January 2014 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] Pharo Success Story: Reflex Vacances

by Nicolas Petton

Yep, especially customer's pictures are pretty big. That's because the DB was already like this when we started the new website. Running ImageMagick's convert on all images would improve things quite a lot :) Nico Sven Van Caekenberghe writes: > On 30 Jan 2014, at 11:17, phil(a)highoctane.be wrote: > >> Hello, >> >> I paid a visit to the site. >> >> Looks good. Images are loading awfully slow tough. >> >> Isn't there any image caching? It seems that they aren't resized and the web browser loads the big pictures for each little picture (it does as a matter of fact). >> >> http://www.reflex-vacances.fr/location-vacances/appartement-le-mans/c755c60d >> >> loads >> >> http://www.reflex-vacances.fr/img/ebb76d6a/Photomaison037.jpg >> >> which is quite big (1.7MB each, 3 of them on that page). >> >> There is also a couple of js requests instead of one big load. >> >> The whole loading on an empty cache is more than 10 seconds. Including Google Maps details, 16. Yikes, that's a hell of a slow beast. >> >> I can't help but think about someone looking at this from a mobile device, even on 3G. >> >> The whole network thing (looking at the Chrome Devtools Network tab) seems to avoid a couple of best practices: >> >> https://developers.google.com/speed/docs/best-practices/rtt >> >> like: >> >> - https://developers.google.com/speed/docs/best-practices/rtt#CombineExternal… >> >> Maybe https://developers.google.com/speed/docs/best-practices/rtt#ParallelizeDown… could help with all of those pictures. >> >> Some ImageMagick preprocessing would definitely help with the images. >> >> I tell you that because we implemented such a kind of site in the past (not Pharo) and had to make it work fast (this was for expensive houses, visitors were kind of impatient). >> >> Also, I just help businesses avoid this kind of slowdown as one of my business offerings, so I couldn't resist ;-) > > I didn't look that carefully, but yes, great analysis. > > Image sizes are very important for a site like that, ImageMagic is your friend, sizes should be in the 100-200-300 Kb range. > > YSlow is also a very good tool. > > Sven >> Regards, >> Phil >> >> >> >> >> >> >> >> >> >> >> >> >> >> On Thu, Jan 30, 2014 at 11:02 AM, Nicolas Petton <petton.nicolas(a)gmail.com> wrote: >> >> Torsten Bergmann writes: >> >> >> runs 100% in Pharo + Iliad + MongoDB. >> > >> > "Look ma - no apache or nginx". >> >> Hi! >> >> Apache is used as a frontend server, and many images are indeed in the >> db, thus served through Pharo. >> >> The Pharo image is a 1.4, the VM is Cog, with a typical Mongo+Voyage >> setup. Most of the DB data is cached by Voyage, which partly explains >> the speed. >> >> Mongo and Pharo are running on the same machine, there's no special >> caching except for what Voyage provides by default. >> >> The Iliad app is served by Kom, and we have many bash scripts to monitor, >> auto-restart and manage the app (we manage SmalltalkHub in a similar >> way). >> >> Cheers, >> Nico >> >> >> -- >> Nicolas Petton >> http://nicolas-petton.fr >> >> -- Nicolas Petton http://nicolas-petton.fr

Jan. 30, 2014

Re: [Pharo-dev] StHub (mostly) down ?

by Nicolas Petton

Indeed. There were 5 minutes of downtime. Nico Sabine KnÃ¶fel writes: > Hi, > > smalltalkhub seems to be down now again. > > Regards > Sabine > > Ups! Google Chrome konnte keine Verbindung zu www.smalltalkhub.com > herstellen. > > > > -- > View this message in context: http://forum.world.st/StHub-mostly-down-tp4739307p4740327.html > Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com. -- Nicolas Petton http://nicolas-petton.fr

Jan. 30, 2014

Re: [Pharo-dev] StHub (mostly) down ?

