Slashdot Mirror

Did you miss the bullet point that said "Yes, I know. I wrote that line in the docs. It's still really irritating"? As an Apache project member that primarily contributes documentation, I think he has RTFM -- in fact, he has WTFM. ;)

That's 'cause Apache is confused about it's proper name. You see, when you request the URL without the trailing slash, and Apache sees it's (the URL) refering to a directory it redirects the client to a URL containing a trailing slash. This is done so that relative URLs will work. To do the redirect, Apache needs to know it's own name. See: http://httpd.apache.org/docs/2.0/mod/core.html#use canonicalname http://httpd.apache.org/docs/2.0/mod/core.html#ser vername

That's 'cause Apache is confused about it's proper name. You see, when you request the URL without the trailing slash, and Apache sees it's (the URL) refering to a directory it redirects the client to a URL containing a trailing slash. This is done so that relative URLs will work. To do the redirect, Apache needs to know it's own name. See: http://httpd.apache.org/docs/2.0/mod/core.html#use canonicalname http://httpd.apache.org/docs/2.0/mod/core.html#ser vername

Actually there is something that is the natural progrssion from Ant - Maven. Maven is useful because it abstracts common build process across multiple projects by using plug-ins. Maven 1.0.2 is somewhat klunky, and work is progressing on Maven 2, but if you are looking for more information, take a look at Maven Project and buy Maven: A Developer's Notebook.

And, to be fair to Ant, recent versions of Ant 1.6 have added mechanisms to reuse build logic. Really, go check for yourself on the Ant project page.

Actually there is something that is the natural progrssion from Ant - Maven. Maven is useful because it abstracts common build process across multiple projects by using plug-ins. Maven 1.0.2 is somewhat klunky, and work is progressing on Maven 2, but if you are looking for more information, take a look at Maven Project and buy Maven: A Developer's Notebook.

And, to be fair to Ant, recent versions of Ant 1.6 have added mechanisms to reuse build logic. Really, go check for yourself on the Ant project page.

>Well, it would be nice if it didn't stop responding when your access log hits 2GB, too...

One should never let the logs get larger than a few hundred MB. Consider rotating the logs, either monthly or weekly (or even more often if your logs gain size quickly).

See:
http://httpd.apache.org/docs-2.0/programs/rotatelo gs.html or http://httpd.apache.org/docs/programs/rotatelogs.h tml

NeoThermic

>Well, it would be nice if it didn't stop responding when your access log hits 2GB, too...

One should never let the logs get larger than a few hundred MB. Consider rotating the logs, either monthly or weekly (or even more often if your logs gain size quickly).

See:
http://httpd.apache.org/docs-2.0/programs/rotatelo gs.html or http://httpd.apache.org/docs/programs/rotatelogs.h tml

NeoThermic

Ah yes, the response of the prototypical "I'm cool because I promote OSS" fanboy, quick to attack anyone who dares question his beloved OSS software (I'm surprised there wasn't some conspiracy minded anti-Microsoft in there too).

Of course if you're wondering who Rich Bowen is, you can find a little bit about him here (or in plain english, he probably knows more about Apache then the parent and 95% of everyone reading this post - combined)

Can someone explain to me the difference between Ant and Maven? I have noticed almost everyone supports Ant, but there are people who also do Maven. Are there any advantages, disadvantages, are the different but overlap greatly, and why does Apache support what appears to me to be two competitive technologies...

Very confusing to someone who just has a high-level understanding of Apache world.

Can someone explain to me the difference between Ant and Maven? I have noticed almost everyone supports Ant, but there are people who also do Maven. Are there any advantages, disadvantages, are the different but overlap greatly, and why does Apache support what appears to me to be two competitive technologies...

Very confusing to someone who just has a high-level understanding of Apache world.

According to this mail the ASF is ok with the current GNU Classpath licensing, whatever that means is a different question.

Yeah because as we all know, IBM's never done anything for Java.

If you'll look at harmony's website, you'll see that they aren't just implementing the Java Virtual Machine as you put it. They are implementing J2SE. J2SE is not just a language. It is a platform.

Platform fragmentation is as, or more important than, language fragmentation*. A language cannot stand alone. You need libraries. Platform fragmentation is what Sun is worried about right now, not language fragmentation. The JVM is not even part of the "open source java" debate, since open source JVMs already exist and Sun is more or less encouraging them!

Please see my other comment here.

