NSwiki.net (old site): getting the info off of it

[Todd McCloud]

Welcome to... the ultimate, concise, effort to get back the information from nswiki.net extravaganza.


What is this?
This is a thread dedicated to getting pages and other information off of the old nswiki.net website and onto new, more stable nationstates' wikis. NSwiki.net was the original wiki site owned and operated by Goobergunch, a former NS player. It was going fairly well (albeit a little spammy) until about October 2013, when the site went off line. It's been that way since.

In January of 2014, I and a few other players were able to get our hands on a few of the wiki files. One file was updated until about 2008 and has already been put back online by Afforess. The other one is supposedly newer, but we haven't really been able to use it. I'm not entirely sure what the problem is, if it's just a lack of communication, or if the file is somehow corrupt, etc. I'm going to be honest in saying I know next to nothing about sql files, php, whatever. But others do. That's where you all come in.


What are we going to do?
We're going to get these pages back on-line again. Somehow. Because, see, here's the bottom line: that wiki page was the collection of knowledge from various nationstates outlets, from gameplay to NS sports to II to various regional RP's. Pretty much if it existed on NS and was a big deal to a person or a group of people, it was probably recorded on that wiki. I liken it to the Library at Alexandria - the ill-fated library that was set on fire in ancient times, turning a wealth of information into ashes. You can't set fire to the internet, so it's gotta be here somewhere.


What can I do?
Well, I have the file. Others do too. We're going to get it working. Somehow. If you're good at this kind of stuff, I want you on board. So if you're good with mediawiki, sql, wiki files, the likes, welcome to the club. If you happen to have another piece of the puzzle, like some important tidbit or another file or something, I want you on board. If you have your own personal backups, add them. Ultimately I'd like for us to have a means to transfer a wiki page from the file and put it on the new nswiki site, or another site.


What shouldn't I do?
This thread isn't about figuring out which wiki is / was better, past grudges, trying to armchair quarterback and tell people that they should've used another wiki, etc. None of that helps here. Just focus on getting this going. All I ask. The NS historians will thank you.

[PrussianEmpire]

/me slowly claps.

Your efforts are commendable.

[Glen-Rhodes]

What "file" do you have? If you have the database file (it would end in .sql), then there really should be no difficulty restoring it on a new MediaWiki installation. That would give you all the articles on the wiki, at least. Images, thumbnails, skins, and extensions would have to be restored from a full backup, rather than a database backup.

[Todd McCloud]

What "file" do you have? If you have the database file (it would end in .sql), then there really should be no difficulty restoring it on a new MediaWiki installation. That would give you all the articles on the wiki, at least. Images, thumbnails, skins, and extensions would have to be restored from a full backup, rather than a database backup.

I have a .gz file. I know that's a winzip extension, but it appears to have been zipped twice. The twice-zipped file is 102 MB. The once-zipped file is around 306 MB. That's about all I know at this time. Once twice-unzipped, the SQL file appears to be there. But, again, don't really know how to open it.

EDIT: I can open the file in notepad and see things like various directories and the likes, but that's about all I know. Takes forever to open, lol.

[Unibot III]

I have a .sql file from Goobergunchia. It's 299 MB of data, but it's unclear from the code to an untrained newb like myself, what the hell the code is referring to.

[Todd McCloud]

I have a .sql file from Goobergunchia. It's 299 MB of data, but it's unclear from the code to an untrained newb like myself, what the hell the code is referring to.

Correction, mine's about 306 MB. Don't really know where that extra 7 MB came into play, but perhaps one of us have a non-corrupt version?

[Glen-Rhodes]

.GZ is just a compression format. You have the SQL file, so that's great. You have to import it into a new database. Do you guys have a server ready for this?

e: The larger database has more data, so that's the one you want to use.

[Todd McCloud]

.GZ is just a compression format. You have the SQL file, so that's great. You have to import it into a new database. Do you guys have a server ready for this?

e: The larger database has more data, so that's the one you want to use.

As I said, all I have is the file. Er, all we have. Apparently there are a few versions of this file.

Now I've heard from Afforess, who was working on this a while ago, that the file might be missing stuff. See, the 2008 file is bigger than the 2013 file (I think). Unless there was a mega-deletion that occurred to remove junk from the wiki, that doesn't make a lot of sense to me.

[Unibot III]

I have a .sql file from Goobergunchia. It's 299 MB of data, but it's unclear from the code to an untrained newb like myself, what the hell the code is referring to.

Correction, mine's about 306 MB. Don't really know where that extra 7 MB came into play, but perhaps one of us have a non-corrupt version?

Is it called Goobergunchia.sql or Goobergunchia1.sql? I've got the latter. If yours is the 2008 version, I imagine it's called Goobergunchia.sql.

[Glen-Rhodes]

If you both want to give me copy of the databases you have, send me a TG and I can import them and maybe see what's going on.

[Commerce Heights]

I’d be happy to help with trying to figure out what’s in these files if anyone wants to send them to me.

[Commerce Heights]

The file which Todd McCloud sent me (goobergunch1.sql.gz · 104 497 391 bytes · MD5 f9e90bcc05725bb2a3c81e654e7424fd) is truncated, ending with this in plain ASCII:

<br />
<b>Fatal error</b>: Maximum execution time of 300 seconds exceeded in <b>/usr/local/dh/web/phpmyadmin/libraries/export/sql.php</b> on line <b>990</b><br />

The decompressed data includes tables up to ‘revisions’ in alphabetical order. The ‘text’ table, which is where the actual article text would be, is completely missing.

