chat.freenode.net #tryton log beginning Wed May 14 00:00:01 CEST 2014 | ||
2014-05-14 01:03 -!- cedk(~ced@23.33-67-87.adsl-dyn.isp.belgacom.be) has joined #tryton | ||
2014-05-14 01:03 -!- cedk(~ced@gentoo/developer/cedk) has joined #tryton | ||
2014-05-14 02:01 -!- digitalsatori(~Thunderbi@116.234.181.110) has joined #tryton | ||
2014-05-14 06:23 -!- asoleh(~asoleh@36.70.82.7) has joined #tryton | ||
2014-05-14 06:44 -!- frispete(~frispete@p54A93A45.dip0.t-ipconnect.de) has joined #tryton | ||
2014-05-14 06:55 -!- sharoonthomas(~sharoonth@118.91.181.250) has joined #tryton | ||
2014-05-14 07:02 -!- yangoon1(~mathiasb@p549F0ECF.dip0.t-ipconnect.de) has joined #tryton | ||
2014-05-14 07:30 -!- sharoonthomas(~sharoonth@122.177.186.220) has joined #tryton | ||
2014-05-14 07:45 -!- pobsteta(~Thunderbi@LPuteaux-656-01-248-71.w217-128.abo.wanadoo.fr) has joined #tryton | ||
2014-05-14 07:48 -!- rpit(~ralf@dslb-178-011-040-234.pools.arcor-ip.net) has joined #tryton | ||
2014-05-14 08:02 -!- sharoonthomas(~sharoonth@122.177.186.220) has joined #tryton | ||
2014-05-14 08:15 -!- sharoonthomas_(~sharoonth@122.177.47.13) has joined #tryton | ||
2014-05-14 08:18 -!- vernichon(~Thunderbi@gex01-1-78-234-55-95.fbx.proxad.net) has joined #tryton | ||
2014-05-14 08:18 -!- asoleh(~asoleh@36.70.82.7) has joined #tryton | ||
2014-05-14 08:27 -!- bechamel(~Adium@62-197-96-189.teledisnet.be) has joined #tryton | ||
2014-05-14 08:35 -!- sharoonthomas(~sharoonth@122.177.47.13) has joined #tryton | ||
2014-05-14 09:09 -!- vcardon(~vcardon@LNeuilly-152-23-15-185.w193-252.abo.wanadoo.fr) has joined #tryton | ||
2014-05-14 09:09 -!- vcardon(~vcardon@LNeuilly-152-23-15-185.w193-252.abo.wanadoo.fr) has left #tryton | ||
2014-05-14 09:13 -!- bechamel(~Adium@cismwks02-virtual1.cism.ucl.ac.be) has joined #tryton | ||
2014-05-14 09:24 -!- sharoonthomas(~sharoonth@118.91.181.250) has joined #tryton | ||
2014-05-14 09:32 -!- nicoe(~nicoe@209.102-247-81.adsl-dyn.isp.belgacom.be) has joined #tryton | ||
2014-05-14 09:50 -!- cedk(~ced@gentoo/developer/cedk) has joined #tryton | ||
2014-05-14 09:59 -!- sharoonthomas(~sharoonth@118.91.181.250) has joined #tryton | ||
2014-05-14 10:05 -!- digitalsatori(~Thunderbi@114.94.153.137) has joined #tryton | ||
2014-05-14 10:11 -!- digitalsatori(~Thunderbi@114.94.153.137) has joined #tryton | ||
2014-05-14 10:24 -!- pokoli(~pokoli@unaffiliated/pokoli) has joined #tryton | ||
2014-05-14 10:48 -!- rpit(~ralf@dslb-178-011-040-234.pools.arcor-ip.net) has joined #tryton | ||
2014-05-14 11:07 -!- asoleh(~asoleh@36.70.82.7) has joined #tryton | ||
2014-05-14 11:10 -!- nicoe(~nicoe@2001:6f8:3aa:0:b699:baff:fe56:2328) has joined #tryton | ||
2014-05-14 11:12 -!- digitalsatori(~Thunderbi@114.94.153.137) has joined #tryton | ||
2014-05-14 11:16 -!- nicoe(~nicoe@2001:6f8:3aa:0:b699:baff:fe56:2328) has joined #tryton | ||
2014-05-14 11:26 -!- digitalsatori(~Thunderbi@114.94.153.137) has joined #tryton | ||
2014-05-14 11:30 -!- digitalsatori(~Thunderbi@114.94.153.137) has joined #tryton | ||
2014-05-14 11:34 -!- corro(~corro@46.22.23.85) has joined #tryton | ||
2014-05-14 11:52 <yangoon1> cedk: I have a question wrt to tests: running python setup.py test on trytond in virtualenv fails with http://pastebin.com/8ga1aqWg, while trytond/tests/run-tests.py just runs fine. | ||
2014-05-14 11:52 <yangoon1> cedk: is there a way to make python setup.py test work? | ||
2014-05-14 11:54 <cedk> yangoon1: you must set DB_NAME | ||
2014-05-14 11:55 <yangoon> cedk: thx, will try | ||
2014-05-14 11:58 <cedk> yangoon: in the environment | ||
2014-05-14 11:58 -!- sharoonthomas(~sharoonth@118.91.181.250) has joined #tryton | ||
2014-05-14 12:09 <yangoon> cedk: is there any possibilty to set db_type via environment? | ||
2014-05-14 12:12 <cedk> yangoon: not yet but it should be part of the refectoring of configuration | ||
2014-05-14 12:13 <cedk> it will be even better when multi-db support will be dropped | ||
2014-05-14 12:18 <yangoon> yes, I know your concerns about multi-db, but that's really killing a great and widely used feature | ||
