914world is showing its age Options

[dgw]

Some folks are asking what it would cost to fix. No offense to them, but I spent years estimating and doing software projects and the real answer is that you will know what it costs when the final bill is due.
Whatever guesstimate Andy came up with, besides really being a waste of his time, would probably be a quarter (or less) of the real cost and there are going to be lots of free hours in whatever solution is selected. All of us here can contribute some bucks, but it should be whatever you are comfortable giving, because whatever we all give is not going to be enough to cover all the cost and effort required.
So, when/if a collection method is set up, give some bucks to the effort.

Oh yeah, social media solution? Never been on one of those sites and don't plan to ever go there.

[rjames]

Is it an option to migrate the current server and historical data is it currently operates to a cloud-based VM(s) to preserve the historical data, and allow it to be accessed as we're doing so now...

Yes, hard stop. Invision Power Board (this forum) uses a MySQL back end which can be migrated to any current forum software (we use XENForo). No loss of data, and no need to create an "archive" location.

The forum will look and feel differently, but all the data will be there. You just keep using it as-is (after we stop waving our canes at the sky about how things are in a different place).

It all depends on what Andy wants/needs to do. - GA

Wow! That sounds like an option. (IMG:style_emoticons/default/idea.gif)

I'm afraid a migration won't be that simple. I've looked into this before.


There are several problems:

- The current forum software is the last version of IPB before they went to a subscription model. Which means it *literally* is from 2006(!). Any migration effort would need to support an IPB version that old.

- Any new forum software needs to have a way to store uploaded images on the hard-drive in a way that is scalable (Preferably a date based folder tree). Simply putting all images into one uploads folder is going to cause one of the issues we're having right now. We have over 1,300,000 images and counting in a single folder. They take up 360GB of space. I have yet to see an OS that won't balk at having that many images in a single folder.

- Any new forum software needs an option to suck down any linked images from 3rd party websites and store a copy of them locally. Otherwise you'll end up with older threads with broken and lost images.

- Any new forum software needs to be able to handle over 3 million posts in a single database table that is almost 3GB in size. I know MySql itself can handle that, but if the forum software wasn't made to handle numbers like that (like our software never was) we'll end up spending all the time to migrate just to have the same issues again (or worse).

- Any new forum software needs to be owned outright. No subscription service. We need to be able to run our own copy on our own server. Running this as a service in AWS is way too expensive.

(IMG:style_emoticons/default/popcorn[1].gif)

To be clear, my suggestion was not migrating the existing content to new software. What I envisioned:

Spin up a server vm that runs the current forum software. Copy the existing database to the the same vm (another partition or separate vm if need be) and point the old forum software that is now running in the cloud to the database. Put the database in read only mode. The 'old site' lives on and can easily be restored if it crashes. It won't grow in size since it's locked in read only mode and therefore should be less prone to crashes. All the old links should work because everything is in the same place as far as the software is concerned (after updating the DNS records).

In parallel, install new forum software and associated database on another vm in the cloud for posts going forward. Add an '914world_archive' link somewhere on the new site that will take users to the old site.
Keeping the archive and links in tact would mean a new URL for the 'new' forum.

[mlindner]

How about we keep the existing site as an archive, as it is. No mods and completely read only.

We setup a new site with new software and port all the logins over from the old site. Once we cut over to the new site, we can have a search link to search the old archived site, and be able to link back to it.

I am a professional sysadmin, and I will donate all the time it takes for server setup and installation. I can even help with the software setup.

I definitely don't want to lose all of the contributions of Howard, Slits, Lapuwalli, and all of our friends that have passed.

And @gregamy , I think IPB is setup to use MSSQL for this board not MySql. Ether way, we can port the data over.

I'm in with Clay. What is needed SirAndy

XENForo charges 195 for the license. I for one don't expect Andy to pay for the software license out of his pocket. And I don't know what the hardware requirements are for XENForo. So we may end up paying for new hardware too.

I do migrations and server capacity planning for a living. So I know this stuff. We have to CAREFULLY plan it out or we are going to be suffering withdrawls from 914World while the server is down.

[URY914]

Sounds like we need to put TC on a time out (IMG:style_emoticons/default/laugh.gif) It's a fun thread to look thru but it's not the important stuff. That's 240,000 images.

I haven't done web design in 20 years either, so I don't have much to offer as a solution.

Anything is possible now that AI writes code.

Wow, it's my fault. Our Thread gets More Traffic than any others and keeps peeps on the site

Yeah its all that damned TC's fault! That lousy SOB!

(Do you realize I have more posts on here than anyone? LOL!)

