Category Archives: Tech
Articles relating to a variety of technical themes. Dividing by Zero is not defined, so this fits in as well as anything else does.
Something Spiffy This Way Comes
Comments are shut down. I’m up to something.
Important Software Updates
Krebs on Security posted on several important software updates last week. (It’s a well-written blog, and very accessible to the average user. Highly recommended.)
Microsoft issued an Emergency fix for IE 6 & 7. Even if you don’t use IE, you still need this because some of the programs Windows uses (e.g. Windows Update) make use of IE.
Apple also published a pretty Huge Mac OSX update. Some of these security problems go back to 2003.
And if you use anything with Java (e.g. OpenOffice.org), there’s an update for that which covers 27 different issues.
Of Flying Pigs
A number of the specifications for how things work on the Internet begin life as an RFC, or Request For Comment. For example, RFC 2616 lays out the basis for the HyperText Transfer Protocol or HTTP which controls how a web browser downloads pages from a web server. (That’s what that “http://” is for at the beginning of all web page addresses.)
Today I became aware of RFC 1925 which discusses “fundamental truths of networking for the Internet community.”
My favorite part is section 2.3 which reads:
With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. It is hard to be sure where they are going to land, and it could be dangerous sitting under them as they fly overhead.
Mozy Backup Failure – Day 5
No update yesterday since I had other things going on in the evening and couldn’t spend more than a few minutes working on getting the restore to work. Mozy does seem to be taking my problem quite seriously though – during the day yesterday I received phone calls from both a product manager and one of their Enterprise support technicians. (My subscription level is Home, I assume this level of attention was due to the nature of the problem – application failure – as opposed to the more common “How do I use the program?” issues which plague most product software support teams.)
The Web Restore I started on Tuesday evening finished sometime in the wee hours of Wednesday morning. I started the download around 10:30 or 11:00 Wednesday evening and when I woke up Thursday morning, the download had failed. I restarted it before leaving for work and when I came home this evening, discovered that this time around, rather than finish the download, my computer had decided to go into sleep mode (the download continued immediately when I woke the PC).
This evening around 8:00, I finished downloading the three self-extracting archive files, this time from the Sunday 4pm backup. I extracted the archives and
So where does that leave things?
The most important thing for me is that I have my photos back. I’m grateful that Mozy’s service worked well enough that I was able to restore them. But Mozy’s not out of the dog house: restoring data shouldn’t be this much of a chore, particularly with a service targeted at home users. The initial restore attempt failed, but the message it displayed simply said, “Restore Completed.” The error message was in smaller text, accompanied by a message saying when my next backup would take place. If I hadn’t paid attention, I might have thought all was well and then, 30 days later, my photos would have been gone forever.
To my way of thinking, a “Restore Completed” message should only be displayed if the restoration was successful. And if it fails, it would be nice if the software made the next restoration attempt automatically rather than reverting to “business-as-usual.”
As for the failure itself, I understand that with any online service, failures do happen from time to time. With something as critical as online backup and restore, I’d prefer to see them a lot less often. For that reason, I’ll be dropping a line to Mozy to let them know I’m willing to work with them to reproduce the failure so they can learn from it and make the product better.
Some takeaways:
- Backup is important. I’ve lost data in the past when I didn’t have a backup regimen in place. (I’m not alone in this, I don’t consider myself to be one, but even a Geek Rock Star can lose data.)
- There is value in online backup – anything which could damage your computer could also damage an external drive. This experience does however demonstrate that online isn’t a good place for the primary/only backup.
- My previous backup scheme was a copy of Norton Ghost and an external drive with backups automatically taking place every night. If I’d still been using an automated solution (in addition to Mozy), I probably would have had everything back almost immediately.
Mozy Backup Failure – Day 3
I’m making progress. Sort of.
I’ve pretty much given up on getting the Mozy backup/restore program to restore my entire photos directory. Every time I try, it runs for a bit and then displays a message saying it ran into some sort of problem communicating with Mozy’s servers — where all my files are hopefully stored.
