summaryrefslogtreecommitdiffstats
path: root/zarb-ml/mageia-webteam/2011-January/000133.html
diff options
context:
space:
mode:
authorNicolas Vigier <boklm@mageia.org>2013-04-14 13:46:12 +0000
committerNicolas Vigier <boklm@mageia.org>2013-04-14 13:46:12 +0000
commit1be510f9529cb082f802408b472a77d074b394c0 (patch)
treeb175f9d5fcb107576dabc768e7bd04d4a3e491a0 /zarb-ml/mageia-webteam/2011-January/000133.html
parentfa5098cf210b23ab4f419913e28af7b1b07dafb2 (diff)
downloadarchives-master.tar
archives-master.tar.gz
archives-master.tar.bz2
archives-master.tar.xz
archives-master.zip
Add zarb MLs html archivesHEADmaster
Diffstat (limited to 'zarb-ml/mageia-webteam/2011-January/000133.html')
-rw-r--r--zarb-ml/mageia-webteam/2011-January/000133.html151
1 files changed, 151 insertions, 0 deletions
diff --git a/zarb-ml/mageia-webteam/2011-January/000133.html b/zarb-ml/mageia-webteam/2011-January/000133.html
new file mode 100644
index 000000000..19b614070
--- /dev/null
+++ b/zarb-ml/mageia-webteam/2011-January/000133.html
@@ -0,0 +1,151 @@
+<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
+<HTML>
+ <HEAD>
+ <TITLE> [Mageia-webteam] Initial hosting requirements for maintainers db
+ </TITLE>
+ <LINK REL="Index" HREF="index.html" >
+ <LINK REL="made" HREF="mailto:mageia-webteam%40mageia.org?Subject=Re%3A%20%5BMageia-webteam%5D%20Initial%20hosting%20requirements%20for%20maintainers%20db&In-Reply-To=%3C1294796195.32187.210.camel%40akroma.ephaone.org%3E">
+ <META NAME="robots" CONTENT="index,nofollow">
+ <META http-equiv="Content-Type" content="text/html; charset=us-ascii">
+ <LINK REL="Previous" HREF="000132.html">
+ <LINK REL="Next" HREF="000134.html">
+ </HEAD>
+ <BODY BGCOLOR="#ffffff">
+ <H1>[Mageia-webteam] Initial hosting requirements for maintainers db</H1>
+ <B>Michael Scherer</B>
+ <A HREF="mailto:mageia-webteam%40mageia.org?Subject=Re%3A%20%5BMageia-webteam%5D%20Initial%20hosting%20requirements%20for%20maintainers%20db&In-Reply-To=%3C1294796195.32187.210.camel%40akroma.ephaone.org%3E"
+ TITLE="[Mageia-webteam] Initial hosting requirements for maintainers db">misc at zarb.org
+ </A><BR>
+ <I>Wed Jan 12 02:36:35 CET 2011</I>
+ <P><UL>
+ <LI>Previous message: <A HREF="000132.html">[Mageia-webteam] Initial hosting requirements for maintainers db
+</A></li>
+ <LI>Next message: <A HREF="000134.html">[Mageia-webteam] Initial hosting requirements for maintainers db
+</A></li>
+ <LI> <B>Messages sorted by:</B>
+ <a href="date.html#133">[ date ]</a>
+ <a href="thread.html#133">[ thread ]</a>
+ <a href="subject.html#133">[ subject ]</a>
+ <a href="author.html#133">[ author ]</a>
+ </LI>
+ </UL>
+ <HR>
+<!--beginarticle-->
+<PRE>Le mardi 11 janvier 2011 &#224; 23:07 +0000, Kosmas Chatzimichalis a &#233;crit :
+&gt;<i> The initial requirements for installing the maintainers db in the
+</I>&gt;<i> mageia server are:
+</I>
+Again I am sorry to be harsh and quite direct, but the requirements
+seems rather &quot;too much&quot;.
+
+I do not think it would be wise for us to host a full environment for
+each application based on every coder preference, especially when taking
+in account that our main product is a Linux distribution, ie basically a
+environnement that serve as a base for running other softwares.
+
+So bypassing this seems rather odd. I would maybe do it for a business
+critical complex application whose fate of my company depend on, but I
+do not think this project belong on this category for the moment, nor
+that it couldn't be fulfilled by what we offer in the distribution so
+far.
