Our webhost, GoDaddy, decided to change the rules on us and has deleted the entire Biblioblog Reference Library database.
Beforehand we had enough space to keep the database working and pruned down to a size that was feasible, but in the course of the last few months, they decided to halve the allowable database size which made it impossible to house all of the data and indices. What's worse is that after requests for a full backup of the database were made, they refused to provide anything but a corrupted partial backup. We tried very hard to make the chunked backups necessary to put the entire database into a format that we could move, but in the midst of it (moving millions of records only tens of thousands at a time) they pulled the plug.
It means that we need to re-imagine the library and figure out where to go from here. The most requested feature was the blog hit counter, which we are going to work and restore as quickly as possible. The actual archive (which grew to a grand total of 15 gigabytes of archived posts, index, and blog reading trend data) cannot be resurrected.
So, stay tuned. We'll be back.
May 17: UPDATE 2
Excellent news! The blog feed scripts are so far running strong. You can check out preliminary progress here at on the Quicklist where the last 24 hours of Biblioblogging activity is listed. The Blogs List needs some pruning and I know that plenty of new blogs have started up since the Library went down, but I'll be getting to those in short order.
Good news! The next version of the Archivist software is nearly ready for prime time. Once that's up, the new Biblioblog Reference Library portal will be soon to follow.
No, no April Fool's joke. :-) Today I am picking through the pieces of the old database fragmentary backups to try and get together a new database schema for the next step in restoring the Library. Work is slow due to my other obligations (family, work, grants, and other hobbies) but I'm hoping to get something basic working by the end of this week. Keep your fingers crossed for me. - Steve
April 1 #2:
I now have a schema developed and I'll be uploading it to the server to test it out. This will require some modifications to the old Archivist and Hit Counter scripts (mostly some trimming) but if those modifications are successful, I can start harvesting hit and post data to tinker with. -Steve
April 1 #3:
A preliminary Archivist script is now ready for testing. If this works, I can begin rebuilding the database. -Steve
Well, after being on the phone for 45 minutes, it turns out that one of the reasons why the server was so God-awful slow is that it's still running on a legacy system. I've been told that I need to get rid of some old MySQL 4.1 databases (I didn't even know I still had a few lying around) and they can upgrade the server account to their "new" Grid system. -Steve
April 2 #2:
All MySQL 4.1 databases are toast, and I'm now on the phone again waiting for the upgrade. -Steve
April 2 #3:
I'm now waiting for the first server migration. This part of it could take up to 24 hours before I can continue. From there, the upgrade to the "Grid" will take another 24-72 hours, and from there we can start things cooking. :-) -Steve
The first step of the migration is complete; the second step is on its way. -Steve