[zydron]

it sounds so little, if you say it that way :P

[Myatu]

Oh, just a minor glitch then... Only 60 countries affected

[zydron]

I'm proud of my fellow dutchmen (not -_-') (RIPE NCC is dutch)

[darkfyre]

This made me giggle:

http://www.networkworld.com/news/201...ernet-for.html

RIPE NCC broke a lot of routing yesterday and I figured it was relevant to this thread

"Over 3,500 prefixes (announced blocks of IP addresses) became unstable at the exact moment this 'experiment' started," wrote Earl Zmijewski, a general manager with Internet security firm Renesys. "Not surprisingly, they were located all over the world: 832 in the US, 336 in Russia, 277 in Argentina, 256 in Romania and so forth. We saw over 60 countries impacted."

[Thelen]

Even the best and most experienced have problems every now and then - OVH is no exception,

I think the biggest question to ask isn't "do they have outages", but instead "how sevre are they, how long do they take to respond, how transparent are they about the outage".

A company with twice as many outages will still appear better as long as they act professionaly and communicate with their clients.

For example, SL has outages all the time (due to multiple datacenter with huge distance in between), but they only affect a small number of clients and they are always up front about it all.

[DigitalDaz]

I decided to give Netdirekt a shot for a month being as I was struggling to get an IP address here that I could send mail from.

I have to say, I have had already had two outages where I have been connected to the server when it happened. I've also had a very busy month and haven't really been doing any work on it.

I keep trying to move away from OVH but always end up back here.

I have our company VOIP on a little Kimi C05G, its not far off its first birthday now, October, I think. Apart from the routing issue yesterday which lost us about 30 minutes, maybe a bit less, we have never had a single issue with it. Its a telesales office so I would soon know if there was any issue, they are a right fussy bunch

Now, if only my email server was as reliable

[darkfyre]

Very much so.

For example:

Ive had three different problems with LW over the past two days.

2 days ago - lost connectivity to my servers in Germany
1 days ago - lost routes between my LW servers

Today - They have reported two major issues reported via their NOC, core routers dropping BGP sessions and network problems in Evoswitch.

OVH have been much better in my experience, ive had one serious routing issue which is still ongoing, apart from that only the odd dropout every blue moon.

Every ISP has routing / network problems... I remember back in 2007 I had a server on PEER1 lines .. they had a 100% uptime promise at the time

On Friday, January 12, 2007, from approximately 3:00 to 3:35 PM PST, one of our core routers experienced a hardware failure when the routers clock-scheduler card malfunctioned. Because the router stayed up, HSRP didn't failover, which resulted in traffic being accepted without valid routes.

Engineering was immediately dispatched to diagnose the event and were able to resolve the cause of the incident. We are continuing to troubleshoot in order to prevent similar situations in the future.


PEER 1 regrets the inconveniences associated with this incident, and will stand behind its Service Level Agreement.

Even the best and most experienced have problems every now and then - OVH is no exception,

[Thelen]

Yes it is very common. FDC is even worse. LW had its share of problems.

Once you get above 300-500Gbps things start to get VERY dicey, especially with older hardware (more points of failure per Gbps).

The top networks are far far smaller, ie SL has only about 400-600Gbps, only half roughly as OVH/LW/FDC and they charge a heap more as well (so can have better gear and management).

They have somewhat skimped if you look at some of Oles posts, but still some of the issues are out of there control (ie the gear not doing what it is supposed to be able to do).

[Euan]

For a DC the size of OVH to have so many router / switch / routing issues?

I have been with a few in the past and never had such frequent replacements / failures on networking devices. Has OVH skimped on buying industry standard equipment and gone for the cheap?