A note on ICECAP

Many have inquired in comments what’s happened to the ICECAP website. My query to Joe D’Aleo was answered this morning, looks like servicide.

I noticed the outage yesterday, AM.  I had emailed and spoke to the hosting customer at [my] hosting support yesterday morning and they said server with ICECAP had hardware failure and they were working in it. I was traveling giving a talk in Chicago yesterday and assumed/hoped it was back up. Found out last evening still out with no estimate for return. I was about to upgrade to a dedicated server with them.
I will talk to manager when he comes in this AM. It was not a DOS [attack] it appears. I am home today so I can stay on their case. Usually the outages last minutes with reboot all that has been needed They assure me they back up everything. God forbid if not. I had 3450 stories stored.
Losing a server is about as stressful as you can get in daily work. Been there, done that. – Anthony
The climate data they don't want you to find — free, to your inbox.
Join readers who get 5–8 new articles daily — no algorithms, no shadow bans.
0 0 votes
Article Rating
35 Comments
Inline Feedbacks
View all comments
E.M.Smith
Editor
May 8, 2009 3:33 pm

FWIW, I did a lot of colo and similar work. Some of the stuff we set up was pretty trick (duplicate live data at Colorado colocation facility and main California site live in real time with redundancy at both sites…) It’s all just a matter of how much money to toss at it.
What I do for my stuff “on the cheap” is each article is on my laptop hard drive and the server. The laptop is backed up via a USB removable hard drive (kept in a metal box mostly – EMP proof) “every so often”. Every so often, the laptop drive is backed up to CD / DVD that are archived elsewhere (fire risk?). So there are 4 copies on 2 media types in 3 physically distinct places… And it costs nearly nothing for the TeraByte slow USB drives and blank DVD’s…
That they have hardware responding to pings is the first step (and it’s IP is configured so it knows who it is 😉 Now they will be working through all those backups… It takes time to put them back together (I know, I’ve done it a few dozens of times…)
FWIW, I’ve done a not running out of STUFF page. Why? Folks on ‘the other side’ often seem to think we are going to hit a resource wall Real Soon Now and I’ve gotten tired of retyping it each time. Besides, this one has pretty pictures and better proof reading 😉
http://chiefio.wordpress.com/2009/05/08/there-is-no-shortage-of-stuff
Maybe I should add a paragraph about the cheapness of hardware and the virtue of RAID-5 with redundant hot swap servers…

E.M.Smith
Editor
May 8, 2009 3:37 pm

Oh, and on the USP thread:
One of the major features they provide is that transients and spikes will kill them and not your computer. Yeah, not an “advertized feature” but one I’m really happy to have anyway… No data lives on my UPS …
“My God Man, that One BIG Fuse!”

Donald Wilson
May 8, 2009 4:36 pm

Icecap is back as of 4:35pm PT, but newest content is April 29, 2009.

E.M.Smith
Editor
May 8, 2009 5:54 pm

Donald Wilson (16:36:48) :
Icecap is back as of 4:35pm PT, but newest content is April 29, 2009.

Sounds like the month ending backup. Give them a bit of time, they may well be loading the daily incrementals with the system live …
Also, FWIW, The Wayback Machine might have some pages. It’s very useful when all else fails…

Editor
May 8, 2009 9:19 pm

Paul Vaughan (13:17:16) :

To Ric Werme:
Someone has impersonated me on your site:
http://www.backtype.com/url/wermenh.com%252fclimate%252findex.html?page=6
Remove my name from the comment.

Interesting. My site is wermenh.com, http://www.backtype.com a is Mountain View, CA business that I was unaware of. They appear to “scrape” web pages here and I assume elsewhere and collect people’s comments in one place.
It’s not my site, I can’t remove your name. Looking through http://wattsupwiththat.com/2009/03/22/natural-drivers-of-weather-and-climate/ I accurately quoted from posts marked as posted by Paul Vaughan. He even thanked me for one of posts with links to past blog entries.

Paul Vaughan
May 8, 2009 11:38 pm

Re: Ric Werme (21:19:34)
Thanks for the reply Ric – and my sincere apologies for addressing you in the imperative.
The Paul Vaughan who thanked you for the links in the earlier WUWT thread – that was me.
Thanks to your note about “scraping”, I now see that they were quoting you quoting me. I’ve never encountered this term “scrape” (but I get what it means intuitively) – do you know if such activity is legal (under all circumstances)?

Paul Vaughan
May 8, 2009 11:46 pm

Actually, they were quoting you quoting me quoting the WUWT article – that explains why I did not recognize what appeared portrayed as “my” words.

Jim Papsdorf
May 9, 2009 4:25 am

ICECAP is back with update of status as of May 8 !!!
Friday, May 08, 2009
SITE OUTAGE
My apologies for the site outage. A very rare major disc failure that affected a lot of clients of the hosting service occurred early on the 7th. The fastest way back was to reload the versions from April 29. They are still trying to bring back old server with info from May 6. Unfortunately the solution eliminated the 25 posts between April 29 and May 6.
Rather than post the backlog of new stories and then have them disapppear when (being optimistic) the site is fully restored, I am only going to add this notice this evening and then reassess tomorrow. My apologies. A lot is going on to tell you about.
Posted on 05/08 at 07:18 PM

Editor
May 9, 2009 6:10 am

Paul Vaughan (23:38:58) :

I’ve never encountered this term “scrape” (but I get what it means intuitively) – do you know if such activity is legal (under all circumstances)?

Copyright law would apply. If Anthony squawked, he might get backtype.com to stop access WUWT. (Please don’t – it could be a useful resource for me.) However, copyright law does allow making excerpts of works, though making excerpts of everything would stretch the law quite a bit.
Given the amount of stuff here and elsewhere that is quoted from other sources, backtype is doing something interesting (think prefetched Google searches) but nothing exceptional.
As for scraping in general, search engines do that. I do that daily to create my Baseball Standings on the Run with the knowledge of the data source. It can cause some consternation in that the scrapers usually skip any ads and some odd .gif files used by some web hit counter systems.
Again, providing a link back to the source is good manners.
Some similar issues exist around “deep linking” where I could put a link to a current weather radar image at Intellicast in one of my pages. The result would be a better page for me, but Intellicast would bear the cost of bandwidth for the image and lose the ad revenue from “normal” access.

Paul Vaughan
May 9, 2009 3:40 pm

Re: Ric Werme (06:10:20)
Thanks for the notes Ric.