Dave's Good Stuff

Part of the packing, cleaning and moving marathon of the last week, which is to continue through the coming weekend, was the move of the family computers and servers. This meant the following very simple steps:

• the set up of the internet connection at the new place,
• changing DNS upstream to point to the new IP,
• shutting down and physically moving the servers,
• physically reconnecting them,
• connecting the wireless router to the new cable router,
• configuring the new cable router to account for the wireless router,
• configuring the external IP on the wireless router,
• firing up the servers,
• enjoy fun website availability and internal DNS goodness.

Every step went well and easy, all the way up until that last step.

Though all the laptops in the house connected just fine to the Comcast cable modem/router, the two FreeBSD servers simply would not. From a logical network topology standpoint, nothing had changed. None of the IP addresses, including the gateway and the subnet, had changed. As far as the server interfaces were concerned, they’d just been turned off and turned back on. Let me emphasize that – from a logical network topology standpoint, nothing had changed.

I had it set up like so:

Net -> Comcast Router -> Servers & Laptops.

But they weren’t working. One worked briefly, but then quit. The other never would work.

If the cat5 cable (any cat5 cable) was plugged in, a ping to the (same as before) gateway would result in “sendto: Host is down”. If the cable wasn’t plugged in, I’d get a “sendto: No route to host”. Clearly there was some awareness going on, and the NIC was functioning at some level, because pinging the assigned IP, localhost, or 127.0.0.1 would all return successful. I couldn’t get any response from the gateway, however, and no other working machines could get responses from the servers. It was weird.

So, I got on the phone with Comcast to ask them about any incompatibilities with the router and FreeBSD. I got some good information about my usable external public IP (unrelated to the problem at hand), and some completely bogus information about having to use static IPs within the DHCP scope on their router (yeah but… what?!). The first level tech support wanted to help, but he just didn’t have the expertise, and so he escalated me to the next level (who is well past their 48 hour self-imposed deadline at the time of this writing). I decided to change things up a bit.

I went to this setup:

Net-> Comcast Router -> Wireless Router -> Servers & Laptops.

Note that this is exactly the same as I had at my old house, with the substitution of the Comcast Router for the Surewest cable modem. Everything from the wireless router on back is identical.

You know what? The laptops all worked (I love knowing even rudimentary networking), but the servers still didn’t work. Having eliminated the router, cables and network configs, the fact that it doesn’t work anymore with the exact same setup as was at the other house tells me it’s a another type of hardware problem.

So I yanked the gigabit NICs right out of the servers and went back to the on-board 100baseTX ports and… get this – it worked just fine.

What I’m concluding is that the D-Link GigaExpress DGE-530T card doesn’t work well with the BIOSTAR N68S3+ and the Diablotek EL Series PSEL400 400W ATX PSU. I base that conclusion in part b/c, in addition to flat out not working anymore, there are times when the machines won’t power on when the DGE-530T is installed without some creative combinations of the case power button and the PSU switch. When those cards aren’t installed, there are no issues. What, I didn’t mention that before? My bad.

Given that when I put these servers together, I did so with as little cash outlay as possible. I’m thinking I’ve been bit by the “get what you pay for” principle. In time, I’ll beef them up a bit with better components. But for now, I’m just happy to be back online enjoying fun website availability and internal DNS goodness.

So, I started playing Baldur’s Gate recently. I never finished it back in the day (though I got close), and barely took advantage of the Tales of the Sword Coast expansion. For some reason, with everything going on in our lives right now, and in spite of the fact that I’m arguably busier now than I’ve been since Finals Week in college, I just couldn’t help running through 5 CDs worth of installation and launching some relatively old school gaming (I consider Atari 2600 and the Age of Darkness and the Age of Enlightenment truly Old School – and even that’s too new for some).

It runs a little fast on my m9700, but not too fast to play. The dialog runs together now and again when there is a lengthy challenge/response conversation playing out, but other than that, I’ve noticed no ill effects of the slightly accelerated rate.

I tried using the G3: Widescreen Mod in order to enjoy larger resolutions, but found that I had issues with mouse scrolling around the area at 16:9 resolutions that were less than the native resolution of my monitor (1920 x 1080). Because of the time consuming way the mod is applied, I didn’t try too many resolution options. I also noticed some frame rate issues at the native resolution in spite of the age of the engine (or perhaps, because of it). So I decided to stick with playing it the way it was originally released and patched. I’m not playing for the graphics, after all, but the experience.

I’ve heard rumors that it’s possible to convert BGI to the BGII engine, though. I might look into that. Graphics aren’t everything, but the BGII engine is so much nicer…

I probably won’t have much time in the next couple of months to do more than tinker with it here and there, but after we’re settled in to the new house, and some of my other responsibilities are managed (not the least of which is a FreeBSD build that’s proving difficult due to a troublesome inability to detect the hard drives once in sysinstall), I’ll be able to devote a little more time to it. I may even dive back into Baldur’s Gate II, and the Icewind Dale series.

