Possible Re-hosting Of Website - Input Requested

Never got the call from CMS2CMS, the "professional" migration place.  But, I have gotten several emails from Wendy @wpbeginner, the free migration place.

So I asked her about their services, and here's her answer:

We offer two kinds of free support - and yes it's free. The first is what we're doing now which just answers questions you have as you're doing the site yourself. The second may be what you're looking for and that's where we actually move your Weebly site for you. However our service only covers basic posts and pages. Since you have a forum that and the scribd pdf viewers probably wouldn't transfer and that is the majority of your content.

If you still want to take advantage of that you don't pay us directly but you do have to sign up for hosting with one of our preferred hosts

Fortunately we don't have too many Scribd pdf viewers, and those we do have can be converted to embedded iframes.  As for this forum, I've tried several ways to embed it, all to no avail.  But I'm working with my friend on the Nabble support forum and I'm hoping he can crack it.  But this one is critical, so I need to make it happen before we get much farther.

Yeah, welcome to WordPress.

Seriously.

Do a Google search on:
how to export weebly to wordpress

The results show lots of help out there, don't be discouraged too much by this one approach you took.

IOW there are other plugins & services out there, experiment around and maybe you can find one that does a better importing job, maybe you can modify it (or the XML output results) to do what you want it to do.
Yes, the PDF files will need to be editted.

If there are identifiable attributes of a post/thread/document/etc. that you want to change to a known value, then a lot can be done wring SQL queries & inserts using MySQL Workbench

https://www.mysql.com/products/workbench/


Anyway, she said that if you run the importer several times it frequently gives duplications in the menu

Yes, that's exactly what happens. You see, most of this stuff (the content of wordpress sites) is entries stored inside of a database and not files on disk, there is no way for an importer tool to (easily) determine if such-n-such db entry already exists or not, so it merely writes a new one (menu entries). Any importer tool can't easily figure this out without access to the database (files on disk are easy). Restore the database from a backup and try again.

But now you know what this one does.

Non-Clickable Pages

Not sure what you're talking about here, reading real fast, I guess....

I hope CMS2CMS can do a professional job 'cause fixing what we have would take quite a bit of time.  But the fact that we got that far, or rather Chris/ckuske got us that far, for free speaks volumes. 


Yes, manual fixing of everything would take forever, don't even think about it... Script it. Look at the XML output some of those things are producing, look at the source of what's producing it, it's not complicated at all. Lots of HTML-like entities, some of which you can fix in vi. Other things require more, e.g. SQL queries & database operations.

Again, welcome to WordPress. And there is worlds of support out there. :)

Well, while both of us were watching baseball Chris/ckuske fixed the forum.  Just like that!  

The issue was that I had put an HTTP forum on an HTTPS site, and ne'er the twain shall meet.  He turned the S off, and away we went.  In fact, I just posted on the This Looks Cool thread from that site.

So now the concept has been proven.  It is possible to migrate the website, meaning the documentation portion, to Wordpress.  And it is possible to embed the forum in it.

The only thing we now need to figure out is whether we've done the migration and want to spend the time cleaning it up.  Or, much more likely, we want to have it migrated by someone that has done it a few times.

I still haven't heard from CMS2CMS so that may not happen.  But WPBeginner says they'll do it for free if we go with one of their hosting partners.  And one of those is BlueHost, which has good rates and comes out on top of the list in a review I read.

So Monday when the WPBeginner folks come back to work I'll talk with them more about what we'll get if we go with them on the migration.  And if CMS2CMS should call, I'll check them out as well.

Onward and upward!

Thanks, Chris.  I'll read the info in the links tomorrow.  Too much computing and I'm ready for bed now.  But this post is coming to you from a Wordpress site.  

Before you get toooo involved, spend some time looking at some themes, find a new look & feel you like. For example, the tabbed pages; I seriously don't expect that stuff to import over real well, you might be better implementing some WordPress-style of doing those things (remember, there are boatloads of plugins & themes available).

Or, look at some of the commercial themes; I've seen some I liked on ridethetockies.com & pedaltheplains.com, though I forget the name of the company that authors/sells them. :( Let Google be your friend. :)

Chris - There's LOTS of research to be done.  Themes, plugins, you name it.  And thanks for the places to look, I'll follow up on them.