* Language fragmentation can still happen if Harmony chooses to implement different JSRs than Sun does for some reason. However it is incredibly unlikely that this would be a bad thing. As long as Harmony stays within the accepted protocols for extending the Java language, and keeps any experimental/unapproved-JSR features cleanly quarantined within the -XX "pragma" flags (both of which things, Microsoft did NOT), this will be fine.

And for those of us who don't read the article and don't know what Harmony is:

Harmony is the Apache Foundations project to create an open source edition of Java SE (Standard Edition).

Have you looked at JMeter? I've used it to test web/server applications, but it can be used for other applications too.

JMeter

The only open source solution I'm familiar with is JMeter. I haven't worked with it much, but in the few projects I've used it, it's been helpful. I't part of the Jakarta project. You can find it at http://jakarta.apache.org/jmeter/index.html

Gotcha. In that case I'd use CC to trigger the builds, but use your existing system to do the build itself. That's similar actually what we do at SAS. CC watches the CVS repo (or in our case, a single trigger file) for changes. When a change occurs, CC triggers the build, but the build runs inside your existing build system. Can you call into the existing system with a script?

The problem we have is that I have to schedule builds on every damn server rather than just "make it so".

Sounds like you've got something rather complicated going... but I'd be tempted to get a "big picture" fresh look now that you've got a working system to see if it can't be cleaned up. On the other hand, if it's working for you...

My latest "innovation" to my company's system is to run a completeness test on the distribution to be shipped by installing it and comparing it to what we wanted. Then run a series of automated smoke tests across the results.

I love it!

If you took the system described in the book, I'd have to build on it by adding a 1001 patch scripts to get it to do the rest of the things that I want. I want a system that I can extend in a rational, maintainable way.

ehhh... I think that if you wrapped your existing system (as I mentioned above) you might be able to use a CI system like CC just for scheduling. That should remove a lot of the complexity you're anticipating.

Maybe I should look at Maven 2. How does it handle scheduled events?

Maven 2 has me excited. :) They'll use Continuum http://maven.apache.org/continuum/ as their CI system. It should also have project to project dependancies. Release is scheduled for August.

I suppose what I really want CruiseControl to do is to add a collect event (or post-build) event, where the results of the build system get updated to the central repository.

I think you want the concept of a Maven repository. Maven publishes build artifacts to a centralized repo so that other builds can consume them.

Umm. Sorry - went of into free-form.

:) It's more fun that way! Sorry for the slow replies... it's been a ~crazy~ day!

People did get to 'see the Apache Web server in action'...

That must have been useful. I mean, considering how you can't download Apache installer packages for Win32 and try it out right on your own Windows desktop, I can see how people would clamor to see this hard-to-get product do its thing.

Oh, wait...

...at hobbyist, hippie projects like Geronimo, that dare to offer "competition" to his product, even to the point of passing all the relevant certifications?

IMHO, projects under BSD license make people/companies more comfortable.

I, as well as other colleagues in my company, usually download, study, and customize the projects from Apache, such as Ant, maven, tomcat and so on. we have never concern the license issue, since we all have known projects on Apache are released under it's BSD-like license, anybody can use and change them with or without publishing their work under this license.

Meanwhile, the case when we met GPL was quite different. A few months ago, Company decided to use a open source database for internal project use. Nearly almost of colleagues proposed MySql as the favorite choice. Things were going on well until somebody studied Mysql's so-called dual-license. The folk said to us: we have to pay for it due to its license.

We argued about this issue for a long time, and with no result. Many days later, Manager talked about it in a routine meeting, then he said: "Silver, please make clear the price of MySql , and tell me; to me, if things is not certain, I will have to choose a safe way, that's it."

IMHO, projects under BSD license make people/companies more comfortable.

I, as well as other colleagues in my company, usually download, study, and customize the projects from Apache, such as Ant, maven, tomcat and so on. we have never concern the license issue, since we all have known projects on Apache are released under it's BSD-like license, anybody can use and change them with or without publishing their work under this license.

Meanwhile, the case when we met GPL was quite different. A few months ago, Company decided to use a open source database for internal project use. Nearly almost of colleagues proposed MySql as the favorite choice. Things were going on well until somebody studied Mysql's so-called dual-license. The folk said to us: we have to pay for it due to its license.

We argued about this issue for a long time, and with no result. Many days later, Manager talked about it in a routine meeting, then he said: "Silver, please make clear the price of MySql , and tell me; to me, if things is not certain, I will have to choose a safe way, that's it."