I wonder how it’ll run on the X79 LGA2011 based machine I plan on building towards the end of the year. If it runs fast on a 6 year old laptop…

With all the major breaches in the news lately (RSA, Gawker, PSN, Lastpass, MySQL, Texas Comptroller, etc. – hit the googles with you!), I finally got angry enough to ask a question of DiscoverCard that has been sitting on the back burner for too long. It’s about what I consider to be absurd password length restrictions on their site, and what that might say about how they’re storing those passwords on the back end.

Let’s clear one thing up: On principle, I believe I should be able to use whatever password I want. If I want to use “abc123″, or “puppy1″, or “;a3Wfzu0J|rqVHj%l]x6PZdQHqhpK39vx5?|fSb9NmFdq”, I should be able to. I should have the right to be as smart, stupid, paranoid or legitimately cautious as I want (here’s the thing) so long as my choice doesn’t affect others using the same system. I’m not the only one to think along those lines.

Principles aside, I don’t have a practical problem with complexity enforcement. I understand that repositories of critical and/or sensitive data and services have a need to shore things up a bit with more complex passwords. That doesn’t stop me from thinking there are design issues at hand if my weak web password can compromise someone else’s data. I also can’t help but think the whole mess is, at least in part, driven by a disgusting, ludicrously naive and juvenile expectation that “everything will always be ok, and I’m entitled to it. It’s my right! So there!” and all the litigation that goes along with that childish attitude.

That all said, it’s one thing to require password complexity, it’s another entirely to not allow it past X characters. It raises some questions…

So I asked DiscoverCard about it: (May 05, 02011 09:41 AM)

Can you please explain to me the exact method by which discovercard.com website logon passwords are stored? The length limit greatly concerns me. With the major breaches of late (PSN, Sony, Lastpass, MySQL, Gawker, etc.) it is critical that passwords be stored securely, with adequate encryption, or at the very least, salted hashing. Thank you.

Admittedly not the best in terms of wording and stated concerns. But there you have it. I can’t help but believe DiscoverCard, of all places, is using top-notch methods for storing passwords, but you just never know, do you?

DiscoverCard’s first reply: (May 05, 02011 10:46 AM)

Thank you for your recent message. I understand your concern about online security, and I will be happy to assist you today!

Our emphasis on privacy provisions and improved Internet security has made using our website safer than ever. We use the Secure Socket Layer (SSL) protocol for encrypting communications with our Cardmembers. SSL uses public-key cryptography to scramble the information sent between senders and receivers. In the unlikely event that third parties try to eavesdrop or intercept this message, SSL encryption prevents them from viewing its contents.

We also have Superior Fraud Protection, which means that when you use your Discover Card to shop anywhere on the Internet, you will not be liable for any unauthorized transactions. For more information about our security, please see the following page:

http://www.discovercard.com/customer-service/safety/site-security.html

I appreciate your business and the opportunity to be of service. Thank you for choosing to use Discover Card.
[signature and legalese removed]

Awesome, but no, that’s not what I asked. So, I completed a survey indicating as much and replied… (May 05, 2011 11:06 AM)

Thank you for the prompt reply. I appreciate the information on SSL, however, that doesn’t answer the question I asked. I asked about password storage (at-rest), not information on the wire (in-transit). My question is restated here for your convenience…

“Can you please explain to me the exact method by which discovercard.com website logon passwords are stored? The length limit greatly concerns me. With the major breaches of late (PSN, Sony, Lastpass, MySQL, Gawker, etc.), it is critical that passwords be stored securely, with adequate encryption, or at the very least, salted hashing. Thank you.”

To which they replied… (05/05/2011 12:06 PM)

Thank you for your recent inquiry. In an effort to provide you the best service possible, I have escalated this issue to the appropriate parties in our company for further assistance. We will respond as soon as we have any additional information concerning this matter. Thank you for your patience.

I appreciate your business and the opportunity to be of service. Thank you for choosing to use Discover Card.
[signature and legalese removed]

I don’t expect another reply anytime soon… but I’ll let you know if I do get one, either in the comments, or as a new post depending on the reply.

… well, maybe not quite yet. Vertical lines have started showing up the screen, though. First there was just the one, way over to the right of the screen and out of the way. This morning, a 2nd showed up right down the middle. Ugh.

My research points to a couple of things that could be the problem. Some say it’s a connector issue. Others claim LCD. I’ll do more research, but I’m inclined to think it’s the screen. Regardless, I reckon there will come a time in the nearish future where I’ll be forced to make a choice. The way I see it, I have a few:

1. I can simply do without.
2. I can buy a replacement LCD for ~$180 and install it myself or, with a little more money, pay someone else to do it.
3. I can buy a new gaming class laptop.
4. I can buy a non-gaming class laptop.
5. I can take up the task of building a new desktop computer.

Anyone have any guesses which way I’m going to go?

I won’t simply do without. Nope. No can do. Next…

I’ll consider replacing the LCD, but the laptop is 5+ years old now, and is definitely showing its age. It’s had a good run, but I have to face the notion that it might be time to put it to pasture. Or at least relegate it to more mundane tasks that don’t require a full time monitor – or better yet, donate it to someone who can’t afford their own new laptop… it’ll still more than serve basic needs, and I have just the person in mind.

