Game server had about 15 people in it, and all but a few dropped.

Uplink Router: CAR2-4.DLLSTX2

Time: time of this post


I personally didn't lose connection to the server...

Looks like broadwing is having issues again, seems to be cleared up for now, you can see spike in mrtg.

Yea im getting destroyed by it to, except im coming accross on gblx

Spoke too soon - still having issues over broadwing.

ibr2.dllstx2 - DOWN
car1-2.dllstx2 - DOWN

Looks like every connection except AT&T...

Now that mrtg has updated, it looks like a good sized ddos..

As long as I come in over sprintlink all is good, but looks like the routes keep changing, and knocking over to horrible routes.

Yeah, I'm showing problems here too. They're intermittent, and I'm trying to do cPanel updates, too :S

Looks like Spring#1 went down and transfered it all to Spring#2, maxing it out... that would be connections going out, that is.

I have customers reporting issues as well.

Check out the inbound on globalcrossing - nice amount of traffic coming in there, def not normal traffic failing over.

Giant spikes everywhere... I don't think it's DoS because it'd show more incoming connections..

Look at the network overview. ibr2.dllstx2 crashed. That's the problem. Layer3 and Sprint are out of the mix for the meantime.

Ah. Hope it gets resolved soon

Hmmm..... Everything is normal according to my server monitors but I'm not sure. I thought GigE connections were redundent. Hopefully, whatever it is, is setting alarms off all over the NOC and they are too busy to check the forums.

I called, he said 15 minute ETA on the fix

My thoughts exactly. I opened a ticket to see if I could get a definate answer, but not sure when I will...

EDIT - Great! But 15 minutes can mean a lot to some people

Ah it looks like they are having problems with a part of their network, if you look at the network map:
http://www.servermatrix.com/about_us.html#network
Everything coming into the second M20 (sprint x2 and level3 x2 and probably broadwing as its not on diagram) is down, everything else is failing over for them being down.

Couldn't get onto SM's forums there for a little bit. Orbit was working at the time, but now Orbit is down and this is back up...

Problems here also.

I "upgraded" from my SM Dual Xeon to the TC2800 about 2 months ago. My SM was in dllstx4, my new TC2800 is in dllstx2. I'm starting to miss my old SM...

Funny.. seems that dllstx4 has more network issues than dllstx2

Also, I called them and they also told me about 15 minutes... we shall see.

Seems that distribution switch is back up.

Wait, I was wrong. I just assumed and we all know where that leads you.

My TC 2800 is in 4, but it seems to be having problems as well. 2 is/was the rox!

SiRi

This wasn't specific to one datacenter or the other, as the network for both runs through the main routers. One of the routers went down and the "downtime" was while routes had to reconfigure.

They are. If they weren't, you would show a downtime of a LOT longer than we experienced. However, router reconfiguration isn't instant and it takes a bit for things to work themselves out when a router goes down.