[Ubuntu/Chrome] Massive latency post-patch - resolved.

"Hello, IT - have you tried turning it off and on again?"

Post-patch, I was experiencing tremendous latency in combat. By latency I mean that everything on the battlefield would hang for 3-5 seconds, resume, and maybe 10-15 seconds later, this process would repeat, with obvious delays on commands.

Diagnostics (ping, tracert, bandwidth metrics) returned no obvious issues with my connection. I reproduced the latency consistently across a handful of battles.

Jumped on my Windows 7 laptop, experienced no such issues. Noticed Chrome was 2 versions higher (30.x vs 28.x on Ubuntu).

Updated Chrome and the headers on Ubuntu (12.10 - headers formerly 3.5.0-39, now 3.5.0.42), everything resumed functioning normally.

Didn't attempt to reproduce the issue after testing on Windows for absolute confirmation (shame on me, apologies), but just in case anyone else runs into this, that's what I did.

Hello Xanath,

Chrome has had some issues lately with its flash plugin. Can you tell me which version of the pepflash do you have installed?

PS: One more thing. I thought in Ubuntu the laster flash version was 11.2. Did it update?

Current version is 11.9.900.117; it did in fact update, my previous was 11.2 r202.

Xanath wrote:

Current version is 11.9.900.117; it did in fact update, my previous was 11.2 r202.

You are probably experiencing this one https://code.google.com/p/chromium/issues/detail?id=309540 . We can only wait for google to update their flash plug in or disable the pepflash plug in. Another solution is to try flash from another browser.

The issue resolved after updating Chrome, headers, and Flash to 11.9; it wasn't degradation of framerate, it behaved much more like latency (or packet loss, the more that I think about it). I came and posted in case anyone else runs into this or if maybe it could alert you guys to anything.

On the "steps to reproduce" page in your link, my performance is as expected (no degradation).

When the battle would "hang," I still had UI functionality, could scroll around the battlefield, and most units would "run in place." Once it "caught up," everything would warp to the expected location from the beginning of the hang until current. Orders issued during the hang were not typically reflected, and any orders issued immediately before or following the hang were accompanied by a delay (abnormal time taken to reflect the click with either the green movement or red attack line).

Xanath wrote:

When the battle would "hang," I still had UI functionality, could scroll around the battlefield, and most units would "run in place." Once it "caught up," everything would warp to the expected location from the beginning of the hang until current. Orders issued during the hang were not typically reflected, and any orders issued immediately before or following the hang were accompanied by a delay (abnormal time taken to reflect the click with either the green movement or red attack line).

Oh now I understood what you said. What you are describing suggests that there is a network lag issue indeed. Although it's strange that it didn't happen in your windows laptop.
We don't have a network issue for sure, so it is something else. If kernel update resolved this issue, I'm out of words