Thanks, UK2… No, really, thanks.
Sarcasm? No, that wasn't sarcasm. I never use that.
You may have noticed most of my websites have been very ugly for the past couple of days. The reason for this is that UK2 have seemingly used some very underhanded business tactics to shut down SimpleCDN, their customer and my Content Delivery Network responsible for delivering much of my static content.
Over the next day or so all of our sites should be fully back online, albeit temporarily without a CDN while we get set up somewhere else.
I'm not in a position to say that SimpleCDN has done nothing wrong, and that it's all UK2's fault (though that is what outward appearances make it seem). However, I would certainly recommend that anyone thinking about signing up with them watch this situation closely and use its final outcome as a determining factor in whether they are someone you wish to do business with.
You can follow the situation from SimpleCDN's perspective here. If you're as frustrated as me by the situation, I urge you to do as I have done and contact UK2 to voice your opinion.
Related articles
- What's a CDN? Content Delivery Network (marketingtechblog.com)
- SimpleCDN has been effectively kicked off the Internet by its ISPs w/o warning (admin.simplecdn.com)
Asus WL-500g Premium Power Supply Issue
I got up this morning to find the Power and WiFi lights were off on my Asus WL-500g Premium router. It had worked fine for an extended Battlefield: Bad Company 2 session before I went to bed only 5 or 6 hours prior.
I tried resetting it to no avail. When it turned back on it was switching between having all LAN and WAN lights on (no power/wifi) and having the Power/WiFi lights on (no LAN/WAN), every second or so.
So I reset it again and put it into Recovery mode. I manually set the IP on my computer and used a TFTP client to re-upload my firmware image and reflash the device. To my happy surprise, it was working again. Problem solved.
Or not. I logged into the web interface to change my settings, which went great. I enabled Wireless, chose to Apply settings, and my router almost immediately reset and went back to the flashing lights again, just like before.
I tried this several times with the same result every time. Trying to avoid going insane, I ended up leaving for work and doing some investigation. It turns out I am experiencing the exact same symptoms as a lot of other people whose Asus power supplies have gone bad.
The problem is that a round-topped capacitor in the power supply is prone to electrolytic leakage. It essentially "blows its top" and its contents leak out into the unit. When there is not enough left in the capacitor, the PSU is not able to provide stable power.
The reason the problem only occurred after I enabled WiFi is that the router requires about 0.75 amps to boot up, but 1.1 amps is used when WiFi is powered on. It seems my PSU has enough juice to boot up, but as soon as it tries to enable the WiFi, there is simply not enough power and it gets stuck.
I am very hopeful (and 99% sure at this point) that it is the same issue I have as well. I'm going to stop by Radio Shack on the way home and get a replacement capacitor (hopefully 1200uF / 10v or close). I will report back my results afterward. If all goes well, I'll be making that post from my home computer through my working router!
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=7e979c64-79e9-471c-b610-7c7de8d56bb4)
Known Server Issues
No need to inform me about the following server issues, which I apologize in advance for:
- CPU time to generate some WordPress and Drupal pages is surpassing 30 seconds. I am actively looking into this and suspect it's some sort of VPS issue with queue times. This is my top priority.
- Caching is not working properly. No pages are being cached on most of my sites. The pages are still working fine, but because of the above performance issue this can become excruciating. I will have caching working properly by the time the previous issue is resolved.
- Our CDN is not set up correctly. We are using SimpleCDN to host many of our static assets, for performance reasons. It is correctly pulling most assets automatically, however you will notice some pictures in our posts are broken, and other links may not work quite right yet. This is being worked on actively as well.
If you have noticed anything other than these three problems, please let me know so I can work on those as well. I am assessing the entire Digital Empire infrastructure which is why things are taking longer than they otherwise would, but the end result should be a more stable, much faster network.
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=fa601dce-409c-4a65-9e53-9870daf19a09)
Apache issues resolved, sites back online
Yes, finally, with a little help from the cPanel team, I got Apache/PHP to rebuild correctly, and my sites are again working properly!
Of course, I am still working on the slowness issue. Even with page caching, Apache and suPHP thrash the CPU for several seconds the first time a large dynamic page is loaded. I'm trying to narrow this down to either server configuration or a VPS issue.
I have definitely experienced this before on an over-allocated VMware array, so it's a possibility my web server is just on a busy host box. But I'm still tuning things on the server and trying to minimize the load in every way I can, because more often than not these things are caused by configuration oversight or lack of proper tuning (and user naivety perhaps, in my case).
Monitoring System In Place
In a further effort to make all sites and services in the DE network stable and ensure the highest uptime, we have deployed a full-scale monitoring server at an off-site location in Germany.
This means that we can get a full picture of our network and service status at a glance at any time. It also means we are notified when anything goes down, is running slowly, or becomes unresponsive. In addition, we have the ability to automatically bring things back online and resolve some issues without having to wait for me to fix them.
This is a big jump forward for our network infrastructure, and this means we are now running four full network servers in four different datacenters spanning three countries to provide our sites and services under the highest quality and most ideal circumstances.
Let me know what you think! Have you noticed a difference in any of our sites and services? Is there anything you'd like to see, or any changes you'd like made? We'd love to hear from you!
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=eadf5368-2b0c-4633-96c0-25c1802ae4ed)
Is it up or is it down?
I'd like to sincerely apologize if anyone tried to visit my blog, or any other Digital Empire site, and was not able to get to it.
For the past hour or so, I've gotten reports of the site being inaccessible sporadically. I have also seen this myself, as all of my equipment has been down for the past 15 or 20 minutes.
As you can see by this post, the issue seems to be resolved for the moment.
My current service provider is undergoing a management change, and is taking care of some long-standing issues, so hopefully these problems were a result of the firewall finally being set up properly.
My servers already ran great, but it was taking weeks to get any request through. I firmly hope (and expect) to receive better service going forward with the new owners.