IPv6 proponents like to talk about backwards compatibility. It’s easiest if you just forget about that and assume pretty much every piece of software on every client and server (and in infrastructure device in between!) needs fixed to know about IPv6. And needs to also maintain all its IPv4 bits too for “backward compatibility” until everything you want to use is IPv6 capable. So…

A different traceroute. A different ping. A totally different network discovery mechanism. You’ll have something arp-like instead of arp and forget DHCP. You now have stateless autoconfig.

You’ll have IPv6 routing table entries and need some way of setting/updating/propagating them (ie: why UL asked if we were doing BGP or what). In the short term if they’re broadcasting that they route some super set of our IPv6 space, then having them static route /126′s to their customers is probably easiest.

You’ll have different DNS. Eg:

$ dig -6 google.com
dig: can't find IPv6 networking

Oops. No IPv6 stack loaded. Ooops no IPv6 address. Oops no IPv6 routes. Anyway, once you get there, good luck finding many names with IPv6 addresses.

Oh and don’t forget you’ll need an IPv6 firewall. An IPv6 aware NIDS. IPv6 rules for same. All your data visualization and filtering tools will need to know about dotted quads and colon hexes.

And to make it blatantly clear…every application that opens a socket will need updated to be IPv6 aware, ie man:

AF_INET6 [] (7) linux ipv6 protocol implementation

tcp6_socket = socket(PF_INET6, SOCK_STREAM, 0);
raw6_socket = socket(PF_INET6, SOCK_RAW, protocol);
udp6_socket = socket(PF_INET6, SOCK_DGRAM, protocol);

The protocol aims to have a mostly similar API, buy you can’t get around the fact that a program has to be either AF_INET or AF_INET6 or both and know how/when to choose between the two.

But then you also need to update config files and your invocations. Did you put ipv4 addresses in scripts cause you didn’t want to have tohave a DNS lookup latency or assume DNS was working…you get tochange all of them to ipv6. Is your apache listening on:

<VirtualHost 209.237.247.201:80>

Make sure it’s also listening on the ipv6 address.

Now a lot of programs have been updated, like the dig example above. But some have been forked. And every Unix has done something different with all of this. So we get to learn a whole bunch of new, non-standard command line command names and command arguments. Ie: does one “ping -6″ or “ping6″ or something else. Is ping6 even installed if it’s not “ping -6″?

One of the most fun parts will be typing in addresses like 3ffe:1900:4545:3:200:f8ff:fe21:67cf. I can’t wait til mom calls cause the home router’s not working right. “Do you have an IP? What is it…I need to know if your machine just assigned itself a default one or got one from the router so you’ll have to read it to me so I can tell…OK that looks good…can you open up a terminal and ping 2001:db8:1f70::999:de8:7648:6e8? What about pinging 4FDE:0000:0000:0002:0022:F376:FF3B:AB3F? Hello? Mom?”

Suffice it to say I’ve been indoctrinated by those who aren’t drinking the ipv6 kool-aide. But we’re almost guaranteed to be stuck with it and all its complexities…so what can you do? Get ready….we’re all about to be colon hexed.

Earlier this month Frontier did a nice job of botching TV service billing and surprising even themselves that they were raising prices 46%.

I’m still waiting for a piece of information that says something positive about Frontier’s acquisition of Verizon’s Oregon FIOS customers.

It came to my attention recently that I haven’t gotten voicemail on my iPhone for a few months. ATT tried all sorts of things (eg: removed and re-added the feature, reset network settings, accessing voicemail by voice/phone and twiddling with various things, all coupled with reboots, and so on). All that without luck, though we did manage to get it so that I could get voicemail albeit not through visual voicemail and without notification that I have voicemail (remember the days when you had to call in regularly to see if you had voicemail?). Ultimately ATT said I needed to talk to Apple and probably get a new phone.

So I did some googling because google’s a great helpdesk if you can filter through the noise.