I've done a bit of playing with the test site.  For instance, I uploaded the 46Mb Four Wheelin' brochure in pdf to the site's media section and then embedded it below the same file saved on OneDrive.  Both were embedded the same way using iframes.

And then I viewed the page.  The file from OneDrive loaded in a second or so, but the file from the Dreamhost site took MANY seconds.

So I downloaded and installed the PDF Embedder plugin that supposedly does things "better" and doesn't use iframes.  Then I embedded the Four Wheelin' file below the 2nd instance and viewed the page.  Again the iframed file on OneDrive was by far the fastest, and the PDF Embedder file was 2nd.  But the iframed-Dreamhost file took forever.

The hope was that locally-hosted files wouldn't be found in media storage when Google crawled the site but would be on the page.  Gonna have to think about this a bit.

I know we are also going to need a backup plugin.  And I'm sure there are plenty of others.  But at least those things are available on Wordpress.  They aren't on Weebly.

Ok, I'm trying to think through a game plan. And one thing that appears to be highly likely is that pages using Weebly's widgets, like tabs and buttons, won't come across correctly. (The jury is still out on that one, but I'm pretty sure that's going to be their verdict.) So the question becomes whether to take the widgets out before the migration or fix the pages after. I've thought of a couple of issues, but I'm sure there are more:

1. Content: If we "fix" the pages before the migration the content will come across where it wouldn't, or doesn't appear to do, if the tabs/buttons are left in. On the other hand, as long as we've not cut over we'll have the Weebly site to refer to and figure out what the content should be. So, in some ways it might be better to fix the pages afterward 'cause we we might be able to use the power of Wordpress to create the pages in a better fashion.
2. Editors: In Weebly I'm the only editor, and to have more requires we spend money to upgrade plans. But in Wordpress it looks like we could have several editors. So if we fix the pages after the migration we could spread the work out - assuming we had others with the skills willing to do it.

So, over to y'all. What other issues or thoughts are there regarding the order of doing things?

Gary, are you able to try a couple of sample pages to see how they test? I understand wanting a smooth and uninterrupted transition, but my gut feeling would be to move it, then fix the problems in the new environment rather than trying to preemptively fix it and hope everything lands sunny side up. Granted that is just me, with zero knowledge of the process and a fair amount of mistrust in computer programs doing what they are supposed to.

We’ve moved all 500+ pages, and those that don’t use the widgets are fine. I’m sure they could be better given all the capabilities of Wordpress, but I happy with just moving over one-for-one. Then we can upgrade later if we want.

Anyway, I’m leaning to your thinking of move and then fix. 👍

Gary,

I had composed a response but lost it when tried to preview where an external link was taking me and then hit the browser Back button and it put me at the Table of Contents. That's annoying.

I told you about your SSL configuration some time ago, I've been clicking through it in Firefox for what seems like years now. Unencrypted HTTP is certainly easier and was good for times past, but the world is moving to encrypted everything at this point.

Performance comparisons need to be fair. Your browser cache could be affecting a lot of your perceived speed, it's best to clear your cache & maybe cookies between runs when making those tests.

Your VMWare virtual machine's capabilities certainly make a difference, these are generally dependent on how much RAM & bandwidth you pay for.

Wireless networks are slower than wired but they aren't subjective, it's everybody or nobody.

There are tools available to measure network performace (e.g. Wireshark or tcpdump) but their use & interpretation can be a bit non-trivial. You will need source & destination IPs awa the protocol being used; time is measured in milliseconds. Maybe your new Chris can look into that if he's interested, he's doing a good job so far. :)

It doesn't surprise me that OneDrive performs well, it's a cloud product produced by Microsoft and of course they want their infrastructure to perform well. Sometimes there are benefits to using cloud services, one of them is sometimes speed.

I know nothing about Dreamhost or their capabilities, they are just a hosting company like lots of them (used to be known as ISPs). They provide virtual machines on a blade server connected to the 'net, just like all of them.

Backups will be easy, the hosting companies will likely point you towards a Linux GUI control panel known as cPanel. It works fairly well and is certainly easier for most people to use. Or, command line tools can be used to create sqldump backups & tarballs of disk files, then copy them to your local HD. Your choice.