IMHO, projects under BSD license make people/companies more comfortable.

I, as well as other colleagues in my company, usually download, study, and customize the projects from Apache, such as Ant, maven, tomcat and so on. we have never concern the license issue, since we all have known projects on Apache are released under it's BSD-like license, anybody can use and change them with or without publishing their work under this license.

Meanwhile, the case when we met GPL was quite different. A few months ago, Company decided to use a open source database for internal project use. Nearly almost of colleagues proposed MySql as the favorite choice. Things were going on well until somebody studied Mysql's so-called dual-license. The folk said to us: we have to pay for it due to its license.

We argued about this issue for a long time, and with no result. Many days later, Manager talked about it in a routine meeting, then he said: "Silver, please make clear the price of MySql , and tell me; to me, if things is not certain, I will have to choose a safe way, that's it."

IMHO, projects under BSD license make people/companies more comfortable.

I, as well as other colleagues in my company, usually download, study, and customize the projects from Apache, such as Ant, maven, tomcat and so on. we have never concern the license issue, since we all have known projects on Apache are released under it's BSD-like license, anybody can use and change them with or without publishing their work under this license.

Meanwhile, the case when we met GPL was quite different. A few months ago, Company decided to use a open source database for internal project use. Nearly almost of colleagues proposed MySql as the favorite choice. Things were going on well until somebody studied Mysql's so-called dual-license. The folk said to us: we have to pay for it due to its license.

We argued about this issue for a long time, and with no result. Many days later, Manager talked about it in a routine meeting, then he said: "Silver, please make clear the price of MySql , and tell me; to me, if things is not certain, I will have to choose a safe way, that's it."

IMHO, projects under BSD license make people/companies more comfortable.

I, as well as other colleagues in my company, usually download, study, and customize the projects from Apache, such as Ant, maven, tomcat and so on. we have never concern the license issue, since we all have known projects on Apache are released under it's BSD-like license, anybody can use and change them with or without publishing their work under this license.

Meanwhile, the case when we met GPL was quite different. A few months ago, Company decided to use a open source database for internal project use. Nearly almost of colleagues proposed MySql as the favorite choice. Things were going on well until somebody studied Mysql's so-called dual-license. The folk said to us: we have to pay for it due to its license.

We argued about this issue for a long time, and with no result. Many days later, Manager talked about it in a routine meeting, then he said: "Silver, please make clear the price of MySql , and tell me; to me, if things is not certain, I will have to choose a safe way, that's it."

It would appear that it *was* a bug in 1.3 (it's not very clear on most of the pages). And it seems to have been fixed in 1.3.33, as you can read here:

http://www.apache.org/dist/httpd/Announcement.html

The poor design decisions will be fixed and the JVM will be eliminated.

I doubt either of those things will actually happen. Most of the standard library is a creeping horror from a security and extensibility standpoint, but the fundamental design hasn't changed noticeably since 1.1-- they've just heaped another cargo ship full of additional cruft into the standard runtime with each major release. Stack-based security with principals attached is clumsy, but it appears to be getting treated as the state of the art by both the Java and .Net camps. They both need to be hit with a capability and design pattern cluestick. I'd say to hit them with a COM stick, but they'd probably use the painful parts like object registries and reference counting, and miss the important lesson: abandon globals and statics, interfaces and parameters are the key.

As for elimination of the JVM, it won't likely happen for a few reasons, most notably that even in a system which entirely precompiles to native code, the JVM is still the basis of the language's functionality specification. A C-like Java compiler can certainly produce correct code so long as the same program does the same thing as it would on a JVM. Also, dynamic loading in Java is specified in terms of JVM bytecode.

The result will be a fast compiled language with an elegant syntax.

LMAO! The syntax is about as elegant as C without a preprocessor. In such a fix, one becomes dependent on souped-up editors to churn out boilerplate code. Velocity works as a nice simple preprocessor; for more intricate macros JSE might be a better idea.

I don't think Java should try to morph into C or C++. In my opinion the JVM is the strongest link in the chain; in exchange for a hit in startup time you get the speed of a compiled language, the flexibility of a symbolic interpreter and the safety of fully checked memory access rolled into one. It's probably not the best thing to build grep or wc on, but it's great for bigger things like daemons and windowed apps.