It became clear that I likely had an old stale setting in my phone that wasn’t getting correctly reset by the iOS v4 software. In particular this probably came from a profile loaded on the phone, either from when somebody briefly had me load the benm.at tether/mms profile or the profile my work had me load to access their VPN. Both of these have since been removed. Yet the phone was configured to use an APN of “wap.cingular” (for at least some things?) which is wrong. By the way this phone is not jailbroken or unlocked. It is factory. I’ve only loaded configuration profiles which as best as I can tell is something that’s supposed to be allowed (ie: Apple gives out a tool to make them).

The internet seemed to indicate a few options:

• Create my own profile using the iPhone Configuration Utility: The documentation is a big read and while I can see that I can create a profile and also specify various settings this seems like a heavy handed approach to getting back to sane APN defaults. Plus there’s just one place to enter the APN as far as I can see and as you’ll see below I see multiple places where the wrong APN appears stored.
• Edit some files in my backups: if you do the following in a terminal (as one line if your browser wraps it):
  

  grep -ri wap.cingular ~/Library/Application\ Support/MobileSync/Backup/

  then you’ll see there are indeed some text/xml config files with the bogus setting. But are these files signed in any way? Can I simply edit them? Not clear. Plus there’s a binary blob of backup which matches also. Will iTunes barf in a worse way on edited files compared to simply missing ones? Not clear. Plus: what are the official correct APN settings? Maybe ATT could tell me though I’m not so sure they’d go talking to me about how to twiddle these. But there are datapoints that indicate the next option works, so…

• Simply remove a couple of those config files (see comment #7 here): I opted for this route. Specifically I did a sync and back up of my phone. Then I removed the file 8218978e4ab0a48035bb92653145a6be872ea858 as well as the file 58dede14f1db4ccc1328ddacc103d5c3c4f0992b, which were the two files matching the bad APN in the grep above. Next I did a restore. This triggered a reboot, during (re)activation I immediately was notified that there were messages in my voicemailbox (I’d left a few test ones there). Then there was an error where iTunes said it couldn’t do the restore because files were missing. The link above seems to indicate this is OK and you’d think it would be…I’ve removed some files of settings so that they don’t overwrite the system ones. Then iTunes continued on with the restore, syncing all of my content back to the phone from the backup. Viola! After deleting the one file, some people still didn’t have some features working based on what I see on google. I haven’t tried tethering, but as best I can tell everything else is working now after I did the above.

From what the folks at ATT said (nearly an hour in the store with them and on the phone with high level tech support), the “restore network settings” option is supposed to fix the problem but is known to not always do so. This seems like a fairly simple problem to fix if the message were to get to the right place in Apple…removing profiles is not fully removing all their settings. Today’s iOS 4.1 did not include a fix though and the situation is clearly made harder to diagnose and fix since iOS 4 took away the ability to view/edit these APN settings on the phone (eg: you need the computer with which you sync and some tweeking there instead of just clicking around on the phone).

Interestingly the second file I deleted isn’t reference anywhere by google, where the first one is. After a subsequent backup of the phone, the “known” file is back, but now it is a binary file instead of xml (bummer and way to go Apple). The second “unknown” file does not exist in this newer backup.

One interesting thing the config utility gives is a console onto the phone. During the above tweeking it showed the following:

CommCenter[31] : Found the APN settings for 3 in persistent store but not in dynamic store – not resetting them
CommCenter[31] : Not modifying APN settings now
CommCenter[31] : Save APN settings for 3 was successful.
profileInstallationHelper[39] : Not changing carrier profile.

I really would prefer to not have to mess with any of this stuff. But if I have to, it sure makes me want an Android and an open platform with a developer friendly community and tooling. If I’d have gone with ATT’s story I likely would have ended up at Apple getting a “new” (refurb?) phone.

While linux already has about 90% of the supercomputer market, this is still huge news for linux. There have been a few percent of the market who’ve viewed linux as still not quite ready for prime time. But, as the BlueWaters web site says:

Blue Waters is expected to be the most powerful supercomputer in the world for open scientific research when it comes online in 2011. It will be the first system of its kind to sustain one petaflop performance on a range of science and engineering applications.