Gary Lewis wrote

 But at least those things are available on Wordpress.  They aren't on Weebly.

Weebly is the Playskool of Internet hosting. Certainly it is easy, and many people find it sufficient for recreational use. WordPress allows virtually unlimited functionality, but you need to operate within its constraints.

Gary Lewis wrote

We’ve moved all 500+ pages, and those that don’t use the widgets are fine. I’m sure they could be better given all the capabilities of Wordpress, but I happy with just moving over one-for-one. Then we can upgrade later if we want.

Anyway, I’m leaning to your thinking of move and then fix. 👍

Please be precise. ;) You want to *copy* and then fix the *copy*, leaving the original site intact.  To *move* a site implies deleting the original source, i.e. only one version remaining.

Regardless, I would copy whenever possible, using the original site as a reference. Themes are written independently by several authors, I don't expect theme-specific features to migrate over well at all... (I think I said that already). Swap DNS when you're ready to turn on your new site, simple as that.

Chris - You are right, I didn't use the right term as we certainly didn't "move" the site as all of the pages are still on Weebly.  But I'm not sure "copy" is quite right either as they didn't come across the same.  At least not all of them did.  Perhaps "convert" is a better term.

In any event, we think we have 100% of the Weebly-created & hosted pages up on Wordpress.  But there are differences, with widgets not having converted well, if at all.

In any event, we are playing around with it to see what works and what doesn't, tweak what we can, fix broken links, etc.  Basically just to get our arms around what the issues are.  But I still plan to talk to WPBeginner tomorrow to see if they think they can do a better job themselves than what we've done with their tool.  If so I'll probably give them the go-ahead to convert it and put it on a Bluehost site.  They don't charge for the conversion if we go with one of their partners, and Bluehost is and has excellent pricing.

However, we won't truly be swapping the DNS.  We have a new URL, bullnosebible, and we'll do re-directs to take people to the new site instead of the old.

As for performance, from what I've seen the same things load at the same speed on Wordpress as on Weebly.  And since we probably won't be changing things the initial user experience will be pretty much the same.  But we then need to consider where files are hosted so we can get Google to crawl them, and at that time we'll also need to do the fair testing you mentioned - with a clear cache.  Good point.

I've read that Google has a plugin that integrates Google Drive with Wordpress, and we could consider doing that if it would be faster and get the pages crawled.  But that's down the road a bit when things are working and we are ready to move on.

And there are plenty of backup plugins as well.  A review I read shows some to be either inexpensive or free and provide the features I think we need, like backing up at given times and dates, backing up on any change, incremental backups, etc.

Anyway, things are moving and it looks like we'll get there.  Plus, some things will be better on the new site.  For instance, this is being typed via the test site, and I just realized that the window into which I'm typing is noticeably wider than on Weebly.  In fact, I just went back to the Weebly view of the forum and started a reply and the editing window is certainly smaller.  I think that's due to some custom CSS I installed on the Wordpress site last night that uses 95% of the available width for the "container".  

Gary Lewis wrote

Chris - You are right, I didn't use the right term as we certainly didn't "move" the site as all of the pages are still on Weebly.  But I'm not sure "copy" is quite right either as they didn't come across the same.  At least not all of them did.  Perhaps "convert" is a better term.

They are 2 discrete and separate operations with different results from my standpoint as a Linux sysadmin, Gary. ;)

I worked as a WordPress sysadmin for ~3 years at the company that owns the Denver Post (~350 blogs hosted on 2 Linux clusters) and so that's the perspective I'm looking at your new site from.

Right now it's a Parked godaddy site, which means it's still in limbo, usually because the contracts aren't finished yet.

Gary Lewis wrote

In any event, we think we have 100% of the Weebly-created & hosted pages up on Wordpress.  But there are differences, with widgets not having converted well, if at all.

Shocking!

Gary Lewis wrote

In any event, we are playing around with it to see what works and what doesn't, tweak what we can, fix broken links, etc.  Basically just to get our arms around what the issues are.  But I still plan to talk to WPBeginner tomorrow to see if they think they can do a better job themselves than what we've done with their tool.  If so I'll probably give them the go-ahead to convert it and put it on a Bluehost site.  They don't charge for the conversion if we go with one of their partners, and Bluehost is and has excellent pricing.

