42 replies to this topic

[Draigun]

Larazard, on June 09 2014 - 01:55 PM, said:

Not sure if it's still the same problem, but it seems like me and most of my friend's list just got booted while in the middle of matches as well.

Can't imagine how frustrating this must be for the devs - wishing them the best of luck in tackling this bug.

Yeah, seems like there are still issues on their end. It could be reason for blame on Amazon's servers, as it was partly the database on HAWKEN, and the servers hosting it. Most of the poor network performance is associated with Amazon's infrastructure, for some reason.

Edited by Draigun, June 12 2014 - 01:03 AM.

[RunaPanda]

Draigun, on June 12 2014 - 01:02 AM, said:

Larazard, on June 09 2014 - 01:55 PM, said:

Not sure if it's still the same problem, but it seems like me and most of my friend's list just got booted while in the middle of matches as well.

Can't imagine how frustrating this must be for the devs - wishing them the best of luck in tackling this bug.

Yeah, seems like there are still issues on their end. It could be reason for blame on Amazon's servers, as it was partly the database on HAWKEN, and the servers hosting it. Most of the poor network performance is associated with Amazon's infrastructure, for some reason.

Regardless who's to blame, it's getting worse and worse day by day. I live rather close to an amazon data center here in Montreal, my ping to and from the center is less than 8ms. But yet not only is logins in game slowed drastically, but logins on the site here are suffering too. I got a window from 3am to 6am to play other wise it's 20+ minutes of waiting on either "Syncing Account" or "Syncing Stats". Another game I play with it too, is hosted on an AWS server I have zero connectivity issues with, yet I try hawken and it seems near impossible to play unless I want to wait forever.

[Draigun]

It may be a case of poor routing from the ISP backbone to HAWKEN services. I play with consumer average speeds, and I rarely have issues, with the occasional rubber banding here and there.

I think it is a combination of HAWKEN's services/database failures, Amazon's servers (the FQDN associated with the master database is hosted on an Amazon server), and poorly optimized routes to the aforementioned services.

However, Amazon has high-grade servers (depending on the specs, a VPS could be a different story), so I find it difficult to believe that Amazon is one of the reasons for this ongoing issue.

Edited by Draigun, June 14 2014 - 11:06 PM.