Solved

Can't get in to race

We had to postpone races from a certain time tonight due to the interruptions to the service. As you can see, we're dealing with more than just viewer loading. There are two or three overlapping situations we need to address, and as a team of three at the core each one takes our full focus, meaning we can't address the others until the one before is done. Tomorrow we will be performing major upgrades to the service back-end, this was planned from the disruptions we saw earlier in the week, and these should make a big difference to stability.

Service updates since launch until now have only been fairly minor in scope. Tomorrow the changes will be broad and significant, based on all the data we've collected. It might be that it takes us a few days to complete all of the work involved, but it should really noticeably change things for the better.

That's what I've been trying to get at. I think in most cases it's actually related to their computer. Ultimately it's our job to make sure they don't get into that situation. But, once they are in it, there's only one way to get out... locally. It's not something we can fix remotely.

That's why I've created this thread to help people really understand how to clear their cache: http://igpmanager.com/forum-thread/7158. Until now I heard of people rebooting their PC and wiping the system cache and stuff, or the browser cache but only for the last few hours. None of those things will work. It must be the entire cache for all time to be sure you've got it all. Hopefully those instructions make it clear.

There are other things we're dealing with on our end right now, and I expect the dust to settle on all of this within a few days. Everything will be in a much better position.

Again though, was yours an individual case? As in, others got in okay.

Hi Ian, the problem was resolved, see this thread: http://igpmanager.com/forum-thread/7169

Sorry about that Jackie. We've moved everything to the cloud before mobile launch and there are some unusual behaviours in these cloud servers which we have been studying and trying to find explanation for.

To give you some idea, every single time we restart the servers, a different set of things crop up that have nothing to do with any kind of consistent or logical pattern. However, I just stopped the service for a minute then started it rather than restarting it, and this does seem to have improved things.

These things appear to have nothing to do with our code, or even the work we are doing on the servers. They are just playing up in random ways with each restart. It's easily the most frustrating thing I've ever worked on. Restart it once, Facebook login stops working, restart it again and Facebook is fine but Google play has issues, restart it again and everything is fine, except one page will have {{placeholderText}} like this.

It's like playing roulette with the server. But as I said, maybe it's the "restart" option which needs to be avoided because a stop/start seems to have got things working well. Maybe that's the way forward. The servers simply don't seem to enjoy the restart option.