[Todd McCloud]

The file which Todd McCloud sent me (goobergunch1.sql.gz · 104 497 391 bytes · MD5 f9e90bcc05725bb2a3c81e654e7424fd) is truncated, ending with this in plain ASCII:

<br />
<b>Fatal error</b>: Maximum execution time of 300 seconds exceeded in <b>/usr/local/dh/web/phpmyadmin/libraries/export/sql.php</b> on line <b>990</b><br />

The decompressed data includes tables up to ‘revisions’ in alphabetical order. The ‘text’ table, which is where the actual article text would be, is completely missing.

I've informed Goob of this. Hopefully he responds, because I'm guessing that he's the only one that can really fix this, I think. Unless someone else has a copy of the data backed up. Would mediawiki or another wiki site have a backed up data file?

[Unibot III]

The ‘text’ table, which is where the actual article text would be, is completely missing.

This was what I thought when looking at it, but I wasn't sure as a non-expert. Um, before anyone here criticizes Goob, honestly, I doubt he knows he didn't send us the full package of data - in eleven years, he's never needed to give it to someone else. He's hard to get a hold of, so this could take a while. Thanks.

[Todd McCloud]

The ‘text’ table, which is where the actual article text would be, is completely missing.

This was what I thought when looking at it, but I wasn't sure as a non-expert. Um, before anyone here criticizes Goob, honestly, I doubt he knows he didn't send us the full package of data - in eleven years, he's never needed to give it to someone else. He's hard to get a hold of, so this could take a while. Thanks.

Aye. Hopefully he responds. If someone else could point this thread out to him, it'd be greatly appreciated.

As stated in the OP, this thread isn't about grudges or blaming people for whatever. We're just here to reach a solution only.

[95X]

I thought the old site was nswiki.net, not .org.

I found a version of my own page that was newer than the 2008 version on archive.org, and saved it to local storage. The Internet Archive might not have every single edit of every single page, but it's a better than nothing self-help end-user solution.

[Glen-Rhodes]

The decompressed data includes tables up to ‘revisions’ in alphabetical order. The ‘text’ table, which is where the actual article text would be, is completely missing.

That's what I ran into as well. He tried to export the database using phpMyAdmin, but that software cannot export a database that size under standard configurations. He should have used the command line or something like MysqlDumper. Unfortunately, unless he still has the full database available to him, there is nothing anybody can do now.

[Todd McCloud]

The decompressed data includes tables up to ‘revisions’ in alphabetical order. The ‘text’ table, which is where the actual article text would be, is completely missing.

That's what I ran into as well. He tried to export the database using phpMyAdmin, but that software cannot export a database that size under standard configurations. He should have used the command line or something like MysqlDumper. Unfortunately, unless he still has the full database available to him, there is nothing anybody can do now.

Hopefully it was just some kind of error converting it and the database is still intact somewhere.

[Todd McCloud]

I thought the old site was nswiki.net, not .org.

I found a version of my own page that was newer than the 2008 version on archive.org, and saved it to local storage. The Internet Archive might not have every single edit of every single page, but it's a better than nothing self-help end-user solution.

Archive.org has some pages, but not all. To access the saved pages, go here. If you find one of your pages, save it under html. There's a good html-to-wiki formatter that does a fairly good job (templates can be difficult though). Also, for some reason, the image directory on NSwiki.net (you're right - it's net not org) is still operational and is located here.

[PrussianEmpire]

If there's anything I can do to help, just send me a TG

[A Colour Out of Space]

The decompressed data includes tables up to ‘revisions’ in alphabetical order. The ‘text’ table, which is where the actual article text would be, is completely missing.

That's what I ran into as well. He tried to export the database using phpMyAdmin, but that software cannot export a database that size under standard configurations. He should have used the command line or something like MysqlDumper. Unfortunately, unless he still has the full database available to him, there is nothing anybody can do now.

Yep, that was my analysis of the situation too. The data seems to be FUBAR.

[Solm]

Hi Todd -

Not sure what files you have or the amount of data you've got from Goob. But as the maintainer of IIwiki, I would be happy to lend my expertise in any way that I can / you need. Just to give you some perspective, though. A backup of IIwiki's database (compressed with gzip .gz) is 693MB. This is generated weekly, so that's about 10,000 articles there. NSwiki had about 50,000 articles or so iirc, so you are looking for a compressed file of about 3.5GB.

EDIT: I would also be willing to help host it as well, if we ever come to that.

[A Colour Out of Space]

The (broken) archive of the 2013 data is here: http://capitalistparadise.com/nationsta ... ch1.sql.gz

It does not contain user specific information besides some email addresses and ip addresses (from the edit history). It's what we've been discussing as being FUBAR.

I also managed to get a separate 2008 copy of NSWiki.net, which was much larger in size (400mb compressed), and is totally there...except being 5 years out of date. That version is hosted at http://archive.nswiki.org

[Todd McCloud]

Hi Todd -

Not sure what files you have or the amount of data you've got from Goob. But as the maintainer of IIwiki, I would be happy to lend my expertise in any way that I can / you need. Just to give you some perspective, though. A backup of IIwiki's database (compressed with gzip .gz) is 693MB. This is generated weekly, so that's about 10,000 articles there. NSwiki had about 50,000 articles or so iirc, so you are looking for a compressed file of about 3.5GB.

EDIT: I would also be willing to help host it as well, if we ever come to that.

I see. Well, is there any way to obtain a backup from anyone besides the host? If not, it look like we're SOL, at least as of right now.


What I can ask of people now is to nicely contact Goob. We have an idea as to why the file's no good, and if he still has it (I'm hoping he does and that it just copied wrong for some reason), we'd definitely be interested in it. I do know Goob said he'd be okay with hosting it way back in January, but his biggest issue was the spam (it got kind of bad there towards the end). That gives me hope that the data's still good. Perhaps he just needs some time to get around to re-copying it? I don't know.

[Omigodtheykilledkenny]

I thought the old site was nswiki.net, not nswiki.org.

It was.