I’m not interested in a new gaming-class laptop. The ROI just isn’t high enough for me. If I’m going to spend that much cash, I want more machine that what’s offered in a laptop. I want more performance, more customizability, more upgradeability, and frankly, more control.

Nor am I interested in a non-gaming class laptop. The ROI might be better, but the performance just won’t bet there. I. Need. The. Fast.

That pretty much leaves me with building my own desktop. I’ve been batting around the idea of building a new computer for a few months now. It’s been 5+ years since I’ve upgraded that aspect of what my wife likes to call my “Command Center”, so it’s about time (and as much as things have stayed the same, my oh my, but have they changed!). I’ve put together a rough budget with a range of costs per component, and I could spend as little as a few hundred, or as much as a few thousand. We have a lot of expenses hitting all at once right now, but after a few months and things have settled down a bit (around my birthday), I think I might be able to make a go at it, even with the addition of a home NAS solution (which I’m still debating, to be honest).

I just hope that my laptop screen holds out until then. I’ve seen what happens when this issue is left alone, and it’s not pretty…

UPDATE 2011.03.09 – I think I got a handle on it. I’m still pursuing the subject of this post, but I’m no longer worried about the backups.

I’m done.

I’ve given up on USB based backup solution. Sunday morning has become my standard “find out what went wrong with the full backups last night and see what I can do to fix them.” I tire of it. Granted, the failures this weekend were because I ran out of room on my little 80GB USB drives. Totally my fault. It was just a matter of time. I wasn’t paying attention and the backups failed. Fortunately, that’s all that happened, as opposed to something more insidious. At least it wasn’t some sort of kernel panic, or soft-updates issue again.

I could easily solve it by spending a few bucks on a larger drive, but that would just be another stopgap. I want a solution that will carry us a few years and then some.

So, I’m thinking NAS*. Something that would serve my family’s needs (which amounts to my wife and I at this point, but we’re really hoping for a little papoose sometime here real soon). That means a lot of storage space. That means seamless connectivity with our existing machines, and that means dead simple to use.

I could spend a few hundred dollars on hardware and many hours putting together my own FreeNAS server from pieces parts (or any one of a number of other free options). Or I could spend a few hundred dollars and a few minutes on an OOTB solution.

I’m leaning towards the OOTB solution.

Sure, it’s not as proudly geeky as a home grown solution, but my gorgeous wife doesn’t appreciate geekery as much as some of you and I do. She appreciates things that work and work now. If I’m going to spend this amount of time and money, she has approval powers – it’s just part of that thing called Happily Married. Frankly, the older I get, the more I agree with her. So, OOTB NAS it is.

So far, though I’m still keeping my eyes and mind open, I’ve narrowed my choices down to:

Synology Disk Station DS410
Synology Disk Station DS410j
Netgear ReadyNAS NV+
Netgear ReadyNAS Ultra
Netgear ReadyNAS Ultra Plus
QNAP Systems 419P+
Thecus N4200Eco/Pro
Seagate BlackArmor NAS 400

Each of them fits my base requirements:

1. Interoperability between Windows, Mac and *nix machines
2. Function as a print server
3. Four drive bays for RAID5 or better (hot swappable a huge plus)
4. Small physical footprint

Each of them will do the job. So now, it’s a question of features, performance, future-resistance and of course, price.

I’ll be researching each of these models (and any others that come across my screen in my research) over the next couple of weeks (or less).

* Yeah, I know NAS != backup. This is just a step in the right direction. For backups of the NAS, I’ll grab a big 1 or 2TB disk, throw it into my dev server, and rsync the data from the NAS to it. I’ll keep my backups scripts running for my server data, but I’ll point them over NFS to the NAS, rather than to flakey USB drives.

Since I rebuilt my systems with FreeBSD 8.1, I’ve been hounded by an error message during weekly level 0 dumps. This only happens on my /home partition, which is significantly larger than all the others combined, and only happens on the full weekly backups. The daily level 1 backups all work flawlessly. Given what I’ve learned, I’m thinking it’s just b/c the level 1 backups are done too quickly…

The Problem

The error message, “(da0:umass-sim0:0:0:0): AutoSense failed” is followed by a slew of write messages

(da0:umass-sim0:0:0:0): AutoSense failed
g_vfs_done():da0s1[WRITE(offset=19495206912, length=131072)]error = 5
g_vfs_done():da0s1[WRITE(offset=19495337984, length=131072)]error = 5
g_vfs_done():da0s1[WRITE(offset=19495469056, length=131072)]error = 5
g_vfs_done():da0s1[WRITE(offset=19495600128, length=131072)]error = 5
g_vfs_done():da0s1[WRITE(offset=19495731200, length=131072)]error = 5
g_vfs_done():da0s1[WRITE(offset=19495862272, length=131072)]error = 5
g_vfs_done():da0s1[WRITE(offset=19495993344, length=131072)]error = 5
g_vfs_done():da0s1[WRITE(offset=19496124416, length=131072)]error = 5
g_vfs_done():da0s1[WRITE(offset=19495075840, length=131072)]error = 5
(da0:umass-sim0:0:0:0): lost device
(da0:umass-sim0:0:0:0): Synchronize cache failed, status == 0xa, scsi status == 0x0
(da0:umass-sim0:0:0:0): removing device entry
/backup: got error 6 while accessing filesystem
panic: softdep_deallocate_dependencies: unrecovered I/O error
cpuid = 1
Uptime 2h25m37s
Cannot dump, Device not defined or available
Automatic reboot in 15 seconds - press a key on the console to abort

… a kernel panic, and a dead system. The keyboard doesn’t respond, so it just sits there until the machine is hard reset manually.

