Euan |
1 person marked this as a favorite. |
I agree with jemstone. Whatever update was implemented, has clearly broken the site in a complex manner. You've spent at least one persons time all day yesterday trying to find a fix, which was to be rolled out first thing this morning from what we were lead to believe.
So either the fix didn't work, or something else is wrong.
I owned a web development company for 15 years, and we'd not leave a site like this for a day if we could avoid it. We'd back up, get a functional site back out there, and figure out what we did wrong on a development server, not the production server.
That said, I wish you the best, and good luck finding the needle in the haystack! It's got to be frustrating, and I sympathize - having been there before.
Good luck!
Erik Keith Software Test Engineer |
3 people marked this as a favorite. |
Has the question been asked as to why we can't just roll back to a previous version and then take the time to debug the changes before re-implementing them?
I mean, I'm sure it has, but years of working in a development environment have conditioned me to ask, regardless.
We were expecting to have it resolved in a relatively short timespan, one of the fixes ended up taking a bit longer to implement than we had hoped drawing the process out a bit. At the moment it looks like we have everything fixed and I'm doing a quick sweep to make sure I don't run into any random encounters. I tend to be a little meticulous, please hang in there everyone. :)
jemstone |
jemstone wrote:We were expecting to have it resolved in a relatively short timespan, one of the fixes ended up taking a bit longer to implement than we had hoped drawing the process out a bit. At the moment it looks like we have everything fixed and I'm doing a quick sweep to make sure I don't run into any random encounters. I tend to be a little meticulous, please hang in there everyone. :)Has the question been asked as to why we can't just roll back to a previous version and then take the time to debug the changes before re-implementing them?
I mean, I'm sure it has, but years of working in a development environment have conditioned me to ask, regardless.
Awesome, glad to hear it.
I do suggest that rollbacks be considered in the future for this sort of issue. Dropping back ten yards and punting (metaphorically speaking) is usually better than sacrificing the site for so long.
CrystalSeas |
3 people marked this as a favorite. |
At the moment it looks like we have everything fixed and I'm doing a quick sweep to make sure I don't run into any random encounters.
Also missing: the indication of which forum the post is from.
So the information about which forum the thread is from (when you have the page "flattened") is not coming back? It was useful because the titles people use on their threads are not always informative.
At least if I knew which forum it was in I could decide about checking it out.
Also, it no longer shows if a thread has been moved by staff from one forum to another, which was useful information as well.
"The House On Hook Street" might be in the GM forum, or the Recruitment forum, or the Advice forum or the Products. It was especially useful to be able to distinguish recruitment threads from other threads.
Guy St-Amant |
1 person marked this as a favorite. |
jemstone wrote:We were expecting to have it resolved in a relatively short timespan, one of the fixes ended up taking a bit longer to implement than we had hoped drawing the process out a bit. At the moment it looks like we have everything fixed and I'm doing a quick sweep to make sure I don't run into any random encounters. I tend to be a little meticulous, please hang in there everyone. :)Has the question been asked as to why we can't just roll back to a previous version and then take the time to debug the changes before re-implementing them?
I mean, I'm sure it has, but years of working in a development environment have conditioned me to ask, regardless.
I would +1 the roll back idea if it can't be fixed before going home tonight. and if roll back is impossible, a warning in the header might be nice.
Steve Geddes |
1 person marked this as a favorite. |
jemstone wrote:We were expecting to have it resolved in a relatively short timespan, one of the fixes ended up taking a bit longer to implement than we had hoped drawing the process out a bit. At the moment it looks like we have everything fixed and I'm doing a quick sweep to make sure I don't run into any random encounters. I tend to be a little meticulous, please hang in there everyone. :)Has the question been asked as to why we can't just roll back to a previous version and then take the time to debug the changes before re-implementing them?
I mean, I'm sure it has, but years of working in a development environment have conditioned me to ask, regardless.
Thanks, Erik.
I'm so glad I've got my job and not yours. :p
Ed Reppert |
1 person marked this as a favorite. |
Interestingly, the one pbp thing I was in "ground to a halt" several months ago. This morning (actually a couple days ago, I missed it) there was a note from the GM about it, telling us we'd pretty much finished it anyway and were "almost ready to level". :-)
jemstone |
1 person marked this as a favorite. |
Thank you for your hard work, Erik. Much appreciated.
Please accept this brand new Gorillaz track as a token of my thanks.
Out of curiosity - are you able to share what sort of version control software/system you're using? I have a burning curiosity, as my Operations Specialist days, while behind me, have conditioned me to wonder about these things.
Chris Lambertz Community & Digital Content Director |
NobodysHome |
3 people marked this as a favorite. |
Being a programmer myself, I sympathize. The guys are probably dealing with about a million lines of code. Even a comma in the wrong place will break things. It's like looking for a very small needle in a hundred haystacks.