As mentioned on Day 1, in early to mid January I copied all my data files (including the photos) to an external hard drive before installing Windows 7. As a matter of fact, I actually did this twice.
So I definitely have my files up through mid-January, and I’ve copied them back to the main drive. But my photos since then are still AWOL. Some of those have been copied to Google’s Picasa service and some to Facebook, but none of them are all in one place except on Mozy.
One thought is to try using the Mozy client program to download a smaller set of photos. Instead of the entire 10 years worth, just download the past six weeks worth of photos. For the most part, I do store my photos chronologically, and a chunk of what’s missing is C:\photos\2010\02. That’s easy to restore.
But not all my photos are stored that way. Two weeks ago, I went to the Farpoint convention. Those photos were in c:\photos\conventions\farpoint\2010. Other photos were stored in other locations.
The problem with restoring the photos piecemeal is that I have to remember all the various locations where they restored. It’s much easier to simply restore c:\photos.
In addition to the software which runs on your computer, Mozy also allows you to download your files from their web site. You log into their web site, go to the “Web Restore” section, and select the files you’d like to restore.
The first thing I tried on Sunday was restoring files via the program installed on my computer. It failed. The second thing I tried on Sunday was the web restore, it wouldn’t let me get as far as the screen where you select the files.
I tried letting the restore program run during the day Monday and when I came home, discovered that it had failed again. So I tried the Web Restore again, this time it worked and about 90 minutes later I received a notification that my files were ready for download. I started the download (3 files, totaling about 7 GB).
When I woke up Tuesday morning, the file downloads had failed. Ugh.
Before leaving for work on Tuesday, I managed to download two of the files and had started the third download running. When I came home, the final download had somehow became paused, but I was able to finish it.
So now I had all my photos in one place, right?
That would have been nice. After unpacking the archived files, I discovered that all the files since mid-January were missing from the restored file set! The client program was able to show the past six weeks worth of photos, but the web restore didn’t download them!
It appears that the Mozy application is backing up the files I’ve managed to restore from my local backup and that’s what the web restore gave me.
Taking a look at the Web Restore’s user-interface, I’ve discovered that you can tell it to restore the files to whatever condition they were in as of a certain date. I’ve also discovered that the last backup for which my photos were present was on Sunday, around 4pm.
I’ve started the Web Restore process running again. This time based on the backup from 4pm on Sunday.
Wish me luck.
Mozy Backup Failure – Day 2
Good news and bad news. The good news is that when I switched to Windows 7 in early January, I thought the “custom installation” would reformat my hard drive, so I copied all my files to an external drive. And not quite trusting Mozy after the last round of problems, I never deleted the backup.
So most of my photos have been recovered.
Of course, I didn’t have a local backup of anything since early January – so for starters, all my Snowmaggedon photos are gone, including the only photos I’m ever likely to get of Wylie “swimming” through snow up to his ears. I’d been counting on Mozy to protect those.
Today’s developments:
I started another restoration attempt before heading to work this morning. When I came home this evening, it had failed with a message about not being able to connect to the network. From the message, it’s not clear whether this is a Comcast issue, or Mozy.
During the day, I received a reply from Mozy’s technical support, suggesting that I try restoring the files via their web interface. I tried that last night – the failure message was generic, and I didn’t write it down, but in essence it said, “There was a problem. Try again.”
More interestingly, around 1:15pm, I received an email from someone who as nearly as I can tell (via Google) works in Mozy’s PR department, asking for my contact information. I’ve sent it to him and I’ll be interested to see what this leads to.
Backup Failure
I’ve discovered that approximately 10 years of photos have been deleted from my system. Possibly something I did wrong, I’ll probably never know for sure.