Bad drive? Possibly. But that would make two in a row, so I’m leaning more towards something system-related, rather than drive related.

That led me to this post on the FreeBSD forums, and that then led me to this post elsewhere on the googlewebs. The latter indicates a difference in the way soft-updates are handled in 8.x vs. 7.x.

A Solution

So… I turned off soft-updates with:

Trying the same command with the drive mounted threw the error:

tunefs: /dev/da0s1: failed to write superblock

I knew it wouldn’t work, I just wanted to see what exactly would happen.

My only question, which the posts I found did not answer, was whether to turn soft-updates off on the source /home partition, or the target USB backup drive. I opted for the target given that it’s only used for backups rather than day-to-day I/O operations, and it’s the quicker and easier than rebooting into single-user mode to disable soft-updates on my /home partition. So I tuned the drive, crossed my fingers and launched the backup process again.

The result:

...
DUMP: 30.73% done, finished in 2:26 at Sun Feb 6 13:45:28 2011
DUMP: 33.06% done, finished in 2:21 at Sun Feb 6 13:45:39 2011
(da0:umass-sim0:0:0:0): AutoSense failed
g_vfs_done():da0s1[WRITE(offset=87491444736, length=131072)]error = 5
g_vfs_done():da0s1[WRITE(offset=87491575808, length=131072)]error = 5
g_vfs_done():da0s1[WRITE(offset=87491706880, length=131072)]error = 5
g_vfs_done():da0s1[WRITE(offset=87491837952, length=131072)]error = 5
g_vfs_done():da0s1[WRITE(offset=87491969024, length=131072)]error = 5
g_vfs_done():da0s1[WRITE(offset=87492100096, length=131072)]error = 5
g_vfs_done():da0s1[WRITE(offset=87492231168, length=131072)]error = 5
g_vfs_done():da0s1[WRITE(offset=87492362240, length=131072)]error = 5
g_vfs_done():da0s1[WRITE(offset=114688, length=16384)]error = 5
g_vfs_done():da0s1[WRITE(offset=87096983552, length=16384)]error = 5
g_vfs_done():da0s1[WRITE(offset=87289675776, length=16384)]error = 5
g_vfs_done():da0s1[WRITE(offset=87482368000, length=16384)]error = 5
g_vfs_done():da0s1[WRITE(offset=87491313664, length=131072)]error = 5
DUMP: 35.36% done, finished in 2:17 at Sun Feb 6 13:46:03 2011
DUMP: 37.73% done, finished in 2:12 at Sun Feb 6 13:46:00 2011
...
DUMP: 96.69% done, finished in 0:07 at Sun Feb 6 13:45:57 2011
DUMP: 99.17% done, finished in 0:01 at Sun Feb 6 13:45:42 2011
DUMP: DUMP: 39140287 tape blocks
DUMP: finished in 12691 seconds, throughput 3084KBytes/sec
DUMP: level 0 dump on Sun Feb 6 10:12:12 2011
DUMP: DUMP IS DONE

No kernel panic. No hard reset required. It just picks up where it left off and goes along it’s merry way. I’m fairly confident that file integrity is being maintained, but I’ll be testing that to be sure.

Other than soft-updates prevented it from recovering from the loss of the USB drive, I’m not sure exactly what the problem is. Why was the USB drive lost to begin with? Is it a timeout issue? An I/O issue related to too much data in the pipe? A RAM issue? I’ve two more GB’s to install, but I’ve been waiting to get more duration data to compare against before installing it.

For now, though I’m going to watch it closely, I’m considering the issue tentatively solved.

As I suspected, my 100MB/month plan reset this morning, and I’m back up to my full allotment.

Which isn’t that much, as I also suspected. See, I’ve been tracking it, and my own data usage on my desklaptop is pretty severe, and I really haven’t used it that much. Admittedly, I’ve used it more in the last couple of weeks as I built out my two servers, and wanted the ease of side-by-side ssh sessions. Lots of research and referencing online tutorials, how-to’s, man(n) pages, and the like. Most of it was primarily text based (not a lot of imagery in man pages – don’t you think they could be prettified a bit?), but the bandwidth usage still adds up. In fact, for me, it adds up to about 100MB per day!

