Network Failure

From status.dreamhost.com:

In reply to:

---

Wed, 12 Apr 2006

17:49 PST - Major Network Failure
A large portion of our network just dropped offline. It's impossible for us to tell exactly which servers are affected but we are working as quickly as possible to remedy the situation.

---

What do they mean they don't know which servers are affected?! What are we paying them for?? Doesn't our money pay for admins? Oh that's right.. our money went in for the 10 year party and video cameras to capture throwing servers off buildings.

Well, considering the sheer amount of servers dreamhost do have, it may be a waste of time to list every single one which is affected.

Anyway, chillout... these things happen, and at least they are quick off the mark. I was working on my website updating some pages and noticed it timed out... checked the emergency status page and they let me know there is a problem they are trying to resolve.

Who are you to deny hard working individuals of a party after 10 years of business. Get a grip....

Nice blog btw DH, I enjoy reading it. Its nice to have a host that isn't faceless, and actually lets people know what is going wrong for when things do happen. Smashing value too, I paid $30 (googled a $96 coupon) for a year of hosting, and am happy to pay the full asking price next year!

By the way im on Arrow, so thats one thats gone offline.

Looks like they stopped announcing one of 4 total prefixes in bgp. Traffic to IPs in other IP ranges should be fine.

(at first, I saw an old history entry):
route-views.oregon-ix.net>sh ip bgp 208.97.128.0
BGP routing table entry for 208.97.128.0/19, version 5574924
Paths: (1 available, no best path)
Not advertised to any peer
12956 1239 3356 10912 26347 (history entry)
213.140.32.146 from 213.140.32.146 (213.140.32.146)
Origin IGP, localpref 100, external
Community: 1239:321 1239:5000 1239:5040 12956:25690 12956:25691
Dampinfo: penalty 372, flapped 10 times in 01:02:51

now nothing:
route-views.oregon-ix.net>sh ip bgp 208.97.128.0
% Network not in table

I'm disappointed about the outage, but these things do happen and I'm glad to know they're on it.

I'm on the Garvey server. I received an email notification that the problem was resolved, however I still can't reach my site, so I guess it's not.
*Edited to add: I got ANOTHER response saying it's resolved when it's not. Anyone else getting this?
Edited by mabel on 04/12/06 06:57 PM (server time).

Seems as though moorpark is down as well.

Desilucy, If you dont like the way dreamhost is run you don't have to be hosted by them ^^. I have been exceptionally pleased with the support. Usually late nights, saturdays, sundays, they are there to help via email. So I know they are working hard right now to figure out the problem.

It looks like Garth is also down.

I know Dreamhost has to be down some time, but now's a bad time for me.

Bad time for me too... :-(

It looks like a DNS issue... My domains are available while my subdomains are not.

Example: my personal blog is available as http://todayinchina.com/thierry but not as http://thierry.todayinchina.com

I hope they will fix this asap! ^_^

http://thierry.todayinchina.com/

This is accessible from here.

I'm back online! ^_^

What about you, guys?

Thanks DH Team!

Something is still wrong with my mail server. I can not send nuke email from that box.

For me the outage started shortly before 0040 UTC and was resolved at 0205 UTC.

I don't know why we're on local time here, but that's 1740 to 1905 PST, so the outage report "Posted: Apr 12th, 2006 - 06:20:58 PM PST" was at least 40 minutes late as seen from here.

I submitted service interruption reports ("panel --> Support --> Contact Support --> Select any websites currently not reachable:") as an experiment. The automatic response reported "No server-wide problem" -- presumably it doesn't know about network trouble.

It still seems to me like good service for the money. If it's not good enough in terms of availability, try pair, and be prepared to pay more.

It's resolved for my sites. I tried to write a clearer title, but ran out of space.

In reply to:

---

I'm back online!

---

Yeah, it appears everything is hunky dory again. The status page reports that they had router problems for about 90 minutes.

I think it may have died again. My site was up for a little while, then went back down again.

My site's DNS is going up and down like a yo-yo. Subdomains are working one minute, then down the next.

It's driving me nuts as I just switched hosts two days ago and I'm still attempting to set everything up. Not the best first impression...

If it's only been two days, give it a bit more time for the DNS to propagate everywhere. It took 5 days for my domain's DNS entries to make it to my mom's ISP. Until it's fully propagated, response will be spotty.


-Scott

ps - If people are seeing problems now, Level3 seems to have been experiencing some major routing issues.