You can’t get a bigger vote of confidence in your OS than that!

All nice in theory, but it’s been a highly manual choice that a programmer has to make: Which size do I make which pages? Knowing the answer probably requires more computer micro-architectural knowledge than most software developers have. Then there are all sorts of additional complications with practical stuff like getting hugepages, managing how many of which size pages are available for use, and being robust in the face of changes to that availability. For the most part there have been very few programs which use hugepages because of this complexity, which in the past even had annoying features like essentially requiring one to reboot in order to change some of these system configuration settings. And it’s just been too hard for the programmer unless they’ve got a fairly simple use of memory, like a large static-sized shared memory region (ie: DB2/Oracle/MySQL).

Things have progressed massively though. Huge page usability on linux has gotten much easier thanks especially to kernel and userspace utility work Mel Gorman has spearheaded over the past couple of years. Mel’s got a series on LWN discussing these advances. The articles cover a bit of the technical details behind how hugepages give your programs benefit, how to use libhugetlbfs with your applications, and how new utilities in libhugetlbfs allow you to quickly and easily do things like manage multiple pools of different pagesized pages, test whether your application benefits from different possible hugepage backings and actually exploit those benefits even without rewriting your application specifically for huge pages.

While you might want the holy grail of an omniscient operating system that automagically always has your application backed by the “right” page sizes in the right places, what we have now can truly be considered user-ready.

Specifically the first part covers the background context of how hardware and low level operating system software deals with memory. The second part gives a quick introduction to some of the interfaces available for using huge pages with the different types of memory regions an application can have (eg: text, data, BSS, heap, stack, shared memory, anonymous mmap’s).

And…as a teaser:

Subsequent articles in the series will be getting into the key commands for hugepage pool management, running an application with hugepages in different configurations and showing off how easy it is now to test/profile an application workload to see if hugepages would provide benefit. Mel’s written up specific examples of testing a couple well known benchmarks with hugepages and discusses how (and how much) hugepages works for certain types of workloads.

Linux systems from Red Hat and others ate Sun up from the inside out, by colonizing their customer base. Sun vs. the Linux world is a wonderful example of the weakness of proprietary licensing and trying to maintain control over software versus the GNU General Public License (GPL) and decentralized development model that Linux uses.

It’s not just Linux v. Solaris though. The same issues around user/developer community (ie: market) came up with Java, OpenOffice, MySQL, SPARC. Open source people talk about “community” and there are all sorts of ways to think about that vague term. But in the end, not being “community friendly” equates to not being friendly to your customer base and it hurts your market share over time. Jeremy’s blog post gives some nice examples of this.

It’ll be interesting to see what Oracle does going forward.

I’d followed the Intel press releases on their Terascale chip with a lot of interest. Turns out that was definitely meant as a research chip only to test some hardware, with all of maybe five people ever having written software for it and that seemingly as an after thought so the marketing would be able to say something more about the chip.

Just the last two months Intel Research has been making some press with their SCC chip (“Single Chip Cloud” computer…what a marketing name!). This one they’re aiming to actually get out into the hands of researchers. They’ve got a bare metal mode, a full linux kernel per core, and Microsoft’s announced something or other too. It’s particularly interesting though in how it is set up to leverage message passing and does not give cache coherency. It should spawn some interesting academic research in the coming year or two.

Mattson’s definitely of the mind that the way to deal with some of the central issues with scaling is to stop trying to have cache coherence. He makes pretty straight forward arguments. It’s interesting the parallels with distributed computing going back a couple decades even, both in basic programming and in reliability assumptions.

