New "Static" Site with 302s
-
Hey all,
Came across a bit of an interesting challenge recently, one that I was hoping some of you might have had experience with!
We're currently in the process of a website rebuild, for which I'm really excited. The new site is using Markdown to create an entirely static site. Load-times are fantastic, and the code is clean. Life is good, apart from the 302s.
One of the weird quirks I've realized is that with oldschool, non-server-generated page content is that every page of the site is an Index.html file in a directory. The resulting in a www.website.com/page-title will 302 to www.website.com/page-title/.
My solution off the bat has been to just be super diligent and try to stay on top of the link profile and send lots of helpful emails to the staff reminding them about how to build links, but I know that even the best laid plans often fail.
Has anyone had a similar challenge with a static site and found a way to overcome it?
-
Wow. I wasn't expecting such a detailed and awesome answer Danny. Thanks so much, I'm in the process of migrating away from S3 anyways (for other reasons) though you're right in that I'm going to miss the cost & load times.
I'm using Middleman for now, though the technical part of my brain is indeed interested in how you're going to accomplish the Jekyll solution. I'll look out for your post!
And thanks for the tip on my site. Another thing to add to the list
Arun
-
Hey Arun,
Thanks for posting! I was beginning to think that I was the only Inbound guy anywhere that had to deal with this kind of issue
Yup, I created the same bug with redirect loops trying to get around the slash issue. The problem is that S3 doesn't consider the slash as part of the rewrite data unless something comes after it.
Ultimately, my number one suggestion would be to go with a different service that allows you to install a Server App like Nginx or Apache. Others have agreed that redirections set up through a server app are the way that they feel the most comfortable that link equity is being passed.
If you're dead-set on S3, which I would understand as the load times are crazy-awesome-insane, I may have a solution for you soon. Our dev team is working on a script for Jekyll + S3 sites that will essentially create extension-less files (i.e. example.com/contact) that contain meta refresh + rel canon.
The script will use a list of desired redirections + rules that is structured the same way an htaccess file would be. I can't speak to how it will get past S3's default 302ing yet, but I know that it will use CURL. Look for a YouMoz post soon from me!
Anyways, I hope my notes here help! I'm gonna try and make that post soon after the script is created. Just as a last note, in taking a look at your site I noticed that a lot of the internal links on your homepage don't have the trailing slash in them. I would definitely start there and add those slashes, and perform a "submit page + linked page" to Webmaster Tools after!
-
Hi Danny-
I've got the exact same issue (static site on S3 redirecting with 302s), and surprisingly can't find a lot of information out there. If I do a S3 metadata based redirect from (for example) /blog to /blog/ I just end up in a redirect loop.
I checked out your site and it still looks like you're working on it. Did you end up figuring anything out? If there's any way that I can help get to a solution I'd be happy to spend some time on it.
Thanks!
Arun
-
Thanks for the reply David!
Yup, I think that this has just been a case of wrapping my head around a new way of doing things (i.e. redirections in the AWS bucket config rather than using .htdocs). Static sites are a crazy combination of complicated and simple!
Thanks! We're using Jekyll somewhat, although we've had issues with the image hosting. I've actually had better results using the local github client + "Mou", a local Markdown editor.
-
Nice! (for speed at least)
I would show your team some examples of external URLs pointing at the non trailing slash versions of your pages and explain the downside of the 302 redirect. Also consider that people and bots visiting those URLs will be adding overhead to your server, and on Amazon that will equal increased cost (small as it may be, the pennies add up!)
Reading the link you provided it looks like the default behaviour of the page metadata redirect under the s3 console is to create a 301 redirect. That makes me think the 302 is coming from somewhere else. Look at the following URL:
http://docs.aws.amazon.com/AmazonS3/latest/dev/HowDoIWebsiteConfiguration.html
It looks like you can add advanced redirects under "Enable website hosting -> edit redirection rules". I'd explore if there are redirects listed there and maybe chat to your developers further.
