ruote tmp/log_2011-07-25.html

2011-07-25 00:58:12 utc jmettraux kellyp: in JSON, there are no "_" prefix
2011-07-25 00:58:25 utc jmettraux kennethkalmer: have a good night !
2011-07-25 00:58:55 utc kellyp jmettraux: when i tried with no _ it doesn't do anything
2011-07-25 00:59:02 utc jmettraux it's only used in Ruby to avoid collisions with Ruby keywords, behind the scenes, '_redo' becomes 'redo'
2011-07-25 00:59:13 utc jmettraux give me more context then
2011-07-25 00:59:25 utc kellyp could this be it's in the on_error handler?
2011-07-25 01:01:03 utc kellyp i see in the log that the on_error and irc participants are run
2011-07-25 01:01:12 utc kellyp but the redo seems to be skipped over
2011-07-25 01:02:14 utc jmettraux ah, it doesn't know what you're referring to then
2011-07-25 01:02:56 utc jmettraux you might want to use a :ref or to rethink your logic, it seems convoluted
2011-07-25 01:03:24 utc jmettraux http://ruote.rubyforge.org/exp/redo.html
2011-07-25 01:03:56 utc kellyp here is a more simple example https://gist.github.com/1103335
2011-07-25 01:04:35 utc kellyp yeh, it's a little messy it's being generated from stuff in our db
2011-07-25 01:05:37 utc jmettraux so, what you're saying is that there is an error in the participant test, but the redo is ineffective ?
2011-07-25 01:05:45 utc jmettraux what is the observed behaviour ?
2011-07-25 01:07:01 utc jmettraux what version of ruote are you using ?
2011-07-25 01:07:30 utc kellyp i believe it's the lastest from master
2011-07-25 01:08:00 utc kellyp here is the log from the dsl version https://gist.github.com/1103339
2011-07-25 01:09:42 utc jmettraux there is a shortcut for that construct : on_error => 'redo'
2011-07-25 01:10:06 utc jmettraux now I will open an issue for that and attempt to reproduce it and fix ti
2011-07-25 01:10:07 utc jmettraux it
2011-07-25 01:13:53 utc jmettraux https://github.com/jmettraux/ruote/issues/32
2011-07-25 01:14:04 utc kellyp the on_error => 'redo' worked, went into an endless loop
2011-07-25 01:15:44 utc kellyp thanks for your help!
2011-07-25 01:15:53 utc jmettraux thanks for the issue report !
2011-07-25 01:29:33 utc kellyp jmettraux: it seems that rewind works for what i need https://gist.github.com/1103364
2011-07-25 01:31:01 utc jmettraux ah nice
2011-07-25 01:32:26 utc jmettraux this should be equivalent : https://gist.github.com/1103370
2011-07-25 01:33:55 utc kellyp ah, thanks
2011-07-25 02:32:16 utc kamalfariz good morning! is it possible to reuse pdefs in another pdef?
2011-07-25 02:35:04 utc jmettraux kamalfariz: good morning ! yes
2011-07-25 02:35:43 utc jmettraux http://ruote.rubyforge.org/exp/subprocess.html (scroll for "URI" or "bound at engine level"
2011-07-25 02:35:48 utc jmettraux )
2011-07-25 23:02:49 utc kennethkalmer jmettraux: morning !
2011-07-25 23:03:07 utc jmettraux kennethkalmer: hey, good evening !
2011-07-25 23:03:21 utc kennethkalmer doing well ?
2011-07-25 23:03:39 utc jmettraux yes, doing alright, and you ?
2011-07-25 23:03:56 utc kennethkalmer doing ok
2011-07-25 23:04:09 utc kennethkalmer fighting 1.9.2 slowness (on boot) and Rails 3 upgrade
2011-07-25 23:04:43 utc kennethkalmer got a slightly fast 1.9.2 with www.rubyinside.com/get-edge-ruby-faster-loading-ruby-1-9-2-now-4973.html
2011-07-25 23:05:10 utc kennethkalmer gonna try https://github.com/skaes/rvm-patchsets in the morning
2011-07-25 23:06:07 utc kennethkalmer before I head to bed
2011-07-25 23:06:18 utc kennethkalmer there is an issue with ruote-redis we're trying to identify
2011-07-25 23:06:27 utc kennethkalmer seems we've got a deadlock of sorts
2011-07-25 23:06:35 utc kennethkalmer new processes are simply not starting
2011-07-25 23:06:43 utc kennethkalmer the messages aren't even written to redis
2011-07-25 23:07:24 utc kennethkalmer I even built a custom 2.2.0 from 8e19e6018dd62a10876f2197a32d6e27c4e7ff59
2011-07-25 23:07:39 utc kennethkalmer still no luck
2011-07-25 23:08:29 utc kennethkalmer anycase, I'll try and create a test for it in ruote-redis
2011-07-25 23:08:52 utc jmettraux strange
2011-07-25 23:08:52 utc kennethkalmer injecting a lock with the same characteristics as what I'm seeing and getting it to pass
2011-07-25 23:09:13 utc kennethkalmer 1 thing might be to blame, and I just realized it
2011-07-25 23:09:34 utc kennethkalmer a week or two ago we changed the server time to 1 Aug to run some "first of the month" simulations
2011-07-25 23:09:38 utc kennethkalmer then changed it back
2011-07-25 23:09:44 utc kennethkalmer wonder if the lock doesn't have a future date
2011-07-25 23:16:06 utc jmettraux should be easy to look at
2011-07-25 23:16:50 utc kennethkalmer i think that is exactly the issue, looking at SETNX example and lib/ruote/redis/storage.rb:325
2011-07-25 23:17:56 utc jmettraux (sorry, I'm in a meeting right now, giving you sparse attention)
2011-07-25 23:18:16 utc kennethkalmer no worries, i'm tired and blathering :)
2011-07-25 23:18:44 utc jmettraux old debug trick of explaining to other person :)
2011-07-25 23:18:54 utc kennethkalmer +1
2011-07-25 23:20:24 utc kennethkalmer i'll purge the storage in the morning and retry, should work
2011-07-25 23:20:32 utc kennethkalmer have an awesome day, ttyl !
2011-07-25 23:22:25 utc jmettraux ok, ciao !