Perl 6 - the future is here, just unevenly distributed

IRC log for #metacpan, 2015-07-14

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

All times shown according to UTC.

Time Nick Message
04:43 theory joined #metacpan
05:51 punter joined #metacpan
06:15 melo joined #metacpan
06:35 theory joined #metacpan
06:36 oiami joined #metacpan
06:45 oiami1 joined #metacpan
06:49 zostay joined #metacpan
08:52 bin_005 joined #metacpan
09:02 bin_005 joined #metacpan
11:01 neilb joined #metacpan
11:24 ksurent joined #metacpan
11:33 haarg https://github.com/haarg/metacpan-web/commit/43586bee0f8e6913a852cd805f7b2c5b070fe007  convert api requests from AnyEvent to IO::Async.  not a priority, but something i'd like to pursue in the future.
11:34 haarg might try to write a IO::Async curl library to go along with it
11:41 ilmari bah, when did github switch to using side-by-side diffs by default?
11:54 ilmari haarg: why not Future->needs_all( … )->get instead of map { $_->get } ( … ) ?
11:54 haarg that's probably more sensible, but the map was more similar to the existing code
12:18 neilb ranguard: I noticed you’re using panopta. we’ve tried a number of these services, and the problem is always false positives. How are panopta in that regard, and in general?
12:31 ranguard neilb: personally I don't like the setup for panopta (doesn't seem intuitative), oalders does it all
12:32 ranguard we also use pingdom for external monitoring
12:33 ranguard but then I don't really like pingdom's interface either, though it's getting a little better
12:33 neilb We’ve been using pingdom. They seem the most reliable of the ones we use, but we get false positives. Yeah, pingdom’s UI is a lot better than it used to be :-)
12:33 neilb we’re also trying layer cake, but their data seems pretty crap.
12:34 neilb no, I meant status cake, not layer cake :-)
12:34 ranguard also depends what you are actualy after? - uptime / traffic monitoring, and/or system monitoring?
12:34 neilb system monitoring
12:36 neilb status cake provides good enough looking graphs etc, but report lots of ghost outages, and their reporting on response times varies wildly. They’re “looking into it”
12:36 ranguard we're using munin at work, just implimenting collectd, rsyslog and ELK, but that will be for live diagnostics more than anything else
12:37 neilb how long has metacpan been using panopta?
12:37 ranguard not what you are after, but I came across http://www.cedexis.com/radar/ recently, never considered that some people use a system on top of their CDN's to dynamically switch CDN's!
12:38 ranguard https://github.com/CPAN-API/metacpan-puppet/commits/a9a47b9fd0c6552e4e1e333f1782f3e8d2e41f6c/modules/panopta/manifests/init.pp suggests about a year
12:38 * ranguard did the final switch to Fastly for $work last week (having used them for a couple of years for page assets)
12:44 neilb yeah, we’re a bit small for that stuff. For now :-)
12:44 neilb So oalders, what prompted you to use panopta? How have you found it, particularly on false positives?
12:50 salva joined #metacpan
12:57 oalders neilb: i was trying to get pingdom to sponsor us and panopta reached out to me, so we tried them
12:58 oalders i haven't seen a problem with false positives
12:58 oalders panopta has some code that runs locally on your machine, so you can track lots of things that pingdom can't
12:59 neilb panopta claim that as a benefit, and their approach sounds sensible. But we all know what bollocks marketing people sometimes put on web sites :-)
12:59 neilb yeah, I liked that feature too
12:59 oalders the panopta support that i've gotten has been rock solid and fast.  you can also implement status pages now
12:59 oalders neilb: http://status.panopta.com/metacpan
12:59 oalders not so pretty, but helpful
13:00 neilb yeah, it’s a bit rough looking :-/
13:00 neilb (ie looks like something I might create)
13:01 neilb ok, looks like we’ll try them out. ranguard++ oalders++
13:01 oalders and they don't open source it either ;)
13:02 oalders neilb: be sure to tell them that metacpan sent you :)
13:03 neilb :-)
13:21 vroom joined #metacpan
13:22 _dolmen_ joined #metacpan
13:26 neilb Just noticed that when you search for ‘DBI’, you get ‘DBIML’ listed at the top of the page, like it’s a regular user. Looks like you’re importing all of 00whois.xml, rather than just the users. Is that intentional?
13:32 oalders hmmm https://metacpan.org/author/DBIML
13:34 oalders i guess we are. i don't know that that's intentional
13:35 oalders is there any value in having the mailing lists in there?
13:35 oalders ie indexing them?
13:35 haarg i didn't even know that was a thing
13:35 oalders likewise
13:36 haarg i can see the value of having them somewhere, but not the normal author pages
13:37 oalders neilb: would be worth opening a ticket on cpan-api
13:37 neilb ok, will do
13:37 neilb there are two types of entry in 00whois.xml: PAUSE users and “official” mailing lists
13:38 neilb Always thought that was a bit odd, but no doubt there are hysterical raisins
13:46 neilb done: https://github.com/CPAN-API/cpan-api/issues/417
13:48 oalders neilb++
14:42 ksurent joined #metacpan
16:01 ksurent joined #metacpan
16:24 theory joined #metacpan
16:41 theory joined #metacpan
16:50 vroom1 joined #metacpan
17:03 punter joined #metacpan
17:10 metacpan joined #metacpan
17:10 metacpan [metacpan-web] haarg created haarg/localstorage-shim (+1 new commit): http://git.io/vmsTK
17:10 metacpan metacpan-web/haarg/localstorage-shim 61b98fc Graham Knop: work around Firefox bug where localStorage can throw an exception...
17:10 metacpan left #metacpan
17:18 theory joined #metacpan
17:25 ether yeah, wtf are lists doing in 00whois, and how did they get in there? :)
17:57 vroom joined #metacpan
18:06 y1mmm joined #metacpan
18:15 neilb joined #metacpan
18:21 Ptolemarch joined #metacpan
18:23 Ptolemarch Leo sent me an email on the 6th that I've only just now seen, saying that he can't ssh into lw-mc-01.metacpan.org, one of the hosts hosted by LiquidWeb.
18:23 Ptolemarch Is this problem still ongoing?
19:02 bin_005 joined #metacpan
19:33 ranguard Ptolemarch: yep, left details in PM, we have capacity now that it's not been urgent :)
19:36 Ptolemarch okay.
19:47 punter joined #metacpan
20:42 melo joined #metacpan
20:50 vroom joined #metacpan
21:18 vroom joined #metacpan
21:27 margeas joined #metacpan
22:28 theory joined #metacpan
22:36 Ptolemarch joined #metacpan
22:53 neilb joined #metacpan
23:50 vroom joined #metacpan

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