I could have used my Chromebook, but it really wasn’t ideal for the task. My wife was, many times, confounded by the sheer amount of LogoRhythms I was ensconced within. Throughout the bulk of the builds, I never had fewer than two sessions going (one for each server), and several times I had upwards of 5 and 6 as I tailed logs there and yonder, and ran compiles and code and configs here*. Though I got around the lack of key based ssh authentication in crosh $ with Match statements in my sshd_config, the lack of window tiling, while not fatal to the task at hand, would have hindered me a touch. It was immensely helpful to have multiple and simultaneously visible windows open at once. I’m also a stickler for the look of my terminal sessions, and crosh doesn’t allow much by way of customization.

A sysadmin platform the Chromebook is not. Then again, it was never designed to be.

Only slightly apropos of the topic, we had a lot of home showings over MLK weekend, at both houses. The Chromebook was handy to have around when we didn’t really have any place to go in particular, or errands to run. It provided for plenty of lightweight and easily portable distraction at bookstores and coffee shops with free Wi-Fi.

You better believe I browsed everything http://. That https:// stuff is for wussies. Live a little, people. I’m sure the shady characters off in the corners with their hoodies on and their screens tilted down and facing the wall appreciated that.

* I could have reduced my workload significantly by building one server and mirroring the drive with dd but… well… I didn’t. Maybe next time.

You may have noticed that my site has been offline for a few days. Or you may have noticed and thought I was simply being quiet. I’m known to some to be that way now and again. Not the case this time, though. I was offline good and proper. Now, after lots and lots of LogoRhythms, I’m back on line…

What happened was this: the utter corruption of my RAID1 setup, resulting in the loss of nearly all the data in the mirrored HD partition in which my web server data was stored. My /home directory was mounted on the same drive, but it wasn’t affected for some reason. My /www directory, home to my web server and all the data it served, was nearly completely lost, with plenty of data unrecoverable. At least I have my initial configurations, and somehow, the daily MySQL backups of all my databases.

No, I didn’t have the data backed up elsewhere. I relied on RAID, figuring that HD failure was my biggest concern, not driver/controller/software failure and the loss of superblocks and disk labels.

Yes, I’m an idiot. I should have known. Lesson learned.

So, over the weekend, with the blessings of my lovely and incredibly understanding wife, I exercised my hardware research muscles, and put together the specs for two identical systems to serve as production (Britannia) and development (Sosaria). Then I went out and bought it all (see lovely and incredibly understanding statement above). Here’s the thing… I spent about a third on both of those systems as I spent on my gaming rig 6 years ago, and they’re both arguably nearly three times as fast (by certain measures and points of view… they screamed through the initial SSH host key generation routines so fast I didn’t even realize it was done until I was sitting at a logon prompt). The hardware is anything but crazy fancy, but it is working.

As an aside, it was interesting to watch the two machines run package installs and updates in tandem through two simultaneous PuTTY instances. Nanoscopic differences in the hardware always had one machine or the other running ahead.

It’s nice to know I still got what it takes to piece together a working computer from separate components, and have it boot the first time I give it the juice. Granted, they were all new parts, so there wasn’t any real hacking involved, but it still feels nice.

I kept costs down by sticking with on-board graphics (who needs a distinct graphics card to push console text?), and already having monitors sitting around. I’m saving space by using a KVM switch, instead of multiple monitors & keyboards. It’s a mechanical switch, so until I get a battery backup in place (not cheap), I may run into issues if a brownout causes unplanned reboots. I’m keeping it set to my prod box in that case, to at least simulate some consistent uptime. I haven’t had a brownout in a few years, either, so I’m due…

The two servers are identical in every way, from the hardware up through the OS and (at least at this time) the software packages installed. That way, I can be sure what I develop on dev will work without issue on prod. In fact, once it’s all set up the way I want, I won’t even need to log in to prod to publish new work. I’ll just check in the changes via Subversion and sip my beverage of choice as it all “just works.”

To help avoid future data loss issues (rare is the perfect solution), my dev box will serve as backup to my prod box by default, and I’ll have some large external drives hanging off my dev slow box for yet more backups (via relatively slow USB 2.0 connections – but backups just have to work, not be fast – I’m still looking into a NAS solution, b/c that would just be cool). I’m thinking about looking into off-site solutions too (cloud?), but haven’t started that yet. I’ve no doubt some of you will have opinions on this set up, which I’m more than willing to entertain, assuming they don’t involve too much more money. :)

I’m excited to finally have a semi-proper development environment again. And develop I will… not only to recreate what I lost over the last 6 months (which wasn’t that much to be honest), but also to create whatever it is I’ll create from here out.

But first, I’ll fix all the broken images, and then I’ll get to creating a new theme for this site…

I’ve never been subject to bandwidth caps or quotas, so I’ve never really paid it much heed. I have been aware of their existance, and have kept them in (the back of my) mind with my own designs in the past by keeping imagery low and abiding by the “Content is King” mantra, and keeping my content in the realm of text.

Now, with the Cr-48 Pilot Program, I’ve been inducted into the frightening world of Bandwidth Caps and Quotas, and it’s been a bit of an eye-opener. My own estimates of how much data I throw about in a typical browsing day fell dramatically short of the mark. To their credit, they do say the free plan is “…enough for hundreds of emails or occasional browsing” (emphasis mine), I just wasn’t listening. That’s assuming your e’mails don’t include attachments or crazy signatures with images (any signature with an image is crazy, but that’s a subject for another post in which I lament the demise of plain text as the primary e’mail delivery format).

