01 Oct, 2006, Guest wrote in the 1st comment:
Votes: 0
The core code behind the site has been upgraded to the latest QSF code. There may still be some residual problems associated with this so keep an eye out for them and report them if possible.
01 Oct, 2006, Guest wrote in the 2nd comment:
Votes: 0
You should have used a live and test service to make this upgrade seamless.
01 Oct, 2006, Davion wrote in the 3rd comment:
Votes: 0
If you mean have a test site and a live site, we do, and use it that way. This is simply a mirror of our development site, but still a tad behind it. Don't know when we'll bring all the additional features online here though. Give us a bit longer to test!
01 Oct, 2006, Guest wrote in the 4th comment:
Votes: 0
Well, what choice do I have, really, but I think that the site being down last night was pretty unprofessional.
01 Oct, 2006, Davion wrote in the 5th comment:
Votes: 0
Well, in the process of updating the core of the entire site, as well as making our upgrades, we felt it best to lock the site up while we did this to avoid any unforseen problems. At any rate, I believe it was only 30 minutes or so. But still, appologies from the MudBytes staff! We will try harder next time.
01 Oct, 2006, Guest wrote in the 6th comment:
Votes: 0
It gives no message, just a "debug_mode" message generated by Quicksilver. Why not replace the message with a little game while you wait.
01 Oct, 2006, Guest wrote in the 7th comment:
Votes: 0
Chalk that up to unforeseen issues. It was supposed to put up a nice little message saying we were down for upgrades, but obviously it didn't so that made it look really bad. At least for the 3 minutes that message was being thrown.

Upgrading forum software isn't always a simple case of throw php files at it until it's fixed. Mudbytes is a partially custom built wrapper to the QSF core code so upgrading the site requires at least a two step procedure: Update the core files, then update our changes to the core files on top of it. Which is why we have a dev site, so it can be done their first to work out the major showstoppers.

Then you have database updates on top of all that for when new features are added, or expansions to existing ones. Letting the site be accessible while those database fields are missing would have been just as unprofessional as forgetting about the debug_mode flag that was added to QSF.

Then the skin needs to be exported from dev, imported here, and checked to make sure we aren't providing interfaces to anything else we're still working on.

At any rate, we do apologize for having the site down longer than expected. I can't promise it won't happen again but we'll try and keep a better handle on it in the future.
01 Oct, 2006, Guest wrote in the 8th comment:
Votes: 0
Why not return the mudlist as a single index on the "oops we're down right now, go play a mud" page?
01 Oct, 2006, Conner wrote in the 9th comment:
Votes: 0
I think that having the site down for a half hour is hardly the end of the world, especially for a core upgrade. While I agree that a nice temp page showing the current mudlist links would be a cool alternative, I have to wonder if you'd still be complaining just as much if it had been due to power outage at their host or some other reason totally out of their control.
01 Oct, 2006, Guest wrote in the 10th comment:
Votes: 0
locke said:
Why not return the mudlist as a single index on the "oops we're down right now, go play a mud" page?


Suppose it was the mudlist itself which was the major focus of an upgrade? Really. It's just best for everyone for the site to be down for a few minutes to allow us to get things sorted out. There are always bugs with any deployment no matter how much testing is done and to think otherwise, especially as a programmer yourself, is just plain silly.
01 Oct, 2006, Conner wrote in the 11th comment:
Votes: 0
Samson said:
locke said:
Why not return the mudlist as a single index on the "oops we're down right now, go play a mud" page?


Suppose it was the mudlist itself which was the major focus of an upgrade? Really. It's just best for everyone for the site to be down for a few minutes to allow us to get things sorted out. There are always bugs with any deployment no matter how much testing is done and to think otherwise, especially as a programmer yourself, is just plain silly.


True enough, but if the webserver itself and the mudlist module are both still working, it's not a bad idea.. or even just a temporary page with links to the muds currently on the mudlist as "suggested places to visit while finish getting the site updated". Otherwise, having the debug page is hardly the worst you could do.
01 Oct, 2006, Guest wrote in the 12th comment:
Votes: 0
It's all moot now anyway since the upgrade only took a few minutes. The only problem was the missing setting, which I corrected within 3 minutes of noticing. Even us admins can't get past those :P
0.0/12