Monthly Archives: March 2011

I switched from GoDaddy to Just Host in mid 2010. GoDaddy had great uptime and excellent customer support, but I preferred the cPanel interface available at Just Host to the proprietary admin area on GoDaddy. I also liked being able to host all of my sites under a single, inexpensive plan at Just Host, rather than paying for separate hosting for each domain at GoDaddy (which I later discovered was an option).

But, after a few problems here and there with Just Host, I thought it was a good idea to keep track of issues that lead to support calls. By early 2012, I had decided not to renew my hosting contract and was looking for alternatives. HostGator seemed to be pretty decent, but I was also curious about going it alone with an Amazon EC2 instance.

Below is a record of the support tickets logged with JustHost for what I think are problems that a competent host would be able to resolve without requiring my attention.

Slowness

[#2916833] MySQL seems extremely slow over the last two days
Sunday, August 28, 2011

Hello there,

It seems like something is really slowing down the MySQL server that I use on http://alephstudios.com.

Basically, the same operation takes the server over 5 times longer than on my low-end laptop. The server is usually far, far faster than this. I’m running into PHP max execution, script timeout type problems with what should be very quick processes.

Would someone please look into this?

I’ve put more details up on my site at //ardamis.com/2011/08/28/justhost-com-what-is-wrong-with-your-mysql-servers/

Thank you,
-Oliver

Downtime

The downtime at Just Host has become increasingly frustrating. I’ve now noticed my sites being down four times in 2011 and have logged support tickets.

[#2743795] Site unreachable…
Saturday, July 2, 2011
http://www.webpagetest.org/result/110702_8P_Z5WT/

[#2419040] Sites down again
Wednesday, March 16, 2011

[#2269339] site timing out
Thursday, January 27, 2011

[#2219224] Site appears to be down
Tuesday, January 11, 2011

Just Host is pretty good about quickly getting things running again, but I wondered how much downtime my sites are experiencing that I wasn’t aware of. In August, 2011, I set up a free account at pingdom to monitor for any further downtime.

Error messages

Shortly after I became a Just Host customer, I discovered a misconfiguration on the server that was preventing my sites from saving PHP session variables. This was quickly resolved.

[#1836474] Cannot save session data to /tmp
Monday, August 23, 2010

DNS problems after server migration

On Wed, Apr 4, 2012 at 12:07 AM, I received the following email from JustHost:

We’re pleased to inform you that we will begin upgrading your account, and all associated subdomains, to a new hosting server within the next 48 hours.

This upgrade requires that we take your websites offline for a short period of time to complete the transfer. Anticipated downtime is between 1-2 hours, though we’ll take every step possible to restore full service as quickly as possible.

In order to facilitate the smoothest transition possible, we’ll be updating internal DNS entries automatically during the transfer, and will make the website accessible at the previous IP address for approximately one week after the upgrade completes.

Should you be managing your own DNS (you know who you are), your new IP address will be available in your account once the upgrade completes. As part of our seamless upgrade process, your website will continue to be available at the previous IP address for approximately one week. However, we suggest updating your DNS entries immediately to ensure proper functionality for your hosted domain(s).

Finally, customers in the European Union will begin paying VAT tax on any future purchases, as required by EU law. VAT will not be charged on any past purchases.

Please do not hesitate to contact our Support team if you have any questions: (888) 755-7585

Sincerely,

Just Host Support

That sounded pretty good.

On Wed, Apr 11, 2012 at 12:02 AM, I started getting alerts from Pingdom.com that my site was going up and down. It eventually went down 3 times for a total of one hour.

On Wed, Apr 11, 2012 at 7:24 AM (7 days, or 175 hours, after their notification email), I received the following email from Just Host Support:

The upgrade of your account, and all associated subdomains, to a new server has completed successfully. Your website(s) are now hosted at a new location, with the following IP address; please keep these for your records.

If you are managing your own DNS entries for your domains, not common, please update them now.

173.254.6.73

Like the previous email mentioned, your website is still available at the old IP address, and will remain available for approximately one week.

While no data loss was encountered during the upgrade, we suggest you browse through all your hosted domain(s) to verify that they are functioning properly. Additionally, note that this process required a change in server timezones, from Central Daylight Time (GMT -5) to Mountain Daylight Time (GMT -6). Any automated tasks you have may require adjustment to trigger them at the desired intervals.

Lastly, this upgrade included several software upgrades and changes. The most noteworthy being a new Domain Manager, a new Billing interface, and the more robust SimpleScripts, which replaces Fantastico. You can directly import your Fantastico installed software to SimpleScripts, which will manage the installations and upgrades as Fantastico did.

If you have any questions about your account post-upgrade, or experience any trouble at all, please do not hesitate to contact our Support Department.

Just Host Support

OK, so it took them a bit longer than they originally anticipated. That’s fine.