Their Plans

To help us preview their new OS, Google, in cahoots (ok, in cooperation) with Verizon, initially offers the following plans:

• 2 year 100MB of data per month for $0.00
• 1 day (24 hour) limitless pass for $9.99
• 30 day (720 hour) 1GB pass for $20.00
• 30 day (720 hour) 3GB pass for $35.00
• 30 day (720 hour) 5GB pass for $50.00

All plans are effective from the moment they’re activated. Though most of the plans are for 30 days instead of calendar month, I believe the 2 year plan is by calendar year. So, starting the 1GB plan on 12/01/10 means it’s good through 12/30/2010 (30 days from date of plan activation). Starting the free plan on 12/01/2010 (as I did), however, means you get 100MB free per 30 day period until 12/01/2012. It would end on 11/20/2012 if it were a 720 day (24 months * 30 days per month) plan. I’m sure this is all documented somewhere, but half the fun is guessing!

I’ll take note when my data cap is reset to 100MB to verify my assumptions, but it should happen on 1/19/2011.

My Plan

I plan to spend most of my time connecting via Wi-Fi, rather than 3G, so I opted for the free 100MB/month. However, burning through nearly 90MB over the space of 3 days while at my parent’s house with *exceedingly* careful usage has got me thinking about how much a typical browsing session uses.

By “*exceedingly* careful” usage, I avoided, as much as possible, all streaming media, all file downloads, any image heavy sites, online gaming, and kept my forays onto the Web as short and sweet as possible. Mostly, to be honest, I was just writing my Initial Impressions post. And that was mostly offline, with occasional textual updates online to make sure I didn’t lose anything (it is, after all, such a fine post… meh). Unfortunately, advertisers aren’t worried about the bandwidth capped or those charged by the MB. There are far too many bandwidth heavy flash and streaming ads out there for my taste, and in some of the most unlikely places.

So… Here are some numbers I’ve put together. It’s not scientific, terribly regimented or documented, but I wanted to get a general sense of how much data is transferred in a typical session for me. I’m using NetWorx on my Win7 laptop, b/c there isn’t a bandwidth usage monitor available yet for Chrome OS that I found (that would be incredibly useful, if someone wanted to put that together). I don’t feel like hacking my router just yet to install DD-WRT, Tomato or whatever other alternate solutions are available, nor do I feel the need to implement Squid, so I’m going to deal with local monitoring for the time being. Most of my time is on Google Reader, Gmail, Google Docs, and my own site. My Google Reader usage is primarily text, but there are some feeds that are fairly image intensive, including a few online comics thrown in for good measure.

My Results

Tonight, over the course of about three hours of intermittent browsing broken up by an episode of Firefly, some financial talk with my wife, and dinner, I burned through 18.8MB of data. No single site I visited carried a heavy media payload, and I didn’t honestly browse around that much. I caught up on a few feeds, read some articles, and fixed some books on Goodreads. Nothing was that intense. Truth be told, I saved this video for my Chromebook so it wouldn’t be counted amongst the bits tallied.

I’ll keep monitoring my usage to see, but it’s certainly looking like 100MB/month truly is a pittance that would fulfill only the most spartan of Internet users.

As an aside, just collecting links by visiting the site to copy/paste the URL, and polishing this post burned another 14MB. Ain’t that something?

When I filled out the Pilot Program survey the day it started on December 7th, the last thing I expected was to be chosen. Yet here I am with a Cr-48 on my lap.

Now that it’s here, I feel something of an obligation to do at least an initial impressions post on it. After all, everyone else has (march ye to the Googles to find more), and it doesn’t seem right to accept this free laptop without at least paying for it with some typing and whatnot.
The Arrival
[SinglePic not found]I opened the Fed-Ex box, and found within the now-famous illustrated box, with the explodey drawings on it. I absolutely love the drawing style. It has a distinct Captain Mark sensibility, which I worshiped as a young lad. I especially like the little mouse.

I couldn’t remember at first where I saw the drawings, but I felt my excitement build for reasons I couldn’t quite explain. Something cool lay waiting within… I just knew it. I just couldn’t quite place what it was. Then it hit me, and I immediately went to get my camera.

On the hardware

I’m not going to spent a lot of time because it’s a basic reference that won’t be sold retail, has little in the way of fancy, and because it’s been done a many times before (forget ye not the Googles, my friend). That said, on the off chance that my vote will count for something, I should at least offer a little something by way of critique.

Like
The first thing I noticed, and one of the things that I continue to love is the rubberized chassis. It’s a little tougher to keep pristine, as the surface tends to hold on to blemishes (oils, spilled coffee, dog slobber, etc.), but for me it’s worth it.