While you are it I spotted two other issues for you to consider. Currently the index.html files in your directories resolve to the same page as your main directory. I would 301 those pages back to the parent directory (slash version). Or you could add canonical URLs pointing back to the parent directory (with trailing slash). I'd make a case for adding canonical URLs to all pages.
Also, you currently have a number of redirect chains e.g.
http://www.strutta.com/resources/posts/share-your-contests-and-sweepstakes-all-over-social-media 301 redirects to http://www.strutta.com/resources which 302 redirects to http://www.strutta.com/resources/.
You need to find the original redirect and change it to 301 redirect to the trailing slash version of the directory. Screaming Frog can help you find these redirect chains.
-
Hi Danny!
I don't have much to add here, I think the guys have it right in that you'll need to figure out how to make the 301 work. I quickly read that documentation, then realized I wasn't a robot, so I found this: http://aws.typepad.com/aws/2012/10/amazon-s3-support-for-website-redirects.html which was a bit more friendly.
I wish I could help you out more, but I'm not using AWS. I'm assuming you'll be able to use wildcard or regex matching somewhere, and that should solve your problem.
Great site by the way, anything you're using to help out with the static blog? (Jekyll, Octopress?)
-
Follow-up answer:
Our new website (Strutta.com) is entirely static, hosted on S3. No Apache, just straight HTML files. No apache means no htaccess.
Instead of using htaccess, we have to use the S3 Console: http://docs.aws.amazon.com/AmazonS3/latest/dev/how-to-page-redirect.html
As far as I can tell, this sets up redirects the same way. Although this doesn't answer my initial question, I'm going to try using the control panel later on today to see if 301ing the directories there to include the / will get recognized before whatever is causing the 302 currently
-
Thanks all,
I think the problem is coming from the fact that we're hosted on Amazon Webservices, and the devs are using the "aws bucket config" settings to institute redirects instead of htaccess. SEO vs Dev Battle time.
-
Hey Danny,
As Maximilian suggested above the best solution is going to be to change those 302s to 301s. I generally like to redirect to trailing slash URLs for directories and non trailing slash URLs for files/pages (that's that standard convention). I find in practice hardly anyone who links organically ever includes a trailing slash when linking to a page, but when it's the homepage I don't worry about it too much, browsers and Google can figure that out.
Basically you need to figure out where the 302 is coming from and hopefully it is in your .htaccess file. If you can edit your .htaccess file you need to change that to a 301 redirect, or you could remove the redirect and just use a canonical URL pointing at the / version of the page. I would prefer to go with the 301 though. Just be sure to look at how these redirects are being implemented and in what order, you don't want to end up with redirect chains either.
Can you get access to your .htaccess file or is the server running something funky?
-
Perhaps this is too obvious, but can you not change the 302 to 301's?
Got a burning SEO question?
Subscribe to Moz Pro to gain full access to Q&A, answer questions, and ask your own.
Browse Questions
Explore more categories
-
Moz Tools
Chat with the community about the Moz tools.
-
SEO Tactics
Discuss the SEO process with fellow marketers
-
Community
Discuss industry events, jobs, and news!
-
Digital Marketing
Chat about tactics outside of SEO
-
Research & Trends
Dive into research and trends in the search industry.
-
Support
Connect on product support and feature requests.
Related Questions
-
"Cookies are required to access this site" in Google Serp?
One of my clients is having an issue where their Google search result title and description are just showing "Cookies are required to access this site." instead of the actual meta values. The problem is only in Google as Yahoo and Bing seem to be fine. You can see in the image below or by running a search your self for "be well bodyworks longmont" I've never seen anything like it and couldn't find any reference to anyone else having this issue... I would very much appreciate any insight as to what is going on. Thanks! c5PGL
Technical SEO | | CampfireDigital0 -
"Yet-to-be-translated" Duplicate Content: is rel='canonical' the answer?
