Welcome to the Q&A Forum

You likely have a .htaccess issue causing a rewrite error.   You may want to examine or replace your .htaccess with a default.   Also, I've seen some plugins cause this error.

What is happening is this:

http://www.atozqualityfencing.com/newsite

is sent to:

http://www.atozqualityfencing.com/newsite/

Note the trailing slash.

But that page is returning a 404 error.

If I go to

http://www.atozqualityfencing.com/newsite/index.php it redirects to

http://www.atozqualityfencing.com/newsite/

So there is likely something wrong in the redirect rules.   I would try disabling all plugins. If that fails, compare the current htaccess to a default one and remove any modifications.

.

From a system operations standpoint, here's a couple of tips about rewrite rules, especially if you are doing a mass rewrite.

When to avoid htaccess?

If you have 100's of rewrites do not use htaccess to rewrite each one individually.  The problem is that every HTTP request passing through has to be matched against the rules.  We recently had a customer's developers put in ~8000 rewrite rules - one per page. The result - the server queue filled up as every HTTP request was evaluated against 8000+ conditions.

If you have more than few 100 rewrites, consider either looking for a way to use regular expressions to dynamically rewrite them or use tools built into your CMS software.

Check your headers.

When you have completed your 301 efforts, you can use curl at the command line:

curl -I http://www.domain.com/

or use a tool like Redbot.org to check your headers to assure the 301 is correctly configured.

Trailing Slashes

I am not sure how or if trailing slashes impact SEO nowadays,  but I noticed in your example URLs, they did not include a trailing slash.  I have seen some CMS systems rewrite

http://www.domain.com/page -> http://www.domain.com/page/

This can cause a double 301.  This is easy to miss.   I've found this to be more common when changing CMS systems but something to check.

You may want to check out: http://moz.com/learn/seo/canonicalization and the trailing slash comments there.  Since you are dealing with a rewrite, you may want to fix that now.

Yes, attackers use code injection techniques to hijack search results. They in fact do use some randomization techniques to make it harder to detect.  Conducting a security scan using a tool like Sucuri.net could be helpful but manually reviewing your .htaccess and code may be necessary.

A 904 error is a timeout.  As such this could be on either end of the crawl or in the middle.  If you have You should have remote monitoring to measure server uptime and performance.  Check the time of the 904 errors with your monitoring.  Also review server logs for that period for other errors.

In some cases we have seen security systems block crawlers - especially at larger companies where web app firewalls are in use.  The result can be that the system thinks there's an attack and could tarpit the crawler.

This is just speculation but something to easily rule out by looking at your server's logs and monitoring for the period in question.

If you use gLTDs as your sub-directories, you can then use WMT to set preferences.

See a further discussion here:

https://support.google.com/webmasters/answer/182192?hl=en#2

From my understanding (as a hosting provider not a SEO), is that increasingly signals coming from on-page or WMT are far more important that server location, IP or other hosting related information.  This makes sense with more and more CDNs.

You may also want to see:

http://googlewebmastercentral.blogspot.com/2013/04/x-default-hreflang-for-international-pages.html

In terms of what to optimize that is up to you, but the important part is to give Google clues as to what pages you want ranked where/how.

A 503 error is a service unavailable error.  I have seen situations where redirects are incorrect and loop.  Depending on the hosting setup, this can trigger various HTTP error codes.

The best way to debug this is by looking at your Apache access logs.  Scan your logs for the 503 errors. Pay attention to the URL being requested as well as the referring URL.

Very likely, there's some looping process and in cases when Apache runs on FastCGI, you can get a 503 error due to too many processes being triggered.

Also, due to how WP handles 404's, I've seen many plugins mask underlying causes. So if you have any plugins that impact error handling, you may need to remove those while debugging.

You can also use http://www.redbot.org/ to check the headers for any page that should be redirected. That tool should return a Location header with a URL.  Visit that Location URL in your browser and make sure it resolves.

The goal here is to try to replicate the behavior. Once you can replicate the behavior, dig into your redirect/rewrite rules and examine the logic to determine why you are seeing the loops or failures.

This could just be a WMT error.  I've seen this from time to time. If it persists for more than 24 hours, then you may want to try digging into it further.

Have you tried fetching as Googlebot

In doing server security work, I have seen some hijacking attempts that will send 301 responses to Googlebot.  In some cases, these scripts are improperly deployed resulting in 404 errors or 500 errors.  So I would first fetch the site as Googlebot.

Then examine your server's access logs for hits from Googlebot.  Confirm that it is indeed Googlebot.  You can verify Googlebot IP ranges with reverse DNS:

https://support.google.com/webmasters/answer/80553?hl=en

If there's no foul play, then check for any browser-based detections routines.  With the increasing use of mobile, I have seen some code fail to handle bots correctly.  The behavior is usually an error not a 404. However, many CMS systemincorrectly send 404 errors for a variety of errors and not just true not found status codes.

I work with a number of high traffic sites (TB's of data each day, 10's millions page views/month).  With many of these sites, we have offloaded static content to either dedicated static content servers (typically cloud based so we can scale up and down) or to content deliver networks.  I've not had anyone report any SEO impact.

In contrast, they often see user engagement (page views/user), repeat visitors, and other traffic metrics improve.  Users like fast sites.  Also, Google apparently likes fast sites too, so while I've not seen it, you could actually get a boost in your SERPs due to faster loading pages.

If you break down a modern web page, you will find numerous elements required. Dozens of images, CSS, javascript as well as the page itself.  All of these items require a request to the web server.

With some graphic intensive sites, I've seen as much as 95% of all web server requests (HTTP requests) be attributable to static content.  By moving these HTTP requests to other systems, you free your primary server to handle the application.  This provides a better user experience and improves scalability.

Content Delivery Networks

I do not use Amazon's Web Services so I do not know specifically what they offer.  But here are two CDN's Ihave used with good success:

Internap:

http://www.internap.com/cdn-services-content-delivery-network/

Edgecast:

http://www.edgecast.com/

One method I look for is called "origin pull." With this method, you do not have to upload files to the CDN. The CDN will fetch them automatically from your site as needed.  I found this is much easier to manage on sites that have frequent content updates.

While I've seen a lot of news about Google cleaning up content farms, link farms, and similar spam, I've also seen some companies start ranking very well for niche terms using these same practices.

Question:

Does Google completely discount links from content farms and similar sites or simply give them low value?

Observation:

I've seen a company start ranking well (top 3) for several terms when they used be on page 2.

When I looked at their links, they are from article farms, directories, do-follow blogs and similar low-vale sources.  Relative to others, they have about 10x the volume of links with the precise anchor text they are targeting.

I wonder in absence of other information that these spammy links still count for something.  Given the low competition for the term, this is enough to boost their rank.

Just thoughts some thoughts as we are working on long-tail strategies for some key terms.