One point that struck me: He said that not many programmers are used to thinking in distributed terms and that most prefer a shared memory model. Probably most of HPC is looking at cache coherent single system images when you’re at the dozens of cores type scale…the scale of these research chips. Mattson comes from a chem background and certainly a number in the audience were HPC types. But maybe their marketing name actually has a bit of foundation in looking at the web space instead of HPC. If you look at today’s really popular web applications, they’re backed by a distributed software model using low end commodity systems that are assumed to be failure prone. So there’s a whole generation of programmers who take it for granted that if you scale up much you need to take the time to architect in a distributed way…they don’t just scale by adding/allowing parallelism and assuming they’ve got a giant machine with a single address space. Probably most of them don’t even program at a level where they know or care what an address space is!

The other thing I took away is that I should probably be paying a little attention to OpenCL.

A large part of my concern about Frontier is whether they’re ready and capable of running a data network at the level FIOS customers have come to expect. While the current FCC bureaucrats thankfully seem to be heading towards enforcing network neutrality, that only covers active disruptions of service by the provider in order to create more billable events to the customer. It doesn’t do anything to insure that a provider is capable of providing reasonable latencies and bandwidths. If the Frontier sale goes through I will still have a 15Mbps dedicated data line, but it’s anybody’s guess what will happen as data goes through Frontier’s network. The Frontier network and its connections to the internet will matter more than the fibre coming into my house.

Given that Verizon had a stake in the Fairport sale going through, I have to attribute some blame to them, not just Fairport, for faking the Fairport network readiness testing done before outside auditors. Which then makes me quite skeptical of Verizon and Frontier assertions in the news media that Frontier is ready and capable and that service will not be degraded as a result of the pending sale.

Unfortunately, not many people are aware of these things or understand how they “get on the internet” and how it can go wrong. The Verizon FIOS service is really quite fantastic currently. There is a lot of room here for future disappointment. Comcast certainly sees the opportunity, is advertising heavily and working hard (seemingly with success) to lure people to what is ultimately a worse product offering.

My weather station sends its data to the internet. Currently the data goes to

• CWOP (and from there it goes on to NOAA)
• The WeatherUnderground

I wanted a way to do things with the data myself though. So I set about making a plugin for WordPress that will pull a weather station’s raw data from CWOP and dynamically update the current weather in a WordPress sidebar widget.

The project lives here, is open source, and hopefully will continue to get some periodic tweeking over the coming weeks to the point where it has the ability to do minimal GUI display of the change in temperature, wind, &tc.

If you’re reading this on my blog (as opposed to in a feed reader or some other aggregator like Facebook Notes), you should see the current weather in the blog sidebar…straight from my roof.

It’s fun being a nerd!

]]> http://dolavim.us/blog/2009/08/19/my-little-ajax-summer-project-a-cwop-plugin-for-wordpress/feed/ 0 http://dolavim.us/blog/2009/06/30/my-weather-station-finally-is-online/?utm_source=rss&utm_medium=rss&utm_campaign=my-weather-station-finally-is-online http://dolavim.us/blog/2009/06/30/my-weather-station-finally-is-online/#comments Tue, 30 Jun 2009 15:22:52 +0000 tpepper http://dolavim.us/blog/?p=525 Continue reading →]]> Last year for my birthday we got a Davis weather station. For Christmas I got the data logger accessory and bought an NSLU2 to run linux and manage the data. I then got busy with school and work and didn’t get back to finishing after getting linux on the slug (which is impressively easy and well integrated). This week I finally got back to playing with it, got the linux install updated and tidied up, got wview built and mostly running. Last night I worked through a few minor bugs and configuration issues and today I see that my data is showing up on CWOP and Wunderground!

I know I’ve got the station sited well with respect to rain and wind, but I’m not sure if it’s going to be great in hot weather (subject to some radiant heat from below). But I’ll be able to compare that well on Thursday and Friday with other stations on wunderground given we’ve got a special weather statement for unseasonably hot weather. Out tomatoes, hops and marionberries are going to love some 90+ degree weather.

]]> http://dolavim.us/blog/2009/06/30/my-weather-station-finally-is-online/feed/ 0