Log for #openttdcoop.devzone on 12th September 2014:
Times are UTC Toggle Colours
00:00:25  *** Supercheese has quit IRC
03:48:01  *** Supercheese has joined #openttdcoop.devzone
07:01:59  *** LSky` has joined #openttdcoop.devzone
09:30:41  *** tycoondemon has quit IRC
11:32:21  *** Supercheese is now known as Guest543
11:32:26  *** Supercheese has joined #openttdcoop.devzone
11:38:29  *** Guest543 has quit IRC
12:50:45  *** tycoondemon has joined #openttdcoop.devzone
14:14:31  *** mrruben5 has joined #openttdcoop.devzone
15:27:26  *** gelignite has joined #openttdcoop.devzone
15:31:21  *** ODM has joined #openttdcoop.devzone
15:33:09  *** Alberth has joined #openttdcoop.devzone
16:21:42  *** oskari89 has joined #openttdcoop.devzone
17:40:37  *** frosch123 has joined #openttdcoop.devzone
17:41:07  *** andythenorth has joined #openttdcoop.devzone
17:42:15  <frosch123> hmm, i guess we need to change something about eints, so it survives a database restart
17:45:54  <planetmaker> nah, itwas fiddled with
17:46:07  <frosch123> no, it's still offline
17:46:19  <frosch123> if the database connection is closed once, it stays closed
17:46:22  <planetmaker> ^Spike^, ?
17:46:23  <frosch123> until eints restart
17:48:14  <Alberth> silly data base :p
17:48:27  <Rubidium> isn't there some 'reopen db connection if needed' wrapper around a connection?
17:48:58  <Alberth> I am wondering why it doesn't bail out with "failed to connect"
17:51:48  <frosch123> Alberth: db_connection is only set once in init()
17:52:06  <frosch123> if the database closes the connection, eints will never open it a second time
17:52:36  <frosch123> i think we need to change get_authentication to check whether the connection is open, and make it try to reopen it
17:52:54  <Alberth> yeah, but usually you get "broken pipe" or so when trying to write to a closed connection
17:53:01  <Alberth> apparently not here
17:53:05  <frosch123> sure
17:53:24  <frosch123> first error is "connecton was closed by admin", next errors are "connection already closed"
17:53:25  <Alberth> hmm, maybe it catches the error?
17:54:04  <frosch123> uhm, yes, eints does not exit :p
17:54:22  <planetmaker> hm :D
17:55:20  <planetmaker> Alberth, frosch123, we've been considering to change our authentication... how much effort would it be to connect to another database, or use some ldap thing to authenticate?
17:55:41  <frosch123>
17:55:44  <Alberth> inifinite!
17:55:46  <Alberth> :D
17:55:52  <planetmaker> not saying we will, but maybe that's needed when we want to integrate rhodecode + jenkins + redmine
17:56:01  <Alberth> nah, should be possible
17:56:14  <frosch123> i am sure there is a python ldap module :)
17:56:22  <frosch123> so it should be easier than the old custom redmine method
17:56:30  <Alberth> or spike will write one :)
17:56:35  <^Spike^> planetmaker db is up but you already came to that conclusion :D
17:56:35  <planetmaker> :D
17:56:49  <^Spike^> i need to perform my maintenance at some point :)
17:56:55  <^Spike^> and for some reason during the day is safest :D
17:57:05  <planetmaker> incredible :)
17:57:19  <Alberth> all children to school, safe to play :)
17:57:23  <^Spike^> :)
17:57:27  <planetmaker> hehe :)
17:57:35  <^Spike^> i hope to get more info out of it to see if we need pgsql optimization or not :)
18:00:24  <^Spike^> but as with most dbs...
18:00:27  <^Spike^> it takes time... :)
18:00:50  <frosch123> you are not talking about eints, are you?
18:00:54  <^Spike^> nop
18:00:56  <^Spike^> in general
18:00:57  <planetmaker> yeah, I also just meant to warn frosch and Alberth that there *might* be some change ahead
18:01:10  <^Spike^> i try to optimize mysql/pgsql as much as possible with hw/virtualization/sw :)
18:01:31  <^Spike^> it's the same with our webserver
18:01:36  <^Spike^> which part of the stuff runs on now
18:01:39  <^Spike^> (like paste for ex)
18:01:47  <^Spike^> i try to get the most out of it even if it's just php :)
18:02:03  <^Spike^> if it improves response time which leads to better user experience
18:02:21  <^Spike^> but it's stuff that takes time
18:09:03  <^Spike^> just so many changes atm in such a short period :D
18:10:00  <Alberth> anyone figured out an easy way to save a paste text?
18:10:10  <^Spike^> what you mean?
18:10:32  <Alberth> clicking "raw" and then "save as" doesn't exactly count as easy, imho
18:10:57  <^Spike^> can also just copy-pasta it... or add /raw to the url
18:10:58  <planetmaker> that was the same with the old paste, wasn't it?
18:11:13  <^Spike^>
18:11:21  <^Spike^> for ex
18:12:10  <Alberth> copy/paste implies I need to open an editor first
18:12:11  <frosch123> Alberth: right click + save as ?
18:12:31  <Alberth> I get an .html save
18:12:51  <planetmaker> for the raw?
18:12:57  <frosch123> er,, i mean "save link as" on "raw"
18:13:05  <^Spike^> i get the same with the old paste unless i pick the /raw version
18:13:18  <frosch123> anyway, you can also do a wget -O - bla/raw | patch -p 1 :)
18:13:50  <Alberth> right-click on"raw" button works
18:14:02  <Alberth> filename is a bit sub-optimal :p
18:38:53  <Alberth> calling init() from an exception doesn't look like a problem
18:39:11  <Alberth> except the init may fail too
19:17:36  *** mrruben5 has quit IRC
20:11:52  *** andythenorth has quit IRC
20:19:28  *** ODM has quit IRC
20:33:10  *** erlehmann has joined #openttdcoop.devzone
20:45:19  <Alberth> frosch123:  how does this look?
20:50:38  <frosch123> it could print some log output when retrying stuff
20:54:17  <frosch123> the _do_command functions is somewhat psql specific now, but i guess we do not care about other dbs
20:54:35  <frosch123> looks fine
20:55:11  <Alberth> good points
20:55:34  <Alberth> will fix them tomorrow
20:57:17  <Alberth> although fixing _do_command completely will be tmwftlb
20:57:43  <frosch123> well, last time we checked there were no py3 modules for mysql
20:57:53  <frosch123> so, it is dead code anyway
21:00:37  <Alberth> given the widespread use of that db, I doubt it no new modules will pop up :)
21:00:53  <Alberth> but even then, we are not using it, so who cares :)
21:01:26  <Alberth> gn
21:02:10  <planetmaker> looks like there meanwhile exists such connector...
21:02:18  <planetmaker> but well... not needed right now :)
21:02:29  <planetmaker> g'ngith Alberth
21:02:44  *** Alberth has left #openttdcoop.devzone
21:17:07  *** frosch123 has quit IRC
21:42:02  *** erlehmann has quit IRC
21:42:42  *** erlehmann has joined #openttdcoop.devzone
22:20:44  *** gelignite has quit IRC
22:24:24  *** oskari89 has quit IRC
22:50:33  *** LSky` has quit IRC

Powered by YARRSTE version: svn-trunk