OK, that sounds good. Bluehost has been around for a long time and has a decent reputation at what they do.

Gary Lewis wrote

However, we won't truly be swapping the DNS.  We have a new URL, bullnosebible, and we'll do re-directs to take people to the new site instead of the old.

Which means (www.)garysgaragemahal.com - or something standing in its place - will necessarily need to be alive so it can redirect users (assuming http 301 Redirect) to the new site.

Is this what you're wanting to do? If so, why? In all cases, *somebody* has to do your DNS.

Gary Lewis wrote

As for performance, from what I've seen the same things load at the same speed on Wordpress as on Weebly.  They should be. And since we probably won't be changing things the initial user experience will be pretty much the same.  But we then need to consider where files are hosted so we can get Google to crawl them, and at that time we'll also need to do the fair testing you mentioned - with a clear cache.  Good point.

Having content hosted "in the cloud" isn't as good from an SEO perspective as having them hosted locally. Many search engine robots won't cross DNS boundaries when indexing a site.

I thought I explained that already.

Gary Lewis wrote

I've read that Google has a plugin that integrates Google Drive with Wordpress, Let the tracking & measuring begin and we could consider doing that if it would be faster and get the pages crawled.  But that's down the road a bit when things are working and we are ready to move on.

I'd try not to. See above.

Gary Lewis wrote

And there are plenty of backup plugins as well.  A review I read shows some to be either inexpensive or free and provide the features I think we need, like backing up at given times and dates, backing up on any change, incremental backups, etc.

Anyway, things are moving and it looks like we'll get there.  Plus, some things will be better on the new site.  For instance, this is being typed via the test site, and I just realized that the window into which I'm typing is noticeably wider than on Weebly.  In fact, I just went back to the Weebly view of the forum and started a reply and the editing window is certainly smaller.  I think that's due to some custom CSS I installed on the Wordpress site last night that uses 95% of the available width for the "container".  

Yeah, there are several backup solutions, Bluehost might even be able to help you with that.

Not trying to be rude or impolite, I'm just looking at it from a different perspective is all. :)

As for window width, yeah. :) There might be several places you have to adjust things but much of it is fairly simple to do.

~~

I'm not going to get back to my truck for a while it seems, gonna be spending a lot of time working before I head to Chicago for my 40th HS reunion. My sister is going with me, and she's getting us into the house we grew up in. It's all fairly nostalgic, TBH.

EDIT: If you send me a u/p I'd be glad to look around. :)

Chris - Thanks for the response.  Below are a couple of short answers before I get into today's status.  But first, why not start a thread on your truck?  Otherwise it is mixed in with the other WHYDTYTT comments.

URL: The "why" is that I think bullnosebible is a lot more descriptive of what we have than garysgaragemahal.  I don't know how all the redirect things will work, although there are plug-ins to do that.  But we will worry about that down the road.

Content Hosting: I understand the advantages of hosting on the site vs on the cloud.  But we can't kill the response time in an effort to get Google to find the content.  So we will have to do some testing.  And if the Google integration thing both gets it found by Google AND solves the response issue then we will have to consider it.

Ok, now for the status, findings, etc.  As far as I can tell there are two possible ways to get a Weebly site migrated to Wordpress - WPBeginner, which is free if you use one of their partner providers, and CMS2CMS, which costs $300.

Let's start with the latter, CMS2CMS: Poo.  Junk.  Etc.  I've had absolutely nothing but a horrible experience in trying to communicate with them.  But I was willing to endure that up front if the product was acceptable.  Yes, our initial effort to migrate 10 pages using their tool was abysmal, but they offered to try 20 pages done by their "pro's".  So I gave them links to each page in the Engine section.  You be the judge: https://demo.cms2cms.com/wordpress-5d9c7f32ad061/

But the lady at WPBeginner has been excellent to work with.  However, she doesn't seem to think they can do any better of a migration than what we got using their tool.  And as previously said, the words came through but the tabs didn't.  And many of our pages use tabs.  So lots and LOTS of rebuilding will need to be done.

I've been trying to clean the WPBeginner site up while I watch the MLB playoffs, and I'm on Page 19 of 28.  So I hope to be "done" with the cleanup that will allow the nav menu to work in a couple of days, at which point I'll give y'all a URL so you can go look.  But right now it is a real mess, and y'all seeing that wouldn't serve any purpose.

