[jonlewi5]

Sure it went down and these things do happen.

The problem was *why* it went down. A logging server for anti-scan protection system getting a full hard disk? That's... amazing.

Thats exactly what i asked earlier in the thread, but got no response.

[_Lemon_]

What are the chances of their entire network going down? Not that high. This is probably the first time it's ever happened. OVH isn't the only DC to go 100% offline. I've seen LayeredTech go offline, and DediWebHost, both whome I had a server with for some time. They were very unreliable having only 1 tier 1 provider (LT had Savvis and DWH had Cogent). Not the same I know, but you get my point.

Sure it went down and these things do happen.

The problem was *why* it went down. A logging server for anti-scan protection system getting a full hard disk? That's... amazing.

[NickW]

... Darn rats ...

I already posted pictures. Oles has the rats covered :P.

[freshwire]

... Darn rats ...

Do you mean tasty rats?

[Myatu]

What are the chances of their entire network going down? Not that high.

... Darn rats ...

[Andy]

What are the chances of their entire network going down? Not that high. This is probably the first time it's ever happened. OVH isn't the only DC to go 100% offline. I've seen LayeredTech go offline, and DediWebHost, both whome I had a server with for some time. They were very unreliable having only 1 tier 1 provider (LT had Savvis and DWH had Cogent). Not the same I know, but you get my point.

[Thelen]

Yea ISP/DC that host their own site and services from their own connection is very stupid and causes lots of lost business.

It really wouldn't cost much to get a CDN based host, aside from great uptime you'd then also have much better speeds and ping response, giving a better view of the company.

But whatever, I'll let you mis-manage your hundred million dollar company

[Neil]

I had those issues aswell, even now 5-10 mins ago it happened again. From what I can tell it was again a whole range that went down.

2012 is coming close now, skynet is taking over

Hi, that an issue with the router see http://status.ovh.net/?do=details&id=287 - this is caused by the ARP issue.

Skynet is already online and running, http://en.wikipedia.org/wiki/Skynet_%28satellites%29

[Tz-OVH]

marks, seeing as any trouble at your DC will lead to the entire ovh domain being left inaccessible, perhaps its not a bad idea to host your website/forums off-site?

[mvegetto]

I had those issues aswell, even now 5-10 mins ago it happened again. From what I can tell it was again a whole range that went down.

2012 is coming close now, skynet is taking over

[marks]

This is the translation in English:

http://status.ovh.net/?do=details&id=282

Unfortunately, yes, there were 2 cuts:

30 minutes (Jul 12 05:31:54 / Jul 12 06:01:31)
25 minutes (Jul 12 06:45:13 / Jul 12 07:10:40)

The situation was corrected very quick, but it's clear that things like these shouldn't have ever happened.

[HandsomeChap]

Ah was wondering why I had some outages last night

[jonlewi5]

Excuse my ignorance, but reading through travaux using chrome to translate, but my understanding was that a script brought down the ovh network, is that right?
If it is, its a bit concerning that a rogue script can kill the network?

Of course, my understanding may be waaaay off the mark

[Andy]

So the WHOLE datacentre went offline twice for 30 and 25 minutes a time?

[marks]

yes, there was a short time downtime:

http://travaux.ovh.net/?do=details&id=4366 (the English translation will be posted very soon).

Also, more updates will be posted with what measures will be taken for this not to happen again

[Andy]

Did anyone else have world-wide connectivity cuts throughout the night? My server was reported down from London, USA, France and Sweden through pingdom.com and through friends. This happened 5 times for 20 minute intervals throughout the night. Probably the worst nights sleep I've had in a while with text alerts coming left right and center!

Anyone from OVH vouch for what happened?

Cheers.