Hi All, We have a partially internationalized site, some pages are translated while others have yet to be translated. Right now, when a page has not yet been translated we add an English-language page at the url https://our-website/:language/page-name and add a bar for users to the top of the page that simply says "Sorry, this page has not yet been translated". This is best for our users, but unfortunately it creates duplicate content, as we re-publish our English-language content a second time under a different url. When we have untranslated (i.e. duplicate) content I believe the best thing we can do is add which points to the English page. However here's my concern: someday we _will_translate/localize these pages, and therefore someday these links will _not _have duplicate content. I'm concerned that a long time of having rel='canonical' on these urls, if we suddenly change this, that these "recently translated, no longer pointing to cannonical='english' pages" will not be indexed properly. Is this a valid concern?
Technical SEO | | VectrLabs0 -
Link rel="prev" AND canonical
Hi guys, When you have several tabs on your website with products, you can most likely navigate to page 2, 3, 4 etc...
Technical SEO | | AdenaSEO
You can add the link rel="prev" and link rel="next" tags to make sure that 1 page get's indexed / ranked by Google. am I correct? However this still means that all the pages can get indexed, right? For example a webshop makes use of the link rel="prev" and ="next" tags. In the Google results page though, all the seperate tabs pages are still visible/indexed..
http://www.domain.nl/watches/?tab=1
http://www.domain.nl/watches/?tab=24
http://www.domain.nl/watches/?tab=19
etc..... Can we prevent this, and make sure only the main page get's indexed and ranked, by adding a canonical link on every 'tab page' to the main page --> www.domain.nl/watches/ I hope I explained it well and I'm looking forward to hearing from you. Regards, Tom1 -
Disavowing the "right" bad backlinks
Hello, From july to november (this year), I gained 110.000 backlinks. Considering that I'm having trouble ranking well for any keyword in my niche (a niche that I was ranking #1 for several keywords and now I'm losing), I'm starting to believe that negative seo is affecting me. I already read several articles about negative seo, some telling this is a myth, others telling that negative SEO is alive and kicking... My site is about health and fitness in brazilian-portuguese language, and there's polish/chinese/english with warez/viagra/others drugs pointing to my domain and a massive links in comments with blogs without comment approval. Considering that all these new backlinks are not on my language and are clearly irrelevant, can I disavow them without fear of affecting my SEO even more ? Everytime you see someone talking about the disavow tool, is always the same warning: "cautiong when disavowing a link, you can hurt you site even more, removing a link that - in some way - was helping you". Any help or guidelines if I can remove this links safely would be greatly appreciated. Thank you and sorry for my english (it's not my native language) 5ZDjUcK.jpg
Technical SEO | | broncobr0 -
Schema markup for products is missing "price": Is this bad?
Hey guys, So a current client of mine has an e-commerce shop with a few hundred products. They purposely choose to keep the prices off of their website, which is causing errors in Google Webmaster Tools. Basically the error shows: Error: Structured Data > Product (markup: schema.org) Error type: missing price 208 items with error Is this a huge deal? Or are we allowed to have non-numerical prices for schema ie. "call for quote"
Technical SEO | | tbinga1 -
How to use rel="alternate" properly for mobile directory.
Hey everyone, For the URL - http://www.absoluteautomation.ca/dakota-alert-dcpa-p/dkdcpa2500.htm - I have the following tags in the header: rel="canonical" href="http://www.absoluteautomation.ca/dakota-alert-dcpa-p/dkdcpa2500.htm" /> rel="alternate" media="only screen and (max-width: 640px)" href="http://www.absoluteautomation.ca/mobile/Product.aspx?id=37564" /> Yes Google WMT is reading these as duplicate pages with duplicate titles, meta descriptions etc. How can I fix this? Thanks!
Technical SEO | | absoauto0 -
URL Structure "-" vs "/"? Are there any advantages to one over the other?
An example would be domain.com/keyword/keyword2 vs domain.com/keyword-keyword2 Are there any advantages / disadvantages to one over the other?
Technical SEO | | nicole.healthline0 -
How to setup tumblr blog.site.com to give juice to site.com
Is it possible to get a subdomain blog.site.com that is on tumblr to count toward site.com. I hoped I could point it in webmaster tools like we do www but alas no. Any help would be greatly appreciated.
Technical SEO | | oznappies0