by Max Leske

good from Bern too. On 30.01.2014, at 13:18, Damien Cassou <damien.cassou(a)gmail.com> wrote: > On Thu, Jan 30, 2014 at 11:49 AM, Sabine KnÃ¶fel > <sabine.knoefel(a)gmail.com> wrote: >> Damien, does it work now? for me down again. > > > working fine :-) > > -- > Damien Cassou > http://damiencassou.seasidehosting.st > > "Success is the ability to go from one failure to another without > losing enthusiasm." > Winston Churchill >

Jan. 30, 2014

Re: [Pharo-dev] StHub (mostly) down ?

by Damien Cassou

On Thu, Jan 30, 2014 at 11:49 AM, Sabine KnÃ¶fel <sabine.knoefel(a)gmail.com> wrote: > Damien, does it work now? for me down again. working fine :-) -- Damien Cassou http://damiencassou.seasidehosting.st "Success is the ability to go from one failure to another without losing enthusiasm." Winston Churchill

Jan. 30, 2014

Re: [Pharo-dev] files.pharo.org/ appears to beOffline!

by Marcus Denker

it seem there was some network problem at Inriaâ¦ On 30 Jan 2014, at 11:51, Noury Bouraqadi <bouraqadi(a)gmail.com> wrote: > > Noury Bouraqadi > Ecole des Mines de Douai > http://car.mines-douai.fr/noury > -- > > > >

Jan. 30, 2014

[pharo-project/pharo-core]

by GitHub

Branch: refs/tags/20627 Home: https://github.com/pharo-project/pharo-core

Jan. 30, 2014

[pharo-project/pharo-core] 0c32b4: 20627

by GitHub

Branch: refs/heads/2.0 Home: https://github.com/pharo-project/pharo-core Commit: 0c32b480faa5ca015008c7765dac570fe0937f52 https://github.com/pharo-project/pharo-core/commit/0c32b480faa5ca015008c776… Author: Jenkins Build Server <board(a)pharo-project.org> Date: 2014-01-30 (Thu, 30 Jan 2014) Changed paths: A NativeBoost-Core.package/CompiledMethod.extension/instance/hasNativeCallPrimitive.st A NativeBoost-Core.package/CompiledMethod.extension/instance/nbArgumentNames.st M NativeBoost-Core.package/CompiledMethod.extension/methodProperties.json M NativeBoost-Core.package/NBFFICallout.class/instance/sender..st M NativeBoost-Core.package/NBFFICallout.class/methodProperties.json A NativeBoost-Core.package/NativeBoost.class/class/prepareForProduction.st M NativeBoost-Core.package/NativeBoost.class/methodProperties.json M ScriptLoader20.package/ScriptLoader.class/instance/commentForCurrentUpdate.st A ScriptLoader20.package/ScriptLoader.class/instance/script600.st A ScriptLoader20.package/ScriptLoader.class/instance/update20627.st M ScriptLoader20.package/ScriptLoader.class/methodProperties.json Log Message: ----------- 20627 12120 Let NativeBoost work without sources https://pharo.fogbugz.com/f/cases/12120

Jan. 30, 2014

[regression reporter]regression occurred

by no-reply＠ci.inria.fr

https://ci.inria.fr/pharo/job/Pharo-2.0-Tests/Run=run%201,VM=vm,label=linux… 1 regressions found. FileSystem.Tests.Disk.FileHandleTest.testTruncate

Jan. 30, 2014

Jenkins build became unstable: Pharo-2.0-Tests Â» run 1,vm,linux #280

by no-reply＠ci.inria.fr

See <https://ci.inria.fr/pharo/job/Pharo-2.0-Tests/Run=run%201,VM=vm,label=linux…>

Jan. 30, 2014

Re: [Pharo-dev] Pharo Success Story: Reflex Vacances

by phil＠highoctane.be

YSlow +1 OptiPNG is your friend: http://optipng.sourceforge.net/pngtech/optipng.html https://tinypng.com/ interesting as well. Phil

Jan. 30, 2014