What I Want

Leave a reply

I’m thinking about how to revamp this website, and also rethink my presence of Facebook, Twitter, Flickr.  I’m not quite sure what I want, but some of the problems I’ve noticed:

  • These days, I often share links and things on Facebook, because that’s where most of my friends are, but then they are locked up on Facebook, which makes them hard to go back to even a few weeks later, and its only going to be worse when I move on from Facebook.  Also, any comments/discussion things I post provoke end up trapped in Facebook.
  • If I post things elsewhere, like Twitter, Flickr, or even directly on this blog, people I know often don’t find out about them.
  • I think I want to create a division between my personal and professional presence on the web.
  • I think I want to create some single-subject blogs, but still have them discoverable on Geekfun.

Just as I’m not entirely clear on the problem, I’m not entirely clear on the solution, but again, I have some ideas.

  • Whenever practical, my interaction should be with a single self-hosted software package so that I have control and flexibility.  At this point, I’m assuming that I would be using WordPress, plus some plugins to automatically push updates to Flickr, Facebook & Twitter.
  • When it isn’t practical to publish through a self-hosted tool, updates, photos, etc posted to other sites/services, like Facebook, should also appear my own site as if it was published there directly.
  • Updates, photos, etc distributed to other services should link back to the original, canonical copy of the update.
  • Some easy way of routing updates

Pieces of the puzzle:

  • WordPress on this and other blogs.
  • Twitter
  • Foursquare, Gowilla, or some other location-oriented service.
  • Posts on other sites via BackType and Diqus.
  • Photos uploaded to Flickr
  • Status updates, links, photos, etc that I share on Facebook.

Amazon.com and AWS Down For Me, UltraDNS to Blame?

Leave a reply

We noticed a bit ago that photos we had stored on Amazon S3 weren’t showing up.  A little digging quickly revealed that things were timing out on DNS resolution for s3.

Looks like they are using UltraDNS as their DNS provider, and UltraDNS looks like it is offline too, at least from here.

Attack, or just really bad luck?

EAS:~ Erik$ dig +trace ultradns.net


; <<>> DiG 9.6.0-APPLE-P2 <<>> +trace ultradns.net


;; global options: +cmd


. 15888 IN NS D.ROOT-SERVERS.NET.


. 15888 IN NS L.ROOT-SERVERS.NET.


. 15888 IN NS G.ROOT-SERVERS.NET.


. 15888 IN NS E.ROOT-SERVERS.NET.


. 15888 IN NS I.ROOT-SERVERS.NET.


. 15888 IN NS M.ROOT-SERVERS.NET.


. 15888 IN NS B.ROOT-SERVERS.NET.


. 15888 IN NS H.ROOT-SERVERS.NET.


. 15888 IN NS C.ROOT-SERVERS.NET.


. 15888 IN NS K.ROOT-SERVERS.NET.


. 15888 IN NS F.ROOT-SERVERS.NET.


. 15888 IN NS A.ROOT-SERVERS.NET.


. 15888 IN NS J.ROOT-SERVERS.NET.


;; Received 228 bytes from 10.210.5.254#53(10.210.5.254) in 45 ms


net. 172800 IN NS D.GTLD-SERVERS.net.


net. 172800 IN NS B.GTLD-SERVERS.net.


net. 172800 IN NS M.GTLD-SERVERS.net.


net. 172800 IN NS E.GTLD-SERVERS.net.


net. 172800 IN NS H.GTLD-SERVERS.net.


net. 172800 IN NS J.GTLD-SERVERS.net.


net. 172800 IN NS F.GTLD-SERVERS.net.


net. 172800 IN NS L.GTLD-SERVERS.net.


net. 172800 IN NS I.GTLD-SERVERS.net.


net. 172800 IN NS K.GTLD-SERVERS.net.


net. 172800 IN NS C.GTLD-SERVERS.net.


net. 172800 IN NS G.GTLD-SERVERS.net.


net. 172800 IN NS A.GTLD-SERVERS.net.


;; Received 499 bytes from 192.36.148.17#53(I.ROOT-SERVERS.NET) in 203 ms


ultradns.net. 172800 IN NS pdns1.ultradns.net.


ultradns.net. 172800 IN NS pdns2.ultradns.net.


ultradns.net. 172800 IN NS pdns3.ultradns.org.


ultradns.net. 172800 IN NS pdns4.ultradns.org.


ultradns.net. 172800 IN NS pdns5.ultradns.info.


ultradns.net. 172800 IN NS pdns6.ultradns.co.uk.