My favorite classroom incident of all time: I was rather infamous at my real-time company as being the author of the device drivers course, as well as being the only one insane enough to teach it. ("Lesson 1: Let's all introduce ourselves. 15 minutes. No practice. Lesson 2: Here's how you initialize a device and get it to interface with our OS. 90 minutes. Practice 2: OK, build your device driver. 12 hours.")
So I was very, very good at debugging the raw C code students were using to code their drivers, until one student flummoxed me for a good 45 minutes. That's pretty hard with only a few hundred lines of code.
We were finally doing a line-by-line comparison with the solution when we found it: One stray period the compiler wasn't complaining about, but that was breaking the code.
45 minutes to find a single period in the wrong place is still a record for me. Usually compilers are better at finding such things when you run 'em in full debug mode...
Unreliable Narrator |
6 people marked this as a favorite. |
Fixes should now be in place for the display of New Posts as well as the My Campaigns Tab.
In the frozen room of Serv'Or, Sir Erik slew the troublesome pugwampis. And there was much rejoicing.
Erik Keith Software Test Engineer |
Thomas Seitz |
1 person marked this as a favorite. |
Fixes should now be in place for the display of New Posts as well as the My Campaigns Tab. I'm looking into the x icon issue currently, as its definitely not working as intended.
It's pretty clear I need to spend more time talking to myself when I'm in the Test Platform.
But the important thing is Krampus is still in his box right Erik?
Erik Keith Software Test Engineer |
1 person marked this as a favorite. |
Erik Keith wrote:But the important thing is Krampus is still in his box right Erik?Fixes should now be in place for the display of New Posts as well as the My Campaigns Tab. I'm looking into the x icon issue currently, as its definitely not working as intended.
It's pretty clear I need to spend more time talking to myself when I'm in the Test Platform.
Oh totally, who would make such a rookie mistake as to let- OH. OH NO.
On that note:
1. The 'X to close' feature will now dismiss the New Post commentary. It won't update in real time when you do, I'm getting that documented for future revision.
2. Subforum information should now be posted during flattened messageboard viewing. (Thanks CrystalSeas!)
CrystalSeas |
2 people marked this as a favorite. |
One more missing piece:
There used to be a 'null' icon (a zero with a slash through it) that let you hide individual threads.
My bad. It's there, but it's not working correctly.
When I click it, the title is hidden (and two 1 pt rules are shown), but it refreshes itself and the topic shows up again. You can't hide threads. When you open a tab with that page, the thread is back
And yes, thanks for the fast fix on the forum information next to the threads
John Woodford |
1 person marked this as a favorite. |
One more missing piece:There used to be a 'null' icon (a zero with a slash through it) that let you hide individual threads.
My bad. It's there, but it's not working correctly.
When I click it, the title is hidden (and two 1 pt rules are shown), but it refreshes itself and the topic shows up again. You can't hide threads. When you open a tab with that page, the thread is back
And yes, thanks for the fast fix on the forum information next to the threads
I found the same thing, but if I navigate to the subforum instead of Online Campaigns or Messageboard I can still hide threads.
rknop |
1 person marked this as a favorite. |
45 minutes to find a single period in the wrong place is still a record for me. Usually compilers are better at finding such things when you run 'em in full debug mode...
Back in 1989 or 1990 I was coding C on a Mac. (So, we're talking like a Mac II here.) The editor didn't do any syntax highlighting.
I spent the better part of an evening having no idea why things weren't working. It was a missing */ -- which meant that a bunch of code was commented out, but I didn't realize it.
Syntax highlighting is a nice thing.
Gisher |
1 person marked this as a favorite. |
NobodysHome wrote:45 minutes to find a single period in the wrong place is still a record for me. Usually compilers are better at finding such things when you run 'em in full debug mode...
Back in 1989 or 1990 I was coding C on a Mac. (So, we're talking like a Mac II here.) The editor didn't do any syntax highlighting.
I spent the better part of an evening having no idea why things weren't working. It was a missing */ -- which meant that a bunch of code was commented out, but I didn't realize it.
Syntax highlighting is a nice thing.
Yes, it is.
John Napier 698 |
1 person marked this as a favorite. |
NobodysHome wrote:45 minutes to find a single period in the wrong place is still a record for me. Usually compilers are better at finding such things when you run 'em in full debug mode...
Back in 1989 or 1990 I was coding C on a Mac. (So, we're talking like a Mac II here.) The editor didn't do any syntax highlighting.
I spent the better part of an evening having no idea why things weren't working. It was a missing */ -- which meant that a bunch of code was commented out, but I didn't realize it.
Syntax highlighting is a nice thing.
I know. Oh, yes, I know all about the dangling comments. Such a pain. Especially while coding on the fly. Even more so if you're coding in an OS like DOS, where the best you had was EDIT.COM when using a non-IDE standalone C compiler. Now that I've become spoiled, I wouldn't trade my Visual Studio compilers for anything in the world.