So, still using hashes for ajax navigation? Yeah me too. Well that isn’t really necessary any more. Since HTML5 is really mainstream these days it’s time to start taking advantage of window.history.pushState API, caniuse.com seems to agree with that. As you can see there is support for the latest version of all major browsers. For browsers that do not support the pushState API yet there is a beautiful JS library with fallback support to using hashes.
For the last few days I have been wondering why I couldn’t get the number of HTTPD processes down to the level I configured in httpd.conf . I looked every where, tried to figure out if I was using prefork.c or worker.c tried multiple values, tested the server under load using Apache Bench and so on and on.
Still after multiple hours of testing, tweaking and changing values nothing seemd to change. Apache keeped spawining 25 to 30 processes which whould use 3 to 4 % memory each. Yes you can do the math that’s 90% just after 10 minutes of uptime. When using OpenVZ or Plesk it’s just waiting for a process to get killed becuase your VPS is hitting it’s memory limit.
So, I looked further, enabled mod_info (great tool for debugging problems with Apache) and there it was. After hours of searching I was looking at the wrong config file. It seems that Plesk installs a file called swtun.conf in /etc/httpd/conf.d/ . This file gets loaded after the prefork.c settings in httpd.conf are set. So, logically it overrules the values in httpd.conf rendering it completly useless. Would have been handy if that was somehow documented somewhere …
Anyway, the values set in swtune.conf are pretty high for a small VPS. Adjusting them to about half of their defaults seems reasonable on non-highend VPSes.
Thanks again for the quest Plesk, not for the first time you’ve taken a few of my evenings away searching for a tiny little configuration value.
There is this big opensource blogging platform. It’s commonly used as a complete CMS solution to many websites. Even this site uses it. Yes, I’m talking about WordPress. Currently WordPress is hosting over 70 million websites. When you are talking about an ammount that massive you would say the source code by now woud look all tidy and neat. Well, I suggest you brace yourself.
if ( !isset( $HTTP_RAW_POST_DATA ) ) {
$HTTP_RAW_POST_DATA = file_get_contents( 'php://input' );
}
if ( isset($HTTP_RAW_POST_DATA) )
$HTTP_RAW_POST_DATA = trim($HTTP_RAW_POST_DATA);
Two things that stand out from the crowd here. The first is the space after $HTTP_RAW_POST_DATA in the first if-statement before the closing parenthesis. Of course, you have a lot of developers working on this. Different people have different habits. Things like this are impossible to prevent. At least, you would think. But wait, aren’t there solutions to this problem? Of course there are! PHP CodeSniffer for example. This tool analyses PHP-code based on a set of pre or user defined rules and gives back a report containing remarks about codestyle violations. It even chains with SVN’s pre commit hook, so rogue developers won’t be able to commit their code!
Second thing is this: What the hell is the second if-statement about? After the first one there is no way that $HTTP_RAW_POST_DATA is not set. No way. Just not. It is not possible that the if-statement evaluates to false. It is always true. Always. If $HTTP_RAW_POST_DATA was not set before the first if-statement then it surely was after the first if-statement. That’s another wasted CPU cycle (on 70 million blogs) right there. Well, at least that didn’t type a space here.
This is just one of the many examples of strange code in the WordPress source code. Don’t get me wrong, I love the application. It’s fast, easy to work with and handy to use. But there are a few things that are just a bit weird. Maybe it’s about time to shake of the weirdness when you are serving 70 million website. Thats all.
Searching the web for ways to remove the mandatory validation of the telephone field in Magento I found quite a lot results non of which actually worked with Magento ver. 1.4.2.0 or up. The biggest problem seems to be to completely remove the validation and mandatory state of the telephone field throughout the whole Magento store. Some solutions fixed it in the checkout, some fixed the client side validation but not a single solution fixed it all.
The final solution contains three steps which are:
If they not already exist within your theme copy the following files from the base/default theme to your theme:
In each of these files, look for the <input> that defines the telephone field and remove the required class from the label, remove the required-attribute class from the input and make sure to remove the *.
Copy the file app/code/core/Mage/Customer/Model/Address/Abstract.php to app/code/local/Mage/Customer/Model/Address/Abstract.php . This ensures that upgrades will not break your modifications. Now open the file and look for the part where it validates the telephone field, it should look something like this:
if (!Zend_Validate::is($this->getTelephone(), 'NotEmpty')) {
$errors[] = $helper->__('Please enter the telephone number.');
}
You can either completely remove this part or comment it out by placing it between /* and */ .
