All North American Data Center Worlds Network Technical Difficulties (Oct. 26)

[worldofneil]

You appear to have a very idealistic view of consumer power in this instance.

Perhaps, but you also seem content with forcing an unrelated company into fixing a problem caused by others... Guess we'll have to agree to disagree (and I'm home now and my ISP doesn't use L3 so I can/have logged in and I want to go and play rather than keep trying to explain things on these forums!)

[Chadamius]

http://www.nytimes.com/2016/10/22/bu...tack.html?_r=0

[whiskeybravo]

http://www.nytimes.com/2016/10/22/bu...tack.html?_r=0

That's from last week. Though, it could be similar for sure

[ChaoticCody]

That's from last week. Though, it could be similar for sure

Reference my post above. This is a Level3 issue. It's most likely not a Ddos attack.

[whiskeybravo]

Reference my post above. This is a Level3 issue. It's most likely not a Ddos attack.

I don't see how a faulty fiber/equipment in Houston would affect people all over the country. Certainly makes bad infrastructure look worse if that is the case lol.

[Chadamius]

Yes, it was 6 days ago...

(The outage map is the exact same as current. Coincidence?)

In the article they explain:

"Such attacks are common, but there is evidence that they are becoming more powerful, more sophisticated and increasingly aimed at core internet infrastructure providers."

They also explain in the article that the reasons for such an attack could be political. (Prevent early Voting, preventing the democratic process, etc..)

(It could be a test to see how effective their hacks are just before the election. And then mount a larger attack closer to election day.)

[LillithCale]

I don't see how a faulty fiber/equipment in Houston would affect people all over the country. Certainly makes bad infrastructure look worse if that is the case lol.

Please see this explaination here:

This isn't a perfect explanation, but it should give you some idea of who and what Level 3 are.

Level 3 is one of a very few Internet 'Backbone' companies that provide the major data transmission services that are the foundation of the Internet as it stands now. These kinds of company have fiber optic networks for moving huge quantities of Internet traffic around the globe very efficiently. ISPs - Internet Service Providers - interconnect with companies like Level 3 to ensure that their customer's traffic is efficiently routed across the internet to the desired destination. Very large ISPs like AT&T, ComCast, Charter/Time Warner and others maintain their own backbones that cover part of the service, but they interface with companies like Level 3 to connect their backbone to the rest of the world. These agreements are called peering agreements because both parties agree to operate as peers, and allow traffic from their backbone to run across their peer's backbone infrastructure.

So, if you think of all the customer network links as twigs on branches. The branches are the ISPs, and the large branches and trunk of the tree are the network backbone providers like Level 3. Since more than one branch is affected by this problem, the problem is not specific ISP branches of the tree, the problem is in the trunk of the tree.

[whiskeybravo]

Please see this explaination here:

Very familiar with who Level 3 is. Doesn't invalidate my point of making a bad infrastructure look worse if one faulty cable can shut down half the world's internet access. Billion dollar industries should have better contingency plans. Or you know, re-invest in the infrastructure instead of handing out bonuses for increasing the bottom line.

[Geist]

Reference my post above. This is a Level3 issue. It's most likely not a Ddos attack.

Most definitely a routing/service outage, and cut or failing fiber at a focal point can be really nasty with all the added technical complications you get at this order of magnitude (and with so many people supposedly being affected downstream).
I'm rather surprised that they didn't immediately escalate this, though - I work in IT as a profession, and my hometown houses one of the main regional/national network hubs for the Italian territory. Even here, the emergence of such a single point of failure would have raised quite a few eyebrows...

[Kosmos992k]

Perhaps, but you also seem content with forcing an unrelated company into fixing a problem caused by others... Guess we'll have to agree to disagree (and I'm home now and my ISP doesn't use L3 so I can/have logged in and I want to go and play rather than keep trying to explain things on these forums!)

Fine, go play, but don't try to tell those who are affected that they should change ISP, since at last count the ISPs involved include AT&T, Comcast, Cox Communications, Charter and Time Warner (add RCN to that list). It's very nice that YOU can play, lots of others can't.

Changing ISPs isn't an option. Calling your ISP is perfectly fine, making them understand the issue is not. I know it's not because I have contacted my ISPs, I have been through this before as well, and it was no different then. Quite how you think a relatively small number of users of a relatively small online game are going to force their SPs to do something is beyond me, it's a David and Goliath situation, although in this case David is about 3cm tall, and Goliath is looking down on the top floors of skyscrapers.

Most definitely a routing/service outage, and cut or failing fiber at a focal point can be really nasty with all the added technical complications you get at this order of magnitude (and with so many people supposedly being affected downstream).
I'm rather surprised that they didn't immediately escalate this, though - I work in IT as a profession, and my hometown houses one of the main regional/national network hubs for the Italian territory. Even here, the emergence of such a single point of failure would have raised quite a few eyebrows...

I completely agree. It's surprising that such a vital link has a single point of failure. These networks are typically designed to remove single points of failure. But even with a single point of failure, I can't understand why they haven't re-routed across partner networks while they fix their stuff.