Help coderanch get a
new server
by contributing to the fundraiser

Galen Palmer

+ Follow
since Aug 09, 2007
Merit badge: grant badges
For More
Cows and Likes
Total received
In last 30 days
Total given
Total received
Received in last 30 days
Total given
Given in last 30 days
Forums and Threads
Scavenger Hunt
expand Ranch Hand Scavenger Hunt
expand Greenhorn Scavenger Hunt

Recent posts by Galen Palmer

Thank you for the quick, comprehensive response.

A (probably) dumb follow up question. I'm the kind that still holds on to his old mobile for years so I'm unlikely to change phone, but can you test on a Android phone even if you don't have phone service on that phone?
13 years ago
What should I look for in a tablet if I want to try out this book? I don't have an android phone so I thought I might pick up a "cheaper" tablet to test things out with. Any recommendations for minimum specs or any other gotchas I might need to be aware of?
13 years ago
Cool. Picked up the e-book last night. Only a chapter into it but so far this is the best Rails practices book I've read since Agile Web Development with Rails. Great job!
13 years ago
Is your book aimed at Rails 2 or Rails 3?

If both, did you find that the anti-patterns changed between version? For example, are there some anti-patterns that are actually hard to implement in Rails 3 that were common in Rails 2?
13 years ago
I appreciate the responses, but I guess I didn't make the main point of my question clear: when migrating an existing app is it worth it to try and "go the 3.x way" or can a legacy app be maintainable in a partially migrated state and still take advantage of all the 3.x goodness?
14 years ago
I've got various old Spring 2.x based apps where the only annotations are for Hibernate and all the other bean wiring is written into the root-servlet.xml file. Is it possible to incrementally migrate an app towards the 3.x way of doing things? Or, is it worth the time to migrate the app all in one go?
14 years ago
Another option for a quick check. Under linux (or a mac) terminal:

should show a tomcat process running.

15 years ago
I think I'm going to re-architect slightly so that the tomcat app is accessible from a named "context" (folder/directory/?) instead from root. In other words, instead of have it under .

I discover one thing that may help people. If my ProxyPass name did not match my tomcat webapp name, then security wasn't working at all (this was a local test using the default Apache 2 setup under Leopard).

In other words:

did not work, but this works fine:

By "work" I mean that the user was not able to log in under the 1st version but was under the 2nd version...
15 years ago
One last note: This is set up under Linux (centos5).
15 years ago
It sounds like you're far enough along that you have a couple of lines like:

In your Apache httpd configuration. Can you post your progress so far? It'll be easier to debug that way.
15 years ago
I'm trying to log into a tomcat app (6.0.x) using apache (2.2.x) as the front end using mod_proxy_ajp. Everything works great except when I try to set up the app as a "default" or "root" app for my domain. In this case, the login fails because it appears that j_security_check is not being forwarded correctly.

For example: I can get works fine. In this case my apache VirtualHost configuration looks like:

However, if I try to make it the default app for my domain (e.g. Apache fails with a connection reset exception. The access_log displays about 10 entries like:

In this case, my apache VirtualHost looks like:

I tried adding a line so the proxy passes look like :

But I got the same behavior as I got without that extra line.

Anyone have any ideas on the correct way to do this? BTW, I really DO want everything for this specific domain to get fowarded to the webapp. My only problem right now is related to the tomcat login.

Thanks in advance!
15 years ago
I have a class where there are bunch of read functions that are accessed frequently and an update function that will be used infrequently but should block all the read functions until it is complete.

I'm not sure about the terminology used in java.util.concurrent.lock so Im not sure if there's a solution in there.

Here's my requirements:
1. Multiple concurrent reads should be allowed
2. The update functions corrupt the inner model so all read methods should placed on hold until the update is complete

I was reading the Semephore docs and wondered if that was a starting point. IN essence, I'd want a version that provided an infinite number of permits for reads (since reads don't interfere with each other) but when one of the update methods was called needed to start it would wait for all existing reads to complete then revoke all permits before starting the update.

The higher level question: is there a term that describes this particular type of problem (or its solution)? Semephore doesn't seem to quite encapsulate what I'm looking for.

The lower level question: how might one solve this?

I stumbled across your post while doing a google search on the error message. No solution for you, but I am noticing the same exception when I add @Transactional to my DAO interface and the error goes away when I remove that annotation.
Sorry, should have escaped the entities:

Use: £ or £ instead of the pound character and it will be sure to work.
I believe that you can replace the pound sign in the quotes to either £ or £

[edit by Eric Pascarello - escaped the two amps so it would not show up as the character.]
[ April 01, 2008: Message edited by: Eric Pascarello ]