+
+&gt;<i> 1. RVM (Ruby version manager)
+</I>
+&gt;<i>From what I know, that would likely mean compiling our own ruby version
+</I>on the server, using its own separate set of gems. In term of work, it
+would like adding a specific chroot, or a special vm of a different
+distribution just to host the application. ( different distribution
+since that would be totally unintegrated with the rest of the servers )
+
+I am sorry to say that I do not really see this as a good idea, because
+that put the burden of taking care of the environnement on sysadmin
+(especially security wise), and totally bypass the collaboration system
+that we have setup ( ie puppet + svn + integrated shared hosting ).
+
+&gt;<i> 2. Rubygems (1.3.7)
+</I>&gt;<i> 3. Rails (3.0.3)
+</I>
+Again, I would rather use the package given by the distribution used on
+the server ( ie a Mandriva 2010.1 for the moment, soon to be upgraded to
+the first Mageia release ). This would mean for the moment rails 2.3.10
+and ruby-RubyGems 1.3.5.
+
+A small minor backport would be of course ok ( ie for rubygems ), but a
+full update of the rails stack is likely be more work, especially since
+it would requires others updates ( like the various active-* parts ).
+
+Moreover, using distribution rpm give everybody the same set of module
+to work with, if the need to host/develop multiple rails applications
+arise ( and I think we cannot exclude this possibility ) without having
+to have 1 set of gems per application. And again, we will not need to
+handle security ourself ( or at least, no need to do the hard work as
+this is the goal of the security team ).
+
+So far, we were perfectly able to do it for every perl web application
+we wrote ( be it epoll, mga-mirrors, catdap ), and also for external
+applications in python and perl ( sympa, bugzilla, transifex ).
+
+&gt;<i> 4. MySQL or PostgreSQL (if everything else is using it).
+</I>
+I would strongly support Postgresql, as we do have a server running, and
+that everything is using it on the web server so far ( ie, the 5
+database enabled application use it, and wiki should also ).
+
+Given the fact that the maintainer db is basically a simple CRUD web
+application, I doubt that this will change much for you as I do not
+expect any advanced mysql features, except the old &quot;autoincrement vs
+trigger&quot; trick. But I also hope that modern ORM used nowadays are able
+to hide that sort of low level details.
+
+Again, if there is no other way but to use mysql, we could, but I do not
+think this is the case.
+
+&gt;<i> 5. Passenger (Apache mod_rails)
+</I>
+As I told in my answer on the git topic, could we avoid using passenger,
+and switch to fastcgi ?
+
+Otherwise, the web server will likely suffer of bloat since each process
+will bundle perl interpreter + modules (due to bugzilla installation
+that requires mod_perl) and a ruby interpreter + modules. And I think we
+should avoid this for performance reasons, since processes are reused
+between each request by apache ( standard mpm-prefork, as I am not sure
+that everything will be thread safe ).
+
+We also use fastcgi to be able to finely tune application, by deciding
+how much server should be started per web application. And so far, every
+application we use are able to do it, except bugzilla who requires
+mod_perl for the moment.
+
+I know that rails support it too, and we will take care of using static
+mode to avoid the issue of slow startup of the server, problem which is
+basically the same for catalyst, who is used on 3 of our applications
+( epoll, catdap, mga-mirrors ) at the moment, the same for django, used
+for transifex .
+
+--
+Michael Scherer
+
+</PRE>
+
+
+<!--endarticle-->
+ <HR>
+ <P><UL>
+ <!--threads-->
+ <LI>Previous message: <A HREF="000132.html">[Mageia-webteam] Initial hosting requirements for maintainers db
+</A></li>
+ <LI>Next message: <A HREF="000134.html">[Mageia-webteam] Initial hosting requirements for maintainers db
+</A></li>
+ <LI> <B>Messages sorted by:</B>
+ <a href="date.html#133">[ date ]</a>
+ <a href="thread.html#133">[ thread ]</a>
+ <a href="subject.html#133">[ subject ]</a>
+ <a href="author.html#133">[ author ]</a>
+ </LI>
+ </UL>
+
+<hr>
+<a href="https://www.mageia.org/mailman/listinfo/mageia-webteam">More information about the Mageia-webteam
+mailing list</a><br>
+</body></html>