2014-05-14 12:19 <cedk> yangoon: nobody ever explains a real usage of it | ||
2014-05-14 12:20 <cedk> it just the usual crying about chnages | ||
2014-05-14 12:24 <yangoon> cedk: What would be the alternative? How would be the way to easily launch several databases? Would there be a way to restart the server from the client with a different database? | ||
2014-05-14 12:24 <cedk> yangoon: no | ||
2014-05-14 12:25 <cedk> yangoon: what is your usage??? | ||
2014-05-14 12:25 <cedk> I don't care about losing functionnality that nobody uses | ||
2014-05-14 12:25 <cedk> indeed, it is for me a good motivation to remove them | ||
2014-05-14 12:26 <yangoon> cedk: I am accessing different databases all the time | ||
2014-05-14 12:26 <cedk> yangoon: so what | ||
2014-05-14 12:27 <yangoon> cedk: AFAIS I would have to launch a dedicated server for each one each time I want to access one | ||
2014-05-14 12:28 <cedk> yangoon: please explain your use case? | ||
2014-05-14 12:28 <pokoli> ACTION listening to this nice conversation | ||
2014-05-14 12:29 <yangoon> cedk: having a production and a test database is a simple setup | ||
2014-05-14 12:29 <cedk> yangoon: that's the worst configuration ever | ||
2014-05-14 12:30 <cedk> someone running his tests server on the same one as the production is just a fool | ||
2014-05-14 12:30 <yangoon> cedk: ok, bad example | ||
2014-05-14 12:32 <yangoon> while developing/testing you create a lot of databases, that you want to access immediately without reconfiguration of the server | ||
2014-05-14 12:33 <cedk> yangoon: creating DB will be done per command line and so you have to setup the server | ||
2014-05-14 12:33 <cedk> and indeed, you don't create so often a database when developping because it is the tests that does it for you | ||
2014-05-14 12:34 <yangoon> and how to access those different databases easily? | ||
2014-05-14 12:35 <cedk> you start the server with the right database | ||
2014-05-14 12:36 <pokoli> yangoon: I usually have several trytond servers while developing, because each server has diferent modules | ||
2014-05-14 12:36 <cedk> when developping one feature, you need only one database | ||
2014-05-14 12:37 <cedk> so still no real example of the need of multi-db | ||
2014-05-14 12:38 <pokoli> cedk: you plan to add db_name to config file or only commandline? | ||
2014-05-14 12:40 <cedk> pokoli: probably both but it will be a uri | ||
2014-05-14 12:41 <pokoli> cedk: like sqlalchemy one? | ||
2014-05-14 12:41 <cedk> pokoli: yeps | ||
2014-05-14 12:41 <pokoli> cedk: nice :D | ||
2014-05-14 12:42 <pokoli> cedk: and I will be in favour of having both options | ||
2014-05-14 12:45 -!- asoleh(~asoleh@36.70.82.7) has joined #tryton | ||
2014-05-14 13:11 -!- nicoe(~nicoe@85.26.36.239) has joined #tryton | ||
2014-05-14 13:12 <corro> cedk: our use case for multi db is having a pool of trytond instances, handling requests for several databases | ||
2014-05-14 13:13 <cedk> corro: nothing prevent to have a pool of pool of trytond | ||
2014-05-14 13:13 <cedk> corro: it will even be better because if one database is overloaded it doesn't affect the others | ||
2014-05-14 13:14 <cedk> as I said previously, it is a infrastructure problem and solving at the software level is wrong | ||
2014-05-14 13:14 <cedk> it is wrong but also inefficient | ||
2014-05-14 13:16 <pokoli> corro: currently you have a pool of trytond listening to all databases? | ||
2014-05-14 13:16 <corro> pokoli: yes | ||
2014-05-14 13:17 <pokoli> corro: and how the tryton cache behave? As AFAIU each trytond will have a cache for each database | ||
2014-05-14 13:19 <corro> pokoli: it works fine, but the cache is not shared, that's true. I tried using a memcached server, but my attempts failed | ||
2014-05-14 13:22 <corro> I saw there was this idea of using memcached before (GSoC IIRC), it would definitely be an improvement for large scale installations | ||
