Up to Something

We win again. I wanted to run Scalix for a client, inside of a VE/CT/whatever, and I tried using Fedora Core 7 to do so. I was unable to make the installer work and didn't see much in the way of help from Scalix. Here is a link to the Bugzilla page. (Login Required) Here are the important bits. Aside from discouragement I didn't get anything from Florian... Nothing except responses that was. No one else seems to give a damn if I even exist.

Whatever. The point is that with some minor extra-effort, Scalix does indeed work inside of an OpenVZ container on Linux. My host OS is gentoo, running 2.6.18-028stab053.

The Scalix package is pretty great, just up the Java memory once you have it running. The instances (two) on a system shared by two different companies work great. We migrated to Scalix from Kolab. Outlook users (all two) are happy. Thunderbird users didn't see too much of a change.

Yay for us.

Cliff's Latest Post references some cool news about PostgreSQL. Namely this:

IT KICKS MYSQLs A** IN PERFORMANCE TERMS AS WELL. (as everything else - e.g. features, consistency, startup-options, configurability, etc.)

No doubt someone is going to argue that this has to do with use pattern/configuration/(insert excuse here). However, read his article, and read the links.

I don't use Ruby much. I use PostgreSQL quite a bit. My blog pisses me off - lots of slow load times, lots of serious lag. Today, I did a "control-R" refresh in Firefox. Eleven (11) Seconds to load the page! It was awful.

Here is what I did....

• VACUUM
• REINDEX

embassy site-packages # vacuumdb -U superpg  -d jmstypo
VACUUM
embassy site-packages # psql -U superpg -d jmstypo \


              -c "reindex database jmstypo;"


NOTICE:  table ......... was reindexed
....
REINDEX

Full load time: 3 Seconds

Referesh time: 1.5 seconds.

---

I am moving this blog in the next month or two. I want to put ruby onto a faster box. A 1ghz p3 just doesn't buy what it used to.

I hate Apache. I really do. I refuse to vindicate that hatred. There are great aspects about it, but the things I want to do are hampered by things like the sewer-refuse-styled configuration syntax.

I like Nginx. It is fast, simple, and is amazing. It does proxy, reverse proxy, rewrite, ssl, and everything else. Cliff Wells cooked up a wiki. As linked before, it cleans up when facing off with Apache, Pound, Lighttpd, etc. It has all the core features that sane people need.

I just wrote a recipie for Trac + Nginx over at Edgewall. Trac + Nginx + PostgreSQL Kicks ass.

Note: I still use Apache for moddavsvn, and a mod_python only application. (But that will be fixed soon enough). SVN is another question entirely.

Command Prompt supplies the best replication available for PostgreSQL. They support it on RedHat and SuSE at the very least. So far, they have been willing to support Gentoo also. That is pretty hot. I am working on getting it deployed. Along the way I'm producing some solid administration documents for my clients, and trying to make some nice wiki howto related to it. I have not succeeded due to as-yet unknown bugs in something that I have done. However, I hope to soon have it up. Mammoth on Gentoo with Replication

In MySQL you can use 'show databases' along with many other convenient 'show' commands. PostgreSQL is a different beast entirely. using real databases can be challenging if you grew up on MySQL.

Connect to the database with psql:

psql -U postgres template1
opsdb ~ # psql -U postgres template1
Welcome to psql 8.0.3, the PostgreSQL interactive terminal.

Type:  \copyright for distribution terms
       \h for help with SQL commands
       \? for help with psql commands
       \g or terminate with semicolon to execute query
       \q to quit

template1=# select datname from pg_database;
   datname
--------------
 template1
 template0
 mydb
 typo
(4 rows)

template1=# quit

Update: [2005-12-16]

After MANY people have pointed out that "dude" \L works in the Interactive Interface. I would like to tell you: This is useless when you are actually using PostgreSQL from within programs / scripts.