I got caught up in the “verification” process that they imposed and quit. So it would be nice if our community liaison did their job and posted relevant information on the official DDO boards.
Then again, divide and conquer is a novel concept the producers are trying. And it looks to be successful.
Here is a summary for you,
During the initial issues on Cormyr, Tolero writes in response to various messages:
1. (yeah that one was also set. The only setting being stubborn is the Saltmarsh one I'm not sure why yet we're still digging into that and that might take until tomorrow to figure out)
2. Yeah that one, and several of the other auto applied ones, am still investigating. We have a small lead on it though that we're chasing.
3. Yeah that is a new mystery. There is a certain setting that can cause that, but I checked that already and that's set correctly. So something more complicated is going on.
After it relaunched:
4. We also changed it so the Harbor Market and Korthos can't, you know, all end up trying to share the same exact server resource
it was a bit heavy
5. Yeah parsing the data to see what it's doing atm
6.
A little game server anatomy as ppl are checking their locs:
- The only valid numbers for server ranges are between 1 through 6 (so 21, 22, 23, 24, 25, 26; 31, 32, 33, 34, 35, 36)
- Ranges 11 - 16 are currently all devoted to the highest traffic public areas, which isolates them from touching dungeon resources
- But not all public areas are isolated in that way. So it’s still possible to be sharing a resource with something public.
- As an area loses people, it reforms later on a different resource. But “busy” areas tend to say on the same server number by virtue of people being in there often.
- Copies of the same place can end up on different servers (both public and private). The big areas (e.g. Harbor) will stay in one server but the smaller ones (e.g. The Twelve) will tend to bop around.
7. Server 22 has been struggling but I can't identify a cause yet. But Keep public is being suspiciously tame as compared to prior days. So I have some ideas of what might be happening but gotta talk to the Engineers to see if my theory holds any water
8. FYI just so folks know what I'm doing with the loc data: so I've hit an important milestone where for the last good while, i can
predict which server is about to experience a lag wave 4 mins before it happens (I take a measure, make the prediction, then confirm if the alarms and locs for that predicted server are popping off). Engineers and I will be meeting up tomorrow to unpack these findings. Thank you for posting your locs! The locs were allowing me to confirm the predictions. This is very important and a huge lead.
The Breakthrough:
9. This is the first time with this lag that I've ever been able to see it before it happened! Normally it's always after-the-fact. And there's enough quantity of servers that it's statistically unlikely for me to be "lucky guessing" that many times in a row
etc, etc