So it looks like the way to migrate the site is to use the results from the WPBeginner tool and go through it page by page and make it right.  That's a HUGE undertaking, but at this point in time I think it needs to be done.  I've gotten to a point on the Weebly site that I loathe adding pages because it is so onerous.

Given that, I think the plan that is emerging from the fog in my brain is to get what we have cleaned up so the nav menu works, and then take a section and make it right.  That'll give me a much better understanding of how much effort it'll be to do the whole thing, and then a decision can be made.

Ok, I just looked at a couple of the sections, the 351/400 portion seems to have migrated very well, the 385 pages, nothing really there to look at.

Bill I think you are missing several things.  First, they didn't do what I asked, and what they agreed to do, meaning convert the whole of the Engine & Exhaust sections, which make up 20 pages.  Instead they converted 9 pages into 9 posts.  Posts, as in blog posts.  Not pages.

Second, of the 9 they did none of them came over intact.  Here's a side-by-side of the lower portion of the 351M And 400 original on the left and their's on the right.  Overlooking the fact that their page isn't nearly as wide as the original, which is a setting on their website, notice to complete lack of the tabs, each of which holds a lot of documentation.  At least the WPBeginner conversion had the tabs, although no content was in them.  But the CMS2CMS conversion doesn't even have the tabs.  

https://wordpress.org/support/article/pages/

Chris - What are you trying to say?  I'm somewhat aware of the difference between pages and posts.  As I'm sure you are aware, we have no posts on the documentation portion of the website, and over 550 pages.
 Anyway, what are you wanting me to see?

All - Time for an update.  Chris/ckuske found a statement that suggests that we might be able to upload an export of a Weebly site to Wordpress.  As it turns out, the way you "backup" a Weebly site is to export it and save the file.  After the most recent addition to the site, which was a week ago today, I exported the site and then downloaded the file.  Chris is going to try to import that file into Wordpress soon, and we'll find out if that works.

But, to give you a bit of background, that file is 1.7Gb in size and takes between 3 and 4 hours to download.  However, of my three Windows machines the only one that is able to download it is this Microsoft Surface tablet.  The other two always have "network failures" part way through the download, and while you can "resume" the download you never get done.  In fact, on this tablet if you download or upload anything else while downloading the export then you'll get a network failure.

So that means that backing up the site is a real bear.  In addition there's no way to automate the backup that I can find.  All of that adds fuel to my fire of moving us to Wordpress where there are lots of plugins that will automatically backup the site.

Some of you may be wondering where the idea of moving the site to Wordpress stands.  After all, it has been almost two months since anything was said about it.

Basically, the move apparently isn't going to happen.  Chris/ckuske did a bunch of work trying to find an automated way to resolve the problems that he found and finally came to the conclusion that it would be less work to rebuild the site page by page on Wordpress than it would be to figure out how to automagically migrate it.  And rebuilding it would be a huge, HUGE pain!

So I went back to my issues with Weebly and determined that I can live will all but one of the problems, and that is the pain it has been backing the site up, as described in the previous post.  But, I also discovered that the problem of having "network failure" during a backup was a lot worse than I'd described.  In fact on this tablet, which had been the least bad one to use for downloading the backup, a recent test had me have to resume the backup probably 30 times over a span of 12 hours to finally get it done.  Clearly that wasn't acceptable.

So I did some looking for ways to fix that problem.  And luckily I seem to have found one - Auto Resume Downloads extension for Google's Chrome browser.  The reviews of it are mixed, but it works a treat for me.  I'm now able to start a download on any of my computers and it will run to completion in a couple of hours with no babysitting on my part.  Yes, there are "network failures" along the way, but in a minute or so the download automatically resumes.

Given that, I've now started adding pages to the site again.  And each time I add a page or documentation to an existing page, I go through the process of causing Weebly to generate a backup and then I download the file and save it.

Gary Lewis wrote

Given that, I've now started adding pages to the site again.  And each time I add a page or documentation to an existing page, I go through the process of causing Weebly to generate a backup and then I download the file and save it.

Gary,

I see you're catching up on some projects today! Did you take a day off from being retired?>..lol.

I'm just kidding, I know you're a busy guy!