2014-05-14 13:22 <cedk> corro: no there are almost nothing that can be cashed outside a transaction | ||
2014-05-14 13:23 <cedk> also I don't see the point to cache something in memcache instead of postgresql | ||
2014-05-14 13:24 <cedk> just tune your prostgresql configuration and/or add a replication pool of server for read | ||
2014-05-14 13:27 <pokoli> corro: i thought so too, but seems that it's not correct | ||
2014-05-14 13:27 <pokoli> cedk: If cache is shared between all of the instances, you can cache all values and invalidate on writes | ||
2014-05-14 13:28 <cedk> pokoli: nothing can be cached outside a transaction | ||
2014-05-14 13:28 <pokoli> cedk: sorry but don't understand why | ||
2014-05-14 13:28 <cedk> pokoli: because it is transaction | ||
2014-05-14 13:29 <cedk> pokoli: http://www.postgresql.org/docs/current/static/tutorial-transactions.html | ||
2014-05-14 13:29 -!- sharoonthomas(~sharoonth@118.91.181.250) has joined #tryton | ||
2014-05-14 13:30 <pokoli> cedk: so you can cache on commit, and then read from cache on the start of the transaction | ||
2014-05-14 13:30 <cedk> indeed in Tryton, there are very few things that are cached outside the transaction but it is so small that there are no benefit to share them outside the process memory | ||
2014-05-14 13:31 <cedk> pokoli: no. What about other transactions | ||
2014-05-14 13:31 <cedk> pokoli: also I don't want to rewrite postresql in Python | ||
2014-05-14 13:31 <pokoli> cedk: but maybe we need two-phase commit to do this, and i don't know if it's too complex | ||
2014-05-14 13:31 <cedk> postgresql is already a cache | ||
2014-05-14 13:32 <cedk> there are no need for a second external cache | ||
2014-05-14 13:32 <pokoli> cedk: makes sense | ||
2014-05-14 13:33 -!- LordVan(~LordVan@gentoo/developer/LordVan) has joined #tryton | ||
2014-05-14 14:19 -!- pobsteta(~Thunderbi@37.162.245.190) has joined #tryton | ||
2014-05-14 14:22 -!- sharoonthomas(~sharoonth@118.91.181.250) has joined #tryton | ||
2014-05-14 14:55 -!- vernichon(~Thunderbi@gex01-1-78-234-55-95.fbx.proxad.net) has joined #tryton | ||
2014-05-14 15:04 -!- sharoonthomas(~sharoonth@118.91.181.250) has joined #tryton | ||
2014-05-14 15:19 -!- buxy(~rhertzog@mail.vm.ouaza.com) has joined #tryton | ||
2014-05-14 16:18 -!- gpodessta(bed26cf1@gateway/web/freenode/ip.190.210.108.241) has joined #tryton | ||
2014-05-14 16:24 -!- alisonken1lap3(~alisonken@18.18.ultimate-int.uia.net) has joined #tryton | ||
2014-05-14 16:44 -!- rhowe(~rhowe@87-127-165-237.static.enta.net) has joined #tryton | ||
2014-05-14 16:49 -!- digitalsatori(~Thunderbi@116.234.181.110) has joined #tryton | ||
2014-05-14 17:09 -!- mariomop(~quassel@host123.186-108-160.telecom.net.ar) has joined #tryton | ||
2014-05-14 17:14 -!- pobsteta(~Thunderbi@pha75-11-82-236-85-232.fbx.proxad.net) has joined #tryton | ||
2014-05-14 17:52 -!- Telesight(~anthony@77-175-159-159.FTTH.ispfabriek.nl) has joined #tryton | ||
2014-05-14 18:07 -!- littlebit(~Thunderbi@ulmg-5d84dbbf.pool.mediaWays.net) has joined #tryton | ||
2014-05-14 18:11 -!- gremly(~gremly@190.85.36.58) has joined #tryton | ||
2014-05-14 18:54 -!- corro(~robin@77-58-75-66.dclient.hispeed.ch) has joined #tryton | ||
2014-05-14 19:02 -!- pobsteta(~Thunderbi@pha75-11-82-236-85-232.fbx.proxad.net) has joined #tryton | ||
2014-05-14 19:08 -!- pobsteta(~Thunderbi@pha75-11-82-236-85-232.fbx.proxad.net) has joined #tryton | ||
2014-05-14 19:19 -!- pobsteta(~Thunderbi@pha75-11-82-236-85-232.fbx.proxad.net) has joined #tryton | ||
2014-05-14 20:20 -!- vcardon(~vcardon@LNeuilly-152-23-15-185.w193-252.abo.wanadoo.fr) has joined #tryton | ||
2014-05-14 20:25 -!- vcardon(~vcardon@LNeuilly-152-23-15-185.w193-252.abo.wanadoo.fr) has left #tryton | ||
2014-05-14 20:51 -!- pablovannini(~pablo@host126.186-109-85.telecom.net.ar) has left #tryton | ||
2014-05-14 21:51 -!- uranus(~uranus@ool-182fa854.dyn.optonline.net) has joined #tryton | ||
2014-05-14 23:17 -!- asoleh(~asoleh@36.70.82.7) has joined #tryton |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!