[Van B]

I’m down for donations as well. I’ve read through this thread and I would agree that both maintaining the history of the forum and bringing it into current software and formatting is the most costly path. But I do believe it is necessary… especially for a site dedicated to an obscure classic car.

[jfort]

I’m catching up. LOOK AT THE NUMBER OF POST IN SUCH A SHORT PERIOD OF TIME! The information contained here is priceless. We all know that. I know nothing relevant to this problem or how to solve it but we have to preserve access to this information one way or the other. Like others, I will contribute financially. Let’s develop a plan, estimate its cost and gauge interest. I hate subscriptions but for this site it would be worth it. Can AI be brought to bear?

[PanelBilly]

I don’t have the skills to help so I just made an additional donation. Had to access the PayPal account directly to do more than the $25 the site indicates but it was pretty easy to do. I say we start building the war chest of cash.

Do as you see fit with it Andy.

[davep]

I agree that we really cannot afford to lose what we already have in the archives, But we do need something current that we can move forward with for at least another decade. It is worth investing serious cash into.

[Krieger]

I agree that we really cannot afford to lose what we already have in the archives, But we do need something current that we can move forward with for at least another decade. It is worth investing serious cash into.

(IMG:style_emoticons/default/agree.gif)

Sir Andy, I have too much money. Please have some. (IMG:style_emoticons/default/biggrin.gif)

[930cabman]

I agree that we really cannot afford to lose what we already have in the archives, But we do need something current that we can move forward with for at least another decade. It is worth investing serious cash into.

(IMG:style_emoticons/default/agree.gif)

Sir Andy, I have too much money. Please have some. (IMG:style_emoticons/default/biggrin.gif)

(IMG:style_emoticons/default/lol-2.gif) (IMG:style_emoticons/default/lol-2.gif)

[StarBear]

I’m down for donations as well. I’ve read through this thread and I would agree that both maintaining the history of the forum and bringing it into current software and formatting is the most costly path. But I do believe it is necessary… especially for a site dedicated to an obscure classic car.

(IMG:style_emoticons/default/agree.gif) Sign me up!!

[TC 914-8]

I think preserving what we have in a separate database might be worthwhile. I think we are a group of very techy people from all levels of technology, we all love to fix things
I had a metal shop teacher (45 years ago) that said “you aren’t building nothing unless you have a plan”. I know Andy, along with other members of this group, know the best path of least resistance. If we build a plan, put a cost to it, time and money, see what it takes to move forward.
Not sure how urgent this is, but we’d all be more than glad to pitch in dollars, time, expertise to make something happen before we crash and burn, not an option.

I’m in to help if needed, minimal IT skills, some $$$, plenty of half-ass ideas. (IMG:style_emoticons/default/sawzall-smiley.gif)

[barefoot]

914world is showing its age (IMG:style_emoticons/default/sad.gif)

The recent outage was not due to some cyber attack or malicious plotting from the darkest depths of the interwebs.

It was simply due to the forum software being overwhelmed by the amount of data stored here combined with the amount of traffic we're receiving.

Fact is, the software we're running on is more than 20 years old, written at a time when the internet was a much different place.
It was never meant to run for this long or handle the amount of data we have accumulated.


So what are our options?

- Keep going until everything crashes and burns into a pile.

- Start over fresh with a new server with new forum software. We'd be losing everything we have.

- Try to fix some of the issues by re-writing parts of the software, reworking the database and changing the way uploaded images are stored. Maybe even update the server itself while we're in there.


While the last choice seems the obvious one, the problem is that i do have a day-job that needs my attention and my time is limited.
It would also mean i'd probably have to shut down the site for a longer period of time, maybe a week or two, and maybe more than once.
(IMG:style_emoticons/default/dry.gif)

I'm not sure if you can impose a limit to image size ?
I frequently have to re-size images for other sites, easy to do on a MacBook.
Barefoot

[NARP74]

I found the problem! Delete this ASAP!
(IMG:style_emoticons/default/thisthreadisworthlesswithoutpics.gif)

[DennisV]

- The current forum software is the last version of IPB before they went to a subscription model. Which means it *literally* is from 2006(!). Any migration effort would need to support an IPB version that old.

- Any new forum software needs to be owned outright. No subscription service. We need to be able to run our own copy on our own server. Running this as a service in AWS is way too expensive.

(IMG:style_emoticons/default/popcorn[1].gif)

Hey Andy. Thanks for all the time you invest in 914World.