It's a myth that it's a myth that Apache was named that. From their website, the FAQ originally said this:
http://apache.org/docs/misc/FAQ.html#name">http:// web.archive.org/web/19980128114236/http://apache.o rg/docs/misc/FAQ.html#name
And then said this:
http://www.apache.org/docs/misc/FAQ.html#name">htt p://web.archive.org/web/20000815061003/http://www. apache.org/docs/misc/FAQ.html#name
Then they changed it to:
http://httpd.apache.org/docs/misc/FAQ.html#name">h ttp://web.archive.org/web/20021017033945/http://ht tpd.apache.org/docs/misc/FAQ.html#name
Now they're trying to get rid of something they've perpetuated for years:
http://httpd.apache.org/docs/misc/FAQ.html#name">h ttp://web.archive.org/web/20030603200610/http://ht tpd.apache.org/docs/misc/FAQ.html#name

and that seems to be the one that's remained until today. Who knows what it'll be tomorrow.

It's a myth that it's a myth that Apache was named that. From their website, the FAQ originally said this:
http://apache.org/docs/misc/FAQ.html#name">http:// web.archive.org/web/19980128114236/http://apache.o rg/docs/misc/FAQ.html#name
And then said this:
http://www.apache.org/docs/misc/FAQ.html#name">htt p://web.archive.org/web/20000815061003/http://www. apache.org/docs/misc/FAQ.html#name
Then they changed it to:
http://httpd.apache.org/docs/misc/FAQ.html#name">h ttp://web.archive.org/web/20021017033945/http://ht tpd.apache.org/docs/misc/FAQ.html#name
Now they're trying to get rid of something they've perpetuated for years:
http://httpd.apache.org/docs/misc/FAQ.html#name">h ttp://web.archive.org/web/20030603200610/http://ht tpd.apache.org/docs/misc/FAQ.html#name

and that seems to be the one that's remained until today. Who knows what it'll be tomorrow.

It's a myth that it's a myth that Apache was named that. From their website, the FAQ originally said this:
http://apache.org/docs/misc/FAQ.html#name">http:// web.archive.org/web/19980128114236/http://apache.o rg/docs/misc/FAQ.html#name
And then said this:
http://www.apache.org/docs/misc/FAQ.html#name">htt p://web.archive.org/web/20000815061003/http://www. apache.org/docs/misc/FAQ.html#name
Then they changed it to:
http://httpd.apache.org/docs/misc/FAQ.html#name">h ttp://web.archive.org/web/20021017033945/http://ht tpd.apache.org/docs/misc/FAQ.html#name
Now they're trying to get rid of something they've perpetuated for years:
http://httpd.apache.org/docs/misc/FAQ.html#name">h ttp://web.archive.org/web/20030603200610/http://ht tpd.apache.org/docs/misc/FAQ.html#name

and that seems to be the one that's remained until today. Who knows what it'll be tomorrow.

It's a myth that it's a myth that Apache was named that. From their website, the FAQ originally said this:
http://apache.org/docs/misc/FAQ.html#name">http:// web.archive.org/web/19980128114236/http://apache.o rg/docs/misc/FAQ.html#name
And then said this:
http://www.apache.org/docs/misc/FAQ.html#name">htt p://web.archive.org/web/20000815061003/http://www. apache.org/docs/misc/FAQ.html#name
Then they changed it to:
http://httpd.apache.org/docs/misc/FAQ.html#name">h ttp://web.archive.org/web/20021017033945/http://ht tpd.apache.org/docs/misc/FAQ.html#name
Now they're trying to get rid of something they've perpetuated for years:
http://httpd.apache.org/docs/misc/FAQ.html#name">h ttp://web.archive.org/web/20030603200610/http://ht tpd.apache.org/docs/misc/FAQ.html#name

and that seems to be the one that's remained until today. Who knows what it'll be tomorrow.

I hate to run your day you know, but that's why it's called Apache in the first place. It comes from the days where the Apache guys were heavily patching the NCSA webserver. Sorry :(

Link

• The current production version of Apache 2.x is 2.0.54. 2.1 is alpha-quality code, the unstable development branch.
• The advisory's dated 5th July, but I certainly haven't seen anything on any of the usual lists about it (and I monitor them as part of my job.)

Not to say it's impossible, the HTTP request smuggling attack vector is real enough - the paper is interesting reading, see http://www.watchfire.com/resources/HTTP-Request-Sm uggling.pdf

http://www.apache.org/dist/httpd/CHANGES_2.1 The 2.1 branch is still classified as alpha software, however.