;; Received 221 bytes from 192.48.79.30#53(J.GTLD-SERVERS.net) in 146 ms


;; connection timed out; no servers could be reached


EAS:~ Erik$ dig +trace s3.amazonaws.com


; <<>> DiG 9.6.0-APPLE-P2 <<>> +trace s3.amazonaws.com


;; global options: +cmd


. 15605 IN NS C.ROOT-SERVERS.NET.


. 15605 IN NS F.ROOT-SERVERS.NET.


. 15605 IN NS E.ROOT-SERVERS.NET.


. 15605 IN NS I.ROOT-SERVERS.NET.


. 15605 IN NS A.ROOT-SERVERS.NET.


. 15605 IN NS H.ROOT-SERVERS.NET.


. 15605 IN NS K.ROOT-SERVERS.NET.


. 15605 IN NS J.ROOT-SERVERS.NET.


. 15605 IN NS B.ROOT-SERVERS.NET.


. 15605 IN NS D.ROOT-SERVERS.NET.


. 15605 IN NS L.ROOT-SERVERS.NET.


. 15605 IN NS G.ROOT-SERVERS.NET.


. 15605 IN NS M.ROOT-SERVERS.NET.


;; Received 228 bytes from 10.210.5.254#53(10.210.5.254) in 41 ms


com. 172800 IN NS D.GTLD-SERVERS.NET.


com. 172800 IN NS I.GTLD-SERVERS.NET.


com. 172800 IN NS F.GTLD-SERVERS.NET.


com. 172800 IN NS L.GTLD-SERVERS.NET.


com. 172800 IN NS H.GTLD-SERVERS.NET.


com. 172800 IN NS K.GTLD-SERVERS.NET.


com. 172800 IN NS G.GTLD-SERVERS.NET.


com. 172800 IN NS C.GTLD-SERVERS.NET.


com. 172800 IN NS B.GTLD-SERVERS.NET.


com. 172800 IN NS A.GTLD-SERVERS.NET.


com. 172800 IN NS M.GTLD-SERVERS.NET.


com. 172800 IN NS J.GTLD-SERVERS.NET.


com. 172800 IN NS E.GTLD-SERVERS.NET.


;; Received 506 bytes from 192.33.4.12#53(C.ROOT-SERVERS.NET) in 44 ms


amazonaws.com. 172800 IN NS pdns1.ultradns.net.


amazonaws.com. 172800 IN NS pdns2.ultradns.net.


amazonaws.com. 172800 IN NS pdns3.ultradns.org.


amazonaws.com. 172800 IN NS pdns4.ultradns.org.


amazonaws.com. 172800 IN NS pdns5.ultradns.info.


amazonaws.com. 172800 IN NS pdns6.ultradns.co.uk.


;; Received 237 bytes from 192.43.172.30#53(I.GTLD-SERVERS.NET) in 220 ms


;; connection timed out; no servers could be rea

JeeNode is an Inexpensive Wireless Sensor and Controller Node

Leave a reply

I’d like to hook up some sensors to monitor temperature, and energy consumption in my home and I’ve been looking for the cheap options.  One possibility is the JeeNode kits, which are available in the US from Modern Devices for $22.50.

The US JeeNode combines an 8-bit Atmega328 microcontroller, with a ISM-band 915 Mhz, or 433 MHz radio along with 4 analog or digital I/O ports and misc supporting circuitry.  There are a variety of I/O modules available as well.  If I understand correctly, the JeeNode is compatible with the Arduino ecosystem, so there is a lot of code and hardware expertise to be tapped.

Assembled JeeNode

Assembled JeeNode

The JeeNode seems like a great option if you want your sensor nodes to have a bit of intelligence.  It is cheaper than the combination of some other Arduino plus an XBee wireless shield.  For my purposes though, an XBee alone may be good enough, since it has enough “intelligence” to be programmed to send readings from its sensor pins back to home-base on a regular interval, or when an interrupt is generated.

copssh, an sshd installer for windows, is a nice idea, didn’t work for me

1 Reply

C:\Program Files\ICW\bin\sshd.exe: *** fatal error – could not load user32, Win32 error 1114I’m using rsnapshot to do efficient file-based backups of disparate on & off-site servers to a big disk on a backup server here in the office.  Up until this point all the machines involved have been running some form of Linux, but I spent today roping a Windows server we have hosted at The Planet into the mix.  The files I’m concerned with are backup dumps produced by MS-SQL.  In the past I used a Windows-friendly file-sync service to move the files, but I’d be happier if I could do everything from my Linux backup server.

