48 [Tux] so far I have found no time to come up with actual pull-requests, but I'm not really in any progress either
06:48 [Tux] the API isn't DWIM in all places
06:50 jberger Ok the scrollback makes more sense on a reread
06:50 [Tux] though I must say that the API for MetaCPAN::Client is WAY more intuitive than the API for Net::GitHub and Pithub
06:51 jberger I've never tried, I've usually just hit ES via HTTP
06:51 [Tux] ES?
06:51 jberger Elasticsearch
06:51 [Tux] ah, elasticsearch
06:52 [Tux] as said: I was challenged to use M::C\
06:52 jberger Oh sure, I'm not saying I'm doing the right thing
06:52 jberger Carry on!
06:52 [Tux] and I am really impressed by the work mickey and the gang did
06:53 jberger mickey++
06:53 jberger gang++
06:53 jberger [Tux]++ while I'm at it, qa is important too
06:54 [Tux] thanks
07:35 punter joined #metacpan
07:53 ranguard [Tux]++ # having someone scream about the issues means we can get them fixed
08:07 neilb joined #metacpan
08:10 Relequestual joined #metacpan
08:20 mickey ranguard: whatever we do, we need to keep the DELETEs to /_search/scroll ... but will be should set all the data indexes to readonly
08:20 mickey s/should/wise to/; # wtf am i saying
08:47 ranguard mickey: ok, so I can limit DELETE to only allow to that path - but wouldn't ES just timeout the scroll ?
08:47 ranguard seems odd that it requires the client to cancel it?
08:48 ranguard also at the moment all of those requests are returning a 500
08:48 mickey it will, but that's (guessing) for improving performance by releasing the resource
08:49 mickey the 500s need to be investigated... we need to log the server error somewhere
08:49 ranguard there is a something in starman_error.log - but it's not exactly useful!
08:50 mickey we don't have the ES console logged?
08:50 ranguard I think slow queries are, but not every query (would impact performance)
08:51 mickey we don't need the queries... we need errors
08:52 mickey there must be an ES config to allow that
09:22 Tux guys, do I need to change anything? (just asking)
09:35 punter joined #metacpan
09:42 mickey Tux: i don't think so... it looks like server-side issue to me
15:49 neilb joined #metacpan
15:52 neilb_ joined #metacpan
16:16 Relequestual joined #metacpan
17:21 neilb_ joined #metacpan
18:33 punter joined #metacpan
21:23 _dolmen_ joined #metacpan
21:28 alh__ joined #metacpan

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