Apache Lenya http://lenya.apache.org/ is an Open Source Java/XML Content Management System and comes with revision control, site management, scheduling, search, WYSIWYG editors, and workflow

This Perchild MPM? "This module is not functional. Development of this module is not complete and is not currently active. Do not use perchild unless you are a programmer willing to help fix it." Doesn't seem like something I could use for a production web site. ;)

or an HTTP server

I really don't want to be a Windows apologist. But that one's about as simple as downloading Apache and double clicking an icon.

While I agree that installing OSX stuff (incl. X11, and I'm impressed) on my sweeties Power Book, this statement is pure hyperbole.

We considered Jython and Groovy when we were adding scripting to an app, since they were the only two scripting languages that came with licenses that were found to be acceptable by our legal department.

Why choose at all? Why not use the Bean Scripting Framework (originally from IBM's Alpha Works, IIRC) and let the customer choose whichever language they prefer? BSF lets you embed any scripting language that has a conforming wrapper. BSF wrappers are available for Groovy, Jython, BeanShell, Rhino, JRuby, Tcl, NetRexx, XSLT, and perl among others (although it's been my experience that some of these aren't all there yet).

"Want to run *two* webservers behind NAT?"
http://httpd.apache.org/docs-2.0/mod/mod_proxy.htm l

I guess you never heard of virtual hosts? HTTP 1.1 requires sending the host you want to connect with, and most modern web servers like Apache can key off that, so you can run 2 completely different web sites from the same web server. Don't give me bullshit that you still want to run an HTTP 1.0 browser.

Stop your bitching. NAT is a great measure that basically saved the Internet for you and me. There will always be clever workarounds that will extend the life of IPv4 probably forever.

And yes, cell phones and PDAs should *NOT* be exposed to the internet. They should all be behind firewalls owned by the service providers. Normal users can't even update their computers with Windows Update, which is a one-click process. How the hell do you expect them to keep up with updating their cell phones after wave after wave of exploits come?

You are an idiot.

Really, it's just basic journalism. Something slashdot clearly knows absolutely bupkus about.

There are plenty. A good Free/Open one is called Eclipse. It is plug-in based and supports Java/Servlet/JSP/C++ and others. You can just download Eclipse and Tomcat and have a complete JSP/Servelt web application framework that works on MS Windows or Linux. Now just buy a JSP/Servlet book or search Google for a JSP/Servlet tutorial and your all set ; )

Quoting from the proposal http://mail-archives.apache.org/mod_mbox/incubator -harmony-dev/200505.mbox/%3C3923A844-DEC5-4CC2-ADE D-B1F144BB6AF5@apache.org%3E:

"We will create directly, via inclusion of independent third-party code, or through contribution :

a) a freely implementable specification of a modular VM and class library that allows for multiple, independent implementations

b) a test suite for interoperability testing of the modules

c) an implementation under the Apache License of the modular VM [...]."

I'll be sure to go. I'm very interested in hearing more about James, their new news and mail server written in Java! http://james.apache.org/

I'd personally love to see the Apache Project coordinate and release a mail server.

Obviously this guy has not heard of Postfix, a truely awesome mailserver

I'd personally love to see the Apache Project coordinate and release a mail server. Is this: http://james.apache.org/ close enough for you? It looks to be fairly complete.

Example from ASF

I'm lucky. I know lots of really fast FTP servers with lots of high quality software on them. And best of all? It's completely free, and legal.

Please be aware: SPF and Sender-Id are NOT THE SAME THING.

I don't think any major ISP has adopted Sender-Id, which is Microsoft's ill-advised attempt to embrace-and-extend SPF in a heavily encumbered way.

2. Microsoft fights SPAM. Slashdot equally outraged.
Conclusion: Microsoft is always evil no matter what they do.

Nope, Microsoft isn't fighting SPAM - if they were they'd be cooperating with the "rest of the Internet", instead of promoting their own proprietary scheme - SenderID - that's so un-open as to provoke this comment from the Apache Software Foundation:

We believe the current license is generally incompatible with open source, contrary to the practice of open Internet standards, and specifically incompatible with the Apache License 2.0. Therefore, we will not implement or deploy Sender ID under the current license terms.

Various other disparate organisations have raised similar concerns, eventually resulting in the IETF ditching Microsoft's proposal.

Microsoft, at least in this case, weren't interested in a working solution; they were interested in a Microsoft-friendly, FLOSS-hostile solution. Which is daft, given the open-source nature of most Internet technologies.