Except for the validation we just removed in the abstract class Magento keeps a validation definition in it’s EAV tables. In the table customer_eav_attribute look for the row with id attribute_id 30 and clear the contents of the column validation_rules. Next in the table eav_attribute loo for the row with attribute_id 30 and make sure that the value of the column is_required is 0.
Completing the above steps will effectively clear the validation and the mandatory state of the Telephone field.
Thanks for this tutorial!
I have only one problem: there is no attribute_id 30 in my customer_eav_attribute table. How can I identify the correct row?
Regards
Matt
I am using 1.5.1.0 and found the table eav_attribute had it as attribute code 17. There was a required field that I changed from a 1 to a 0 and checkout worked great!
You can see app/code/core/Mage/Customer/sql/customer_setup/mysql4-data-upgrade-1.4.0.0.7-1.4.0.0.8.php or other mysql.php file
find ‘sort_order’, and
In datebase table ‘customer_eav_attribute’ , you can search ‘attribute_id’ by ‘sort_order’
I have performed all these steps and I am still getting a pop-up that states “Please enter the telephone number.” Can someone please help me with this. I am running Magento 1.4.
Attempting this in Magento 1.7.0.0 is a straight *fail* I’ve been pestering every possible sollution i’ve found online and none seem to remove the input boxes and validation.
Anybody with a good sollution for the new magento?
Hi.
I tried it on 1.5.0 and ist not working.
Thanks for help
@john:
change in file /app/code/core/Mage/XmlConnect/Block/Customer/Address/Form.php
$contactInfoFieldset->addField(‘telephone’, ‘text’, array(
‘label’ => $this->__(‘Telephone’),
‘required’ => ‘true’,
‘value’ => isset($telephone) ? $telephone : ”
));
to
$contactInfoFieldset->addField(‘telephone’, ‘text’, array(
‘label’ => $this->__(‘Telephone’),
‘value’ => isset($telephone) ? $telephone : ”
));
this work for me after i doing step 1 & 2 of this tutorial. Magento 1.7.
Hi, i’ve done all the described steps (Magento 1,7) and as some of you i still get the popup ‘Please enter the telephone number’. The same with the region, which isn’t necessary in germany. Any ideas or solutions to that?
Thomas
Addition to my comment from a few days before: on the system I made the above described changes I had the compilation acvtivetd. After changing the file template/persitent/checkout/onepage/billing.phtml and recompiling Magento (System > Tools) everything works fine.
Every server-admin is familiar with the situation: Your servers load is flaky but you don’t see any processes eating your CPU capacity, at least not constantly. What is going on? Probably the first thing you do is Google. And yes, you will find lot’s of articles explaining you that what the number means. They will all tell you that it represents the average number of processes in the run-queue over a certain period of time.
You will learn that the health of the number is determined by the number of cores in your system. For every core the load should be 1 or lower in order to considered to be healthy. So when you have a quad-core processor with a load of 3 that should not necessarily pose to be a problem. I am explicitly putting in the word necessarily in the previous sentence because this is where many articles make a mistake. Many resources on the web tell you there is nothing to worry about as long as your load average is lower then the number of cores in your server. I oppose to this statement.
As you might know processes have a number of different states which they can be in. One of those states is IO/WAIT. This means that the proces is waiting for access to an IO-device (for example the harddrive) . During this state the proces is considered to be in-proces thus being in the run-queue and therefore representing 1 in the load average of your server. As you can see, in this situation the number of cores in the server is completely irrelevant because even a load of 2 can mean your quad-core server is slowing down because of the lack of speed in it’s IO-device(s).
The previous example can be recognized when the output of the command top is being watched. When you see a high load but low CPU usage you can almost certainly say that IO is causing the number of processes in the run-queue to be high.
Hi ,
I have a linux box with 8 GB memory with 4 Processors ( Quad-Core AMD Opteron(tm) ). When checked with cpuinfo I found each processor has 4 cores. So which means that my box has a total of 16 cpu cores. Normally for a single core machine, a load average of 1.0 would mean that the processor is 100% used. Does this mean my server load can go as high as 16. As of now the monitoring is set in such a way that we get an alert when the load goes beyond 5.0 . Pasting the /proc/cpuinfo from the server.
processor : 0
model name : Quad-Core AMD Opteron(tm) Processor 2374 HE
cpu cores : 4
processor : 1
model name : Quad-Core AMD Opteron(tm) Processor 2374 HE
cpu cores : 4
processor : 2
model name : Quad-Core AMD Opteron(tm) Processor 2374 HE
cpu cores : 4
processor : 3
model name : Quad-Core AMD Opteron(tm) Processor 2374 HE
cpu cores : 4
Will my server be able to withstand more load with 16 cores or is it something related to memory.
Thanks in advance!