I also love the chiclet keyboard. I like the weight. I like the design of the thing. It’s functional. It’s also quiet. I’m also becoming fond of the layout, including the replacement function keys above the top row, and the embedded power button. I can’t say I’ve used the search button that much, except by way of testing it’s functionality, and I miss my dedicated “Home” and “End” keys (though the shortcuts CTRL-Alt-Up and CTRL-Alt-Down respectively, are easy enough to get to after they’re committed to muscle memory).

I love the matte surface display. I’m used to the glossy display of my Alienware Aurora m9700, and I have to say I don’t miss the reflections. At 1280×800, it’s as wide as I need, and provides more vertical space than a lot of smaller laptops today, which stop at 766. That’s just not enough room for me. It’s no 1920×1200, but for a small netbook style laptop, it’s plenty.

The battery life is advertised at about 6 hours with maximum brightness, and 8 hours at minimum. I’ve not pushed it to the limits, but I’ve spent most of my Chromebook time on battery vs. AC power, and have been satisfied. There’s a nice little display in the top right of the UI that gives you battery charge percentage and time remaining. I have no reason to suspect it’s inaccurate.

I would be remiss if I didn’t mention the quality of speakers. I messed around with Grooveshark a couple nights ago and, while they’d never suffice for long term use, the speakers aren’t horrible for laptop speakers. I haven’t tested my headphones with the Cr-48 yet, but I reckon they’ll work fairly well.

At first, as a whole, the Cr-48 felt (though I hate to use the word) cheap, and more like a toy than a computer. The more I use it, though, the more solid it feels. I understand it’s a little heavier than most netbooks, but as I have zero experience with other netbooks, I can offer no thoughts on the matter.

Not so Much

I’ve never been a fan of touch pads, and as such, don’t have much experience with them. Though I do like the two finger scrolling feature, I find the use of the touch pad, as a whole, less than ideal. I’ve had a hard time getting the two-finger-tap to bring up the context menu, and my cursor tends to jump around a lot. I’m not sure if that’s me just fat-palming it, or some sort of software glitch. However, they win points with the simple and easy differentiation between the touch pad and the surrounding chassis. There’s enough of a difference so that you know where the boundaries are, without it being a distraction otherwise.

As a whole, it’s a little underpowered for the more intensive flash driven web sites. For instance, Bejeweled on Facebook (yes, I’m a fan) is barely playable due to the lag. Other than that, for basic web use, the hardware is more than sufficient.

The final point regarding hardware, and this is very much specific to the Cr-48, is the size of the bezel around the display. I would love to see a laptop with a screen that fills up the lid nearly completely.

On the OS

Chrome OS basically amounts to an RDP session with the Chrome browser set with `alternate shell:s` There’s more to it than that, obviously, but that’s essentially what it is. Anything you can do from within the Chrome browser, you can do in Chrome OS. Most anything you can’t do outside the Chrome browser, you can’t do in Chrome OS. It’s been said that if you’ve worked with Chrome, then you’ve worked with Chrome OS. That’s mostly true.

Like
I like the general simplicity. Being little more than a browser, there’s not much to do except log in (Google accounts only for now) and start working. You can add different apps (or rather, bookmarks) via the a Web Store, and you can adjust a few things under the hood much the same way you do with the Chrome browser through the wrench icon to the right of the Omnibox, but for the most part you log in and get to work (or play, if that’s your preference). The downside is that there’s not much for the power user like myself to configure.

Along with the general simplicity is the fact that it takes about 14 seconds to go from a completely cold state to being out and about, doing whatever it is you want to be doing. The SSD, combined with a distinct lack of extraneous features makes for an extremely fast boot time.

I like the multiple window support (Shortcut: CTRL-N). It desktop parlance, it’s basically launching a new browser instance, which is easily accessible via the “Next Window” button above the “6″ key (SHIFT-Next Window chooses the previous window).

From a security standpoint, they’ve covered their bases fairly well as well. Like the browser, updates are done behind the scenes automatically without any user input. The trade off for never having to worry about out-of-date code is the chance that a faulty update will be released that crashes the OS, or corrupts data in some way. I’ve not had that happen, and haven’t heard of it, either. But it’s certainly possible.

Still talking security, the sandboxing feature keeps different processes isolated (on a per tab basis) from each other and the base linux-based OS, and encryption keeps the data on the SSD safe from potentially prying eyes.

The OS designers gave a nod to the CLI junkies among us with crosh>. The Chrome Shell launches in a new window that’s accessible via the “Next Window” button. It’s pretty basic, and sticks to the overall network-only philosophy by providing very little besides those basic diagnostics and an SSH client. It’s a start, and if Google isn’t lying about the OS constantly improving, we can expect good things there as well.

I’ve mentioned shortcuts. Those and quite a few more are available via the CTRL-ALT-/ shortcut. It brings up an interactive screen that displays all the shortcuts available. There are quite a few, which presents a learning curve, but not an unmanageable one.

Not so Much
The sensitivity of the touch pad sometimes causes the cursor to jump around on the screen, this could be my own lack of experience with such things, but given how often the unexpected jumps happen, I’d like to be able to adjust the sensitivity more. The plus side is that it forces me into a proper typing posture with my hands elevated well above the chassis and touch pad.

