Thanks to all who worked on this for their efforts!
Over the years, I kept a log of the thread titles and their numbers for every thread I posted to, beginning Aug 5, 2008. Putting the thread number in the Archive Viewer search box worked for one thread, but four other thread numbers checked all failed with a message like this:
Thread 2502435 is indexed, but its markdown object is missing.
What does this error “markdown object is missing” mean?
This:
It’s a lot of data to process.
First of all, thanks to everyone involved!
I think something went missing along the way, though; the post ID. For instance I have found this link here:
http://ubuntuforums.org/showpost.php?p=12397979&postcount=9
which seems to reference a post directly, without thread ID. I too had noticed that the thread id was part of the file name, but the post ID seems nowhere to be found; looked inside some random posts to see if one could grep for it, but no dice. 
Any chance the old database is still around to get the post ID’s too?
Perfectly normal, just some posts got moved around.
12397979 was a post by acidreign which got merged into a different thread in between being linked on the UEFI page, and the site being shut down.
It’s the post by acidreign on 2012-12-10 in thread 2091348.
Yes, but the problem is that you can’t know every single one of them. 
I think the post ID should be something one can search for; it’s the ID criterion after all, even if posts get moved around; it’s like an inode in a filesystem, which can have multiple (hard) links pointing to it; a thread being the equivalent of a leaf directory.
Sure. It was a hastily hacked together fun way for @paddylandau to find his favourite threads. Not a robust, well-designed archive. There are gaps in functionality.
I wonder if it would be possible to just convert the old forum, if it’s still around, into a (flat) filesystem tree, in which TID’s (thread ids) are the primary directory names and the titles are symlinks to them. Posts would just be files with their PID (post id) as the start of the filename. To view a full thread it should then suffice to do the equivalent of:
cat $FORUM_ARCHIVE_ROOT/TID/*
(redirected to a file named like the ones in the GH archive now)
Since PID’s are usually monotonic, that should be in the correct order of posts in a thread; might need some zero-padding, though, because:
$ mkdir "$TMPDIR"/test
$ cd "$TMPDIR"/test
$ touch {1..10}
$ ls
1 10 2 3 4 5 6 7 8 9
# comment: since it's the shell that's doing the globbing in the cat command
$ echo *
1 10 2 3 4 5 6 7 8 9
# but there could be another way:
$ seq 10 | sort -n
1
2
3
4
5
6
7
8
9
10
# so make that:
find $ARCH_ROOT/$TID -not -name 'thread-*' -print0 | sort -nz | xargs -0 cat > $ARCH_ROOT/$TID/thread-$TID-<title>
# but for that to work, the post filename MUST NOT contain any non-numeric chars :/
> printf "%s\n" post-{1..10} | sort -n
post-1
post-10
post-2
post-3
post-4
post-5
post-6
post-7
post-8
post-9
Thank you, Aaron! That is amazingly useful to allow “web-walking”.
Is there any way that the search box could handle multiple words as separate parsing strings, similar to using sequential greps? That would really go a long way. The existing search box joins the distinct strings into one long string, which was not the intent. This request is not looking for implementation of a quoted pattern, such as “this that”.
Thank you for your consideration.
I’ve updated the site to provide a browsing experience, in addition to posting old URLs or thread IDs.
Full text search is next.
Thanks a lot to everybody who has contributed to this revival of the Ubuntu Forums as an archive!
I have tested some of my favourite threads at https://ubuntuforums.popey.com/, and they work well. I would be happy to modify the references (for example here and at AskUbuntu) so that they will work again.
Is this ‘only’ a temporary web site during the archiving process? In other words, are there plans that the Ubuntu Forums archive will be moved to some other website (for example run by Canonical, maybe even using the original web address)?
I have no plans to shut it down again. I will clean up the code and put it somewhere when I am a bit less busy. The goal to be to make it possible to contribute to the site, and in case anyone else wants to stand up their own copy. It’s not hard to do. I’ll make sure it’s documented.
Wished I’d have copied the information I wanted before the Ubuntu Forum shut down lesson learned. (sigh)
So far everything I’ve searched for seems to be unavailable. 
Perhaps it will be available in the future. 
So true that all good things come to an end.
Have you looked on the internet archive? There’s a lot of stuff there.
https://web.archive.org/web/20260000000000*/https://ubuntuforums.org/
If there are whole sections of the forum that are missing, I’m sure it’s possible to yoink posts out of the Internet archive and backfill anything into our on data.
I have not as I did not know there was Ubuntu forum threads there so I will give that a look at.
Thanks @popey
The thing is that we were told, before the shut-down, that Ubuntu Forums would remain open to the public, albeit as read-only. It was quite a shock when it was removed from the public eye without any warning.
You have hackers to thank for that.
It really had to be shutdown due to attacks .
I hate hackers and scammers with a passion. The worst things after politicians.
Couldn’t the forums be put into a read-only mode on an immutable basis?
What has been done so far is all that is going to be done except possibly for someone making them easier to access.
Anyone can download the files themselves.
Even had this been done, the forums themselves were still vulnerable to any number of vulnerabilities on the underlying OS or the platform itself. That’s why ‘hackers’ is a problem. Because even in an ‘immutable read only state’ the actual vulnerabilities wouldn’t be patched or sealed off, hence the extra problems.
I think the only option safe left was as is done here - someone to download the files themselves or make a ‘searchable’ tool version like popey did.
Unfortunately from a security perspective, anything less than taking down the forums completely would leave security holes that Canonical couldn’t accept (and the broader security community would raise red flags over).
It sure was a shock when it just disappeared.
