Tim, You raise a good point about CMS functionality exceeding that of flat pages, and I understand that some CMS features wouldn't function. However for our public facing site the majority of the site is essentially static pages that would be reproduced perfectly. Another portion is dynamically generated based upon a get request where all the information is in the URL, so I think these could probably also be reproduced. For our site there would only be a relatively small amount of content that required more dynamic interaction with the server and couldn't be flattened
We do have fairly resilient load balanced systems, however these are not infallible, they are also difficult (and expensive in licensing) to replicate reliably outside of our data centre, to cater for a total loss of connectivity to our data centre. What I would like to do is use Apache's proxy-balancer to do some of the current load-balancing but be able to failover to flat pages (reasoning this is far better than nothing), in the event that all of the load-balanced nodes fail. And to be able to mirror the flattened pages offsite in case our data centre lost network connectivity. Thanks, Paul >-----Original Message----- >From: Tim Ruehsen [mailto:[email protected]] >Sent: Tuesday, October 16, 2012 9:14 AM >To: [email protected] >Cc: Paul Beckett (ITCS) >Subject: Re: [Bug-wget] wget mirror site failing due to file / directory name >clashes > >Am Friday 12 October 2012 schrieb Paul Beckett (ITCS): >> I am attempting to use wget to create a mirrored copy of a CMS >> (Liferay) website. I want to be able to failover to this static copy >> in case the application server goes offline. I therefore need the >> URL's to remain absolutely identical. The problem I have is that I >> cannot figure out how I can configure wget in a way that will cope with: >> http://www.example.com/about >> http://www.example.com/about/something > >You can't make a failover copy with wget like tools. Maybe except for very >simple web sites, but a CMS isn't that simple. >On a web server there will be many essential resources that are not available >via remote access (e.g. scripts, servlets, server configuration, database, >...). >What I want to say is: Even if you solve this (minor) problem of not being able >to map URL paths to the local filesystem (a problem that occurs from time to >time which can generally be solved by transforming the URL into a key/value >pair. AFAIK, wget doesn't have such a feature yet), you will stumble over the >next problem that prevents your copy to be a failover copy. > >It sounds that you have administrative access to your company's web server. >So why not using any of the thousands of "professional" >backup/failover/redundancy mechanisms for such use cases ? >E.g. a filesystem and database cluster - today there should be out-of-the-box >solutions. > >But maybe I don't get your intention... > >Tim