The lack of granularity and feedback in the Wi-Fi settings was troublesome at first. I had my wireless security set to WPA2, and found that the Chromebook simply would not connect. I had to drop back to WPA before it would recognize it. However, I spent too long in trial and error before I figured this out.

Tap-to-Click on the touch pad is disabled by default, and as a exceedingly rare user of touch pads, I was held up by the force it takes to actually click the touch pad without tap-to-click being enabled. I’m not sure why tap-to-click was disabled, but I’m sure there was a reason. Plugging in a USB mouse, however, was simple and instantaneous.

Which brings me to another wish… When I have foregone the use of the touch pad and have a mouse plugged in, there is no (obvious) way to adjust the speed or acceleration settings on the mouse. The mouse presents an extremely fast cursor that is somewhat difficult to control. It’s so fast, in fact, that I’ve found myself preferring the touch pad over the mouse, which is something I’ve never done before in all my years working with laptops.

Generally speaking, the downside of simplicity is lack of granularity. As a power user, I’d like to be able to adjust things more to my liking.

• The ability to sort the Apps section on the New Tab default page. As of this release isn’t available.
• The ability to control default click behavior of icons on New Tab default screen. It defaults to the current tab. In lieu of a start menu, or finder menu, or dock, I would like an icon to open by default in a new tab, or even a new window should I so choose. You can force it by shift-clicking an icon, but that’s not the default behavior.
• The ability to move docked panel items (Scratchpad, chat windows, music players, etc.) to the sides or top, rather than being locked to the bottom.
• The ability to attach more than one file at a time to an e’mail.

Though I really do like crosh> the lack of support for public/private key based authentication for the ssh client is, to my mind, a step backwards in security in favor of convenience. I’ve disabled password authentication on my own server (where I ssh to the most, by far), and have been forced to consider the issue of allowing password auth, if I want to connect via my Chromebook. From within my own network, I don’t see a problem with it, but if I’m connected to another, public network (either a coffee shop somewhere, or via the Verizon 3G plan), it could be an issue. My hope is that they allow key based authentication at some point.

Not at All
The single most frustrating thing (if I may be so bold as to allow myself to be frustrated with a free laptop) was the “easy” and “seamless” automatic setup of the Verizon 3G connection plan. It failed, forcing a call to Verizon Customer Service. As we all know, calls to customer service nearly always try one’s patience.

I called Verizon at the customer service number listed on the failed activation screen, (800) 786-8419. The first person didn’t know what I was talking about, and wanted to set something up through my already existing Verizon Account (or rather, my wife’s account, which meant she had to be there). So I called her upstairs, and she spent some time on the phone setting something up. I don’t know what it was, and I’m not convinced the CSR did either. Regardless, it didn’t work, so that person sent me over to an “Accounts” rep, who had trouble finding any information about the advertised connection plans.

I should mention that yours truly made the process a little more time consuming by misreading the MEID number, and inserting an extra “0″ (zero) after the “A”. Totally my bad, and I now have an eye appointment set up for January.

But back to the available advertised connection plans associated with the Googlebook… Seriously, it’s all listed right there on Google’s own Chrome OS site, complete with a data usage calculator to help you choose the right plan for you. That this wasn’t common knowledge to basic Verizon technical support speaks to a distinct lack of coordination between Verizon upper management and the phone jockeys (or, in other words, business-as-usual for any corporation – admittedly, I’m a tad jaded on that issue). It’s likely that as more people sign up, the process will become smoother. Apparently, I was one of the first.

Accounts was unable to complete the order, having determined that it amounted to a pre-paid plan, with Google being the purchaser. To Verizon’s credit, each person I was transferred to was friendlier than the last, and none of them were unfriendly. Accounts transferred me to the pre-paid mobile broadband accounts division to complete the transaction.

She was able to get me set up, but in keeping with the not really simple and certainly not seamless procedure, I had to go out to crosh> and manually activate the modem with the `modem status` and `modem activate` commands (the later of which is fairly long, which is made even longer by having someone explain it to you slowly over the phone. It went something like this:

crosh> modem status
This provides , the dynamic modem path. Then…
crosh> modem activate-manual /org/chromium/ModemManager/Gobi/[number] mdn [device phone #] min [the Verizon phone # to connect to] spc 000000 system_id [System ID #]

In typical *nix fashion (ChromeOS is linux based), there was no feedback, so the command was successful. We chatted for a bit on the phone (at which point I learned that I was one of the first), and after a while, all was well, and I was connected.

A bit less easy and seamless than advertised, but I have confidence that each call like mine will make subsequent calls easier. Wait. No I don’t. Customer service calls will never be easy and seamless.

Generally Speaking

Overall, I’m very happy with my Chromebook. The nits I have, though they may seem legion, are minor. Given Google’s track record with constant improvements, I do have confidence that my wishlist, and many other things besides, will be answered in time. Until then, I’ll keep submitting feedback and enjoying my Chromebook. My wife, however, has taken to referring to it as “my other woman.”

These are just my initial impressions… I’ll have more thoughts later, I’m sure.

[Show as slideshow]