Not to worry, I’ve been Mozy for most of the past year. Sure, there’ve been some glitches in that time (for example, not quite a month ago, I couldn’t complete a backup for over a week, and the time before that I had to start complaining in a public forum) but eventually everything got fixed. So, nothing to worry about.
So I attempted to restore my deleted photos. The initial estimate said it would take about 5 hours to restore. That’s to be expected, it’s a lot of files after all.
Half an hour later, Mozy says “Restore Completed — Cannot connect to server.”
HOW THE BLAZES IS THAT CONSIDERED “COMPLETE”? Complete failure maybe, but not a complete restore operation.
Based on this event, and my previous bad experiences with Mozy’s support system, I hereby retract my previous endorsement of this product.
Computer Archaeology
One of my favorite movie lines comes from Indiana Jones and the Last Crusade. It’s the very end of an action sequence in which Indy rescues his father from the Nazis. The two of them take a moment to catch their breath and Doctor Jones, Sr turns to his son and asks, “You call this archaeology?”
It’s been a day or two since I last had any prolonged fisticuffs with the forces of evil, but every so often I also make a surprising discovery.
This evening, while trying to figure out where I’d stashed my copy of the data access drivers for Microsoft Access, I discovered a file sitting in a dusty corner of my hard drive.
The file is named msie302m95.exe, it’s 10.8 MB in size, and dated July 31, 1998.
I’m nearly certain it’s a standalone installer for the Windows 95 edition of Internet Explorer, v3.02.
IE 6.0 is the current bane of existence for web developers. It has a number of quirks, lackluster support for web standards, and it’s been around since 2001. And because it was bundled with Windows XP and most people won’t upgrade their web browser (not even for free!), it’s still the most common web browser on the web in general.
IE 3.x dates back to the days before Microsoft won the first browser war. This was the first browser from Microsoft to support JavaScript and among the last to still support the <marquee> and <blink> tags. (These days, both of those effects are largely unlamented, but if you feel desperately eager to create something similar to the average Geocities page, you can reproduce them with a very small amount of JavaScript.)
I don’t miss IE3. But one thing’s very clear from this discovery:
I really need to clear out my hard drive a bit more often.
Software Reinstall
Choosing to go from 32-bit Vista to 64-bit Windows 7, also meant choosing to do a “custom” installation of Windows. Meaning that I have to reinstall all my programs.
Still more to be added over the next few days, but here’s what I’ve got so far.
The List:
General Computing
- Open Office – compatible with MS Office
- Mozy Home Unlimited (paid version) – after a horrendous support experience this summer, I’m not as big a fan of this company as I used to be, but I do like the concept of automatic offsite backups.
- Avast! Anti-virus – free alternative to the various paid security suites. (The main deterrent to malware is still a healthy degree of skepticism when clicking links.)
- Foxit Reader – alternative PDF reader
- VLC media player – plays dang near everything in terms of video or audio, and doesn’t bring unwanted baggage.
- Paint.Net – Image editor.
- 7-Zip – Fairly universal archive/extract tool. Even opens ISO files!
- InfraRecorder – Burn CDs and DVDs
- Picasa – photo management.
- VMWare Workstation – For software I don’t trust, for experimenting with other software environments (e.g. Linux), and other situations where my primary computer isn’t the best choice.
Internet Tools
- Firefox – Web Browser
- Firebug – Debugging tool for use with Firrefox.
- Thunderbird – because I like desktop email
- Putty – Client for Telnet, SSH and FTP, plus it’s scriptable. (I’m not sure where to find this anymore.)
- Chrome – Web Browser from Google.
- Adobe Flash Player
- FileZilla – Both regular and Secure FTP.
Developer Tools
- Visual Studio (Commercial version, but a free “Express” edition is also available).
- Subversion – Source code version control
- TortoiseSVN – Graphical User Interface for Subversion
- Notepad ++ – Semi-replacement for Windows Notepad.
- Microsoft Access – Because I haven’t moved everything to mySQL yet.
- Apache web server