A few thoughts / questions. I am sure you have reviewed these at some stage, so I don't mean to armchair admin, but I don't see them in this thread:

• Poking around a bit about Invision migration seems to indicate folks have had success migrating IPB to both Xenforo and Discord. It's encouraging there seem to be options to save the existing content.
• Why no subscription service? I understand philosophically, but practically this is pretty much the way of the world now. It would not necessarily need AWS. Bulletin Board as a Service would presumably save you admin headaches and worries about updates and security. Xenforo appears to start at $100 per month. Would the annual supporter donations not be enough to cover that?
• Xenforo provides a paid migration service to their hosted Cloud offering. Has this been looked at? While it would have a price attached, they would have a vested interest in making sure things came over successfully and would not require as much of your time. Find out this cost and let the forum members put their money where their mouth is. We cover it or don't migrate.
• I'm another that would not want to see this evolve to a Facebook entity
• Might the move to another platform allow segregation of free and paid content? On 356 forum, anyone can read posts. It is only with membership that you can search, post, or reply to posts. The membership fee obviously helps cover forum hosting expenses.

FWIW, these thoughts come from experience I had migrating a very old trouble ticket system and CRM system to a cloud hosted service. It was still a PITA, even working with the vendor and a consultant to design and implement the migration.

Thanks again for keeping the forum alive.

[mgarrison]

First off, thanks Andy for all you do for this site - simply amazing! (IMG:style_emoticons/default/smilie_pokal.gif)

I don't do any social media either, so would be really bummed if things went towards FB, etc. I'm in for a donation as well, or even an annual membership fee to cover costs...? 914 world and it's members and their knowledge has been priceless for me and my build. Keeping the existing data seems like a high priority to me, even if it's read-only going forward.

[Artfrombama]

With all due respect;
I'm thinking part of the problem everyone has avoided discussing is that maybe it's time to "pass the admin torch" to a qualified member with the time, motivation and expertise to take what we have, preserve and improve.
What say you @SirAndy

[Nogoodwithusernames]

I see the option of keeping this site as a "read only" archive tossed out there a few times, and while certainly better than letting all the info disappear I think it will also basically kill much of the old info being in a separate "dead" site. I think all this existing info being transferred over to a new site would be important to keep everything connected and alive, to be able to revive a dead thread/build with a new question for example.
Obviously this would incur MUCH more work/cost but it seems there is enough support to make it happen, and would be the ideal solution, if feasible.

I am now 30 so I'm too old to be up-to-date with modern tech (maybe I just don't care enough) and too young to be able to ignore it entirely (IMG:style_emoticons/default/blink.gif)

Just my $0.02 for what that is worth these days.
(IMG:style_emoticons/default/driving.gif)

[930cabman]

Some folks are asking what it would cost to fix. No offense to them, but I spent years estimating and doing software projects and the real answer is that you will know what it costs when the final bill is due.
Whatever guesstimate Andy came up with, besides really being a waste of his time, would probably be a quarter (or less) of the real cost and there are going to be lots of free hours in whatever solution is selected. All of us here can contribute some bucks, but it should be whatever you are comfortable giving, because whatever we all give is not going to be enough to cover all the cost and effort required.
So, when/if a collection method is set up, give some bucks to the effort.

Oh yeah, social media solution? Never been on one of those sites and don't plan to ever go there.

"because whatever we all give is not going to be enough to cover all the cost and effort required."

We cannot afford to upgrade??

I really have NO clue, but some guys know a potential solution. I for one drive/repair a 1/2 century old jalopy because I am an optimist. There must be a solution somewhere. Can no one estimate the $$ required?

[NARP74]

Maybe a few more admins so Andy does not bear all the time and responsibility. He has probably done more than we know to "keep the dump running."

My last favorite site died this way, single point of failure, he left to run a company, just keep the site going, it crashed and we lost everything. They decided to move on and the void was filled many ways, never the same.

If this was a corporate eval I was on it would fail. Anyone still using Win 95, 98 or Millennium, Server 2000 to run a successful business. While this is not a business, it could be treated as one.

For many good reasons, seems like we have kicked the can down the road and done just enough to keep it running. Again, NOT a dig at Andy. The time has come to pay up though before it is too late.

Most corps I have worked with or at adopted an N-1 architecture for critical infrastructure. Not bleeding edge, let others work out the bugs, no more than a stable version behind. Then create a migration plan and get it done every few years.

Maybe start a committee, some zoom calls, get into the weeds, come up with some plans. Go public with it and get to work funding it and doing the work.

Times, hardware, software, applications, services have all moved on, Moore's Law anyone? Not transistors and it has slowed down lately but we are still behind.