When you are trying to use the script /var/lib/heartbeat/hb_standby or the script var/lib/heartbeat/hb_takeover you might experience that nothing is happening. In that case you are likely to see the following message in your log (which probably is /var/log/messages) : info: Standby mode only implemented when nice_failback on .
I know what you are thinking, just add nice_failback on to your /etc/ha.d/ha.cf configuration file and everything will be jolly good. Please go ahead, you will notice this doesn’t solve the problem. This is because of the fact that the authors of heartbeat changed the configuration parameter but failed to change the log-message. The correct addition to your /etc/ha.d/ha.cf is: auto_failback off. Just to be sure everything keeps working I would recommend to leave the nice_failback on line in place in case you might ever down grade to a version of heartbeat that is pre 2.x .
Above information was found on the linux-ha mailing-list
This is a quick not on how to move your WordPress site to a different domain. Usually this is no a task performed often, however as a developer you will have the need to do so on a regular bases when publishing a site.
This post is very quick writeup on how to do so. First step, obviously: fix the settings in wp-config.php . Next run the following queries:
UPDATE wp_options SET option_value = replace(option_value, 'http://www.old-domain.com', 'http://www.new-domain.com') WHERE option_name = 'home' OR option_name = 'siteurl';
UPDATE wp_posts SET guid = replace(guid, 'http://www.old-domain.com','http://www.new-domain.com');
UPDATE wp_posts SET post_content = replace(post_content, 'http://www.old-domain.com', 'http://www.new-domain.com');
That should do the trick. Off course change new-domain.com and old-domain.com to the appropriate values.
This post is more a note to myself then it is a blog post. That is not a problem because in fact this whole blog is based on that very principal. Today, I was installing CentOS 5.5 onto a particularly large volume and I encountered some errors during the creation of the partition scheme concerning the placement of the /boot patition.
It seems to be that the version of Grub that is embedded into CentOS 5.5 does not have support for booting from a volume larger then 2 TB. Having encountered this problem before with CentOS 5.4 I was struggling to remember the solution. After unsuccessfully messing around with some commonly mentioned results I gave up.
The most easy solution to this problem: Create separate, smaller, volume for your /boot partition in the configuration of you storage controller. Yes, I know that this solution is only applicable to you who have RAID controller available in their machine. But common, every server should!
Trying to install some extensions today on a Magento 1.4.1.0 installation I experienced the following error using Magento Connect:
pear/Console_Getopt requires PEAR Installer (version >= 1.9.1), installed version is 1.7.1
If you experience errors similar errors while using Magento Connect or Magento’s ./pear script try upgrading your PEAR installation using PEAR. Login in via SSH to the server running your Magento installation. Change your working-directory to the root of the Magento installation and run the following two commands:
./pear channel-update pear.php.net./pear upgrade --force PEAR
Doing that should upgrade your PEAR installation to the latest version. After that you should be able to install extensions using Magento connect as always.
I am a enthusiast user of Cacti because it gives me a good insight in the health and functioning of my servers. Aside from that it’s a great tool to discover the cause of a problem when an incident occurs. We are all familiar with the extreme flexibility of Cacti and it’s data and graph templates but today I discovered a feature which I hadn’t seen before.
The Cacti is the graph exportation feature to be precise. It gives you the opportunity to export all or a subset of your graphs periodically using various methods. It can export your graphs via FTP, SFTP or locally on the server that is running your Cacti daemon. Of course you can specify authentication and path information for your export but even better you can control a lot more! You can arrange everything from the timing of the exports until the size of the exported graphs.
Possible applications of this feature are easy to imagine for example automatic offsite access to your graphs for when your Cacti server becomes unreachable and you need to backtrack the problem. Another application can be the providing of authentication free access to your graph tree for example to your internal network.
You can find the graph exportation settings when logging into your Cacti administration panel using an account with enough rights to administer settings on your Cacti installation. After logging in you navigate to the console tab, from the left menu choose Settings under the section Configuration. On the page that appears select the tab Graph Export.
Edit:
Be careful with the export interval you choose. Cacti has several options to create a timing scheme, for example hourly, daily or per poller-run. Be aware of the fact that when you have large number of graphs the exportation can be quite a burden on the load of your servers I/O!
Erma, and 6 others are discussing. 
Vinod is discussing.
Hi!
Thanks for this, I had been looking for a way to remove the requirement from the admin for a along time now after having managed to get the frontend working without phone one my own.
It seem that the db change for “customer_eav_attribute” is not required anymore (magento v1.4.1.1), matter of fact, that table doesn’t have a “validation_rules” column. Also in the “eav_attribute” table, the telephone entry was row 29 for me, but I guess one can simply search for phone to find it..