The solution seemed obvious, get an ssh server and an rsync client working on windows so I could treat it like any other machine.  I tried using copssh, which starts with openssh and adds just enough cygwin to get it running on Windows, and wraps it all up with some utility scripts in an easy-to-use installer.  I ran into a few little hitches with passwordless public-key authentication, but after uninstalling and reinstalling, everything seemed to be working well.  I was able to ssh in to the server without entering a password and run commands.  Next step was to install rsync, I went with cwrsync, another cygwin-based port of the unix standard software.

Then the problem began, I tried running rsync from a Linux machine against the windows machine and it failed with the following error when I used a non-administrator account:

C:\Program Files\ICW\bin\sshd.exe: *** fatal error - could not load user32, Win32 error 1114

A little searching suggested I wasn’t the only one.  If I sshed in to a shell and ran ‘whoami’ I saw that I was actually using the service account that had been created for sshd, rather than the account of the user I’d tried logging in as.  More digging didn’t give me much hope.  I found some tweaks to the cygwin environment used when starting the service, but that didn’t help.

So, I ended up giving up on copssh, uninstalled it and cwrsync and just installed cygwin, and used it to install openssh and rsync.  Cygwin packages openssh with some scripts that take care of installing it as a service.  I didn’t use these instructions, but they seem to give a good overview.

So, my advice, just use cygwin it was faster than the “shortcut” I tried.

Surprise, there really is a Google Phone, Maybe.

Leave a reply

Rumors of a Google-branded phone based on their Android OS have kicked up again after a Wall Street Journal article, cited unnamed “people familiar with the matter” confirming the existence of the project.  Latest rumors have the device arriving as early as next year.

What’s most notable is that the rumored Google-phone will launched without a carrier partner, and sold unlocked, which is an unusual approach in the US.   That means no carrier subsidies, which typically knock a couple hundred dollars or so off the price of smartphones.  Many are taking that to mean that the Google phone will be priced like unlocked smartphones are now, which is just ridiculous.

Just as carriers subsidize phones because they plan to make a lot of money locking people into a two-year contract, Google has other revenue streams it can tap if people are using one of its phones.  For a start, its a good bet that they are already paying money out to mobile carriers and device makers in order to make Google the default search and map provider on the iPhone, and other devices.  For a Google-branded phone, that money could instead go to subsidizing the cost of the phone.  That’s just the start, they could also include more-intrusive mobile advertising, though I suspect that isn’t going to be necessary as the mobile ad market grows.

I’m interested to see where this all leads.  Apple managed to pry control away from mobile carriers when it launched the iPhone.  It certainly seems like it will be another step towards loosening the hold mobile carriers, have on users in the US at least.  Of course, this just assumes that there really is going to be a “Google phone,” and that this latest frenzy isn’t just a big hoo-hah over T-Mobile’s next android phone, which is entirely plausible.

Will Google’s New Storage Prices Put Pressure on Amazon S3?

Leave a reply

Google  just cut the cost of additional storage for Picasa & Gmail from $20/year for 10GB to $5/year for 20GB, and now you can buy up to 16TB of storage.

This service is targeted at end-users, and it is limited to Photo storage on Picasa, and mail storage for GMail, but people suspect that it will be expanded at some point to provide more general purpose storage, which seems like a no-brainer once ChromeOS is released.

It doesn’t compete directly with Amazon’s S3 cloud-storage, which is mostly targeted at developers, and can store any file, but I have to wonder if Google’s move isn’t going to force Amazon’s hand on storage pricing.  Amazon may not compete directly with Google in this regard, but a lot of Amazon’s customers are, in one way or another.  Google’s old pricing was somewhat higher than Amazon’s, which left developers with room to use Amazon’s storage to compete with Google.  The new Google pricing totally inverts that.

Amazon’s pricing is due for revision anyway.  Cost of raw storage have declined significantly since S3 debuted, and I’m sure Amazon has been able to lower their operating costs significantly as well, but these savings haven’t really been passed along.  They cut their prices somewhat a year or so ago, and instigated a tiered pricing scheme that rewards large customers for growth with declining marginal costs for storing more data, but the new Google pricing takes things to a “whole nother level.”

It’ll be interesting to see whether S3 and other “cloud” storage providers, and the applications that build on them are about to follow suit.

.