kid_drew

I'm stupid. One of the apps is a wordpress blog which was still configured to be at a subdirectory of my webroot. This is solved.

I'm just about to scream. This makes absolutely no sense to me. Hopefully someone else has some insight into this. I've set up virtual hosts plenty of times in apache/linux....no problems. I'm trying to do the same thing in Windows and I'm getting all kinds of problems. NameVirtualHost *:80 <VirtualHost *:80> ServerName www.example1.com DocumentRoot i:\webroot </VirtualHost> <VirtualHost *:80> ServerName www.example2.com DocumentRoot i:\webroot\example2 </VirtualHost> This is my config verbatim with the domains replaced with example1/example2. I get no error messages, but both www.example1.com and www.example2.com direct to the i:\webroot directory. Like I said....this makes absolutely no sense. Does anyone have any ideas?

After more research on my end, I'm convinced I need to do a rewrite and not a VirtualHost. Thanks for the reply texerasmo, but I'm not sure what to take out of what you wrote. I know how VirtualHosts are set up in terms of mapping multiple domain names to one box, but I specifically want this URL mapping: http://www.example.com/site1 should look for code in i:\webroot\drupal http://www.example.com/site2 should ALSO look for code in i:\webroot\drupal Neither of these apps have unique domain names. The URLs are exactly as I have typed them (with one domain replacing example.com of course). I could very well be wrong, but without a more specific example from you, I'm not convinced. -Drew

Hey everybody, So I'm trying to set up a WinXP multi-site drupal box, which allows you to create multiple sites using the same code base and different databases. I can get the databases set up correctly, but my problem is redirecting with .htaccess. I'm a PHP programmer, not a system admin, and I don't know all that much about mod_rewrite. Anyway, here's the situation. Webroot is located at i:\webroot and drupal is at i:\webroot\drupal. I need multiple subdirectories to pull the code from the i:\webroot\drupal webroot directory. So, to illustrate: http://www.example.com/app1 will pull from i:\webroot\drupal http://www.example.com/app2 will also pull from i:\webroot\drupal http://www.example.com/app3 is not a drupal app, so it will pull from i:\webroot\app3 as you would expect it to. Do I do that using mod_rewrite, VirtualHost, VirtualDocumentRoot? I'm a bit lost with all the options. Thanks -Drew

Hey guys, I have two large apps on physically separate mysql databases, and I want to combine the two servers into one to save money. The last time I tried this, the two apps competed for resources to the point that a half second query took 30 seconds. But on two separate servers, they aren't using anywhere near the capacity of one server. So I must be missing a setting or something that would allow two apps with completely disjoint queries to run on the same machine. It has to be some kind of a caching issue, I'm assuming. Can anyone give me any advice? Also, feel free to ask me any clarification questions. -Drew

CentOS I'm not sure about the processor, actually. I didn't set up the server. I know it's 64 bit, so greater than 4GB is possible.

We have a MySQL database backing up two applications getting well over 3M pageviews a day. Our database is getting hammered. The issue seems to be memory. Is 4GB reasonable for a site of this size?

Hey guys, I had a very active database and the disk ran out of space. Now, the database is down and giving me a 2002 error. The process isn't even showing up in the process list, and restarting it fails. I don't even know where to start. Can anyone give me any ideas?

I've never used INSERT DELAYED actually, so I'm not sure if I can afford it or not. As I understand it, MySQL will bundle a bunch of inserts together to throw in at one time. Seems that might actually speed things up.

A few million a day.

select * from trans where user_id=xxxxxx is very typical. Sometimes I query for the transaction type too. So select * from trans where user_id=xxxxxx and trans_type=xx.

Do you mean when I run them as root? Only select/where queries. select * from trans where....

Only inserts.

Hey everybody, I have a site that is pushing 1M users with ~2M pageviews per day. It's built on a form of artificial currency and involves a lot of transactions of this artificial currency. I log every monetary transaction into one table that, as you can imagine, is rather large (32M rows). I move old transactions daily into a non-live table to keep the size manageable, but here's my issue. I intentionally didn't build indexes in this table because it is only ever read from the MySQL monitor. It is only written to by users, so read does not need to be quick. The problem is that when I do need to read it (maybe twice a day) it takes FOREVER to query because it has to go through every row. So I'm wondering two things: 1. Should this table be InnoDB or MyISAM? 2. How much of a performance hit will I take if I build indexes on the table?

Hey guys, I'm trying to code up a simple game where the user passes from one page to the next through a series of links. How can I prevent a user from cheating by typing in the URL of a page that is not in the logical flow of the game? Some kind of cookie trick, I'm sure, but I can't seem to wrap my head around it. -Drew

For those of you who run live active sites, how do you handle code development/testing without potentially breaking a piece of code that your users are interfacing with? Do you have a whole separate development server and then pass tested code to the live server? How do you redirect requests to an "updating server" page while you are making software upgrades?

Yeah, that isn't working. It seems like it grabs the image that was there when the script started loading.....or something. I'm totally confused by this.

Hey guys, I have a Java applet that, on destruction, writes a postscript file to my server, and my server then converts that ps file to png to view in the next page that loads. The problem is that the write from the java applet to the server takes longer than it takes for the page to load, and the conversion happens on the old file instead of the new one. So the page has to get reloaded to see the new version of the png. Is there a way to somehow delay the loading of the next PHP page? -Drew

I already checked it and, like I mentioned before, they're working just fine everywhere else. It isn't a permissions issue.

I have some code to set and retrieve cookies from the browser, and it's all working in Firefox, but not in Internet Explorer. The cookie is not even showing up in my directory. Cookies are working correctly everywhere else, so it isn't a permissions issue. Is there a difference between the way cookies are handled in IE vs FF?

Does anyone have a simple example script that will take an image filename as a query string argument and output the image without using HTML? So I don't want any <img src=""> calls in the PHP script. An example to extract an image from a database blob would probably also be helpful. -Drew

Give me an example of an ajax script that would do this. Note that this image is not the only thing I'm displaying on the page and it will have to be embedded in html. Is that possible?

Hi guys, Ok, so here's the setup. I have a drawing program written in Java. The user can draw whatever they want and then save their progress to an eps file (vector). I convert the eps file to gif with ghostscript (called via a perl script) to be able to view a thumbnail in the browser. If the user navigates away from the Java applet, the drawing will automatically save and convert. The problem is if a user navigates straight from the drawing page to the thumbnail page, sometimes the conversion does not happen in time and the thumbnail comes up as the previous version (cached). I have tried adding a sleep statement to the perl script, which doesn't work because it is a background process. I have also tried adding a sleep statement to the top of my thumbnail script, and that doesn't seem to work either. Anyone have any ideas?

That's using html to display the image, and that isn't working.

Headers don't seem to be doing the trick. I read somewhere that the header cache calls are simply "suggestions" and don't actually force anything. How ridiculous is it that this is so diffucult? I've even tried the trick of making the image call unique like this: <img src="foo.gif?<?php echo rand(1,3000)?>"> And that doesn't even work. How do I make php display the image?