What wasn’t fine was that my sites all failed to resolve. Instead of my sites, I got a JustHost ad-supported placeholder page. Later that afternoon, I called JustHost support and was told that while the file migration was complete, there was a problem with the DNS (as in they pointed it to the wrong server) and we were now waiting for the correct DNS entries to propogate. After a few hours of being completely horrified that my sites were replaced with a hideous landing page, everything began working normally. Or so I thought.

On Sun, Apr 22, 2012 at 5:45 PM, I logged a support ticket because the password protection on one of my subdomains (which had worked fine before the server migration – see below) stopped working. I was still prompted to enter a username and password, but instead of then taking me to the site, the server was throwing a 500 error. I tried removing the password part of my .htaccess file, and the site worked fine. I put the password part back and tried changing the password for the subdomain, but it continued to throw the 500 error.

My email to Just Host support:

Hey there, folks.

I have set an directory password on one of my subdomains – rankings.alephstudios.com – and added (what I believe are the necessary lines to my .htaccess file). I’m getting the expected password prompt, but after I enter the password, I get a 500 internal server errror.

Here are the log entries for this event (my IP is 69.211.123.27):

[Sun Apr 22 16:11:49 2012] [error] [client 69.211.123.27] (13)Permission denied: Could not open password file: /home/alephst1/.htpasswds/public_html/rankings/passwd
[Sun Apr 22 16:11:49 2012] [error] [client 69.211.123.27] (13)Permission denied: Could not open password file: /home/alephst1/.htpasswds/public_html/rankings/passwd

It looks like a permissions or ownership problem on the htpasswd file or parent directory.

The password protection is primarily to keep spiders out of that directory, so I can give you guys the username/password if you need it.

You can email me at xxxxx with questions.

P.S. The Apache log also shows tons of these errors, not from my connection:

[Sun Apr 22 16:43:49 2012] [error] [client 66.249.73.205] Request exceeded the limit of 10 internal redirects due to probable configuration error. Use ‘LimitInternalRecursion’ to increase the limit if necessary. Use ‘LogLevel debug’ to get a backtrace.
[Sun Apr 22 16:43:50 2012] [warn] RewriteOptions: MaxRedirects option has been removed in favor of the global LimitInternalRecursion directive and will be ignored.

I figured that was a pretty good email. I was able to clearly describe the problem behavior, supply some evidence of a record of the problem behavior, and propose some logical places to begin troubleshooting the behavior. As someone who has worked in technical support in the past, I’m pretty careful about how I interact with reps.

This is the response I got:

Hello,

The log file shows errors or issues for the entire server your account is on. The only ones that affect you are the ones associated with your connection, ie IP address, and are typically in red.

A 500 internal server error can be displayed for multiple different reasons. It can be thrown when there is an incorrect redirect or rewrite rule stated in the .htaccess file. You can troubleshoot this issue by simply renaming the .htaccess file, effectively deactivating it. After renaming it, try reloading the page that was giving the 500 error, (you may need to clear your cache). If you need help renaming the .htaccess file, please follow these steps:

Login to your Cpanel
Scroll down to Files
Click on File Manager
Choose Web Root, also make sure that you also check the box to show hidden files.
Click Go
On the right side, look for the .htaccess file. Right click on it, and click rename file (rename it to something like .htaccess.old)

It can also be caused by improper file/folder permissions Typically folder should have permissions of 755 and files should be 644 You can check and change file/folder permissions using the File Manager in the cpanel or using an FTP Client such as Filezilla.

If this does not resolve the issue, please check your cPanel -> Error Logs. Please Note: The MAIN error_log is a shared log meaning that errors from all websites hosted on the server will be displayed as well as those for your website, be sure to reference your IP when searching this log. If you are unsure of your WAN IP address please visit http://whatismyip.com

It can also be caused by a plugin or script that is trying to do something that the server doesn’t allow; checking the error logs as stated above should identify the script or file that is causing the issue.

Thank you,
XXX
Level I Tech Support Engineer

Completely not satisfactory as a response, even from a Level 1 Tech. The errors I quoted in my email were in red. I took pains to provide useful information and got a stock response back without any apparent effort.

So, I replied with this:

I logged a support ticket because the password protection on one of my subdomains (which had worked fine before my sites’ recent server migration) stopped working.

I was still prompted to enter a username and password, but instead of then taking me to the site, the server was throwing a 500 error. I tried removing the password part of my .htaccess file, and the site worked fine. I put the password part back and tried changing the password for the subdomain, but it continued to throw the 500 error.

I looked at the server logs and, as I pointed out in my original ticket, found entries from my IP address indicating a Permission denied error for my htpasswd file.

Would someone please confirm that the files and directories have the correct ownership to allow me to use password protection on my directories?

Thanks,