The question came up today about relative speeds of scp, tar and rsync (the latter two using ssh as a transport mechanism). While anecdotes and rumors are great for defining security policy (think TSA), I wanted some more concrete numbers so I ran a test.

I set up a script to copy a directory 5 times from my laptop to a server on the same subnet. I routinely pull 3MB/s from that server (over wifi), so bandwidth wasn't an issue. I used /var/lib/dpkg as my source directory. It weighed in a 57MB and contained 6896 files. Because rsync will compare changes between source and destination, I made sure to nuke the directory off the server after every run.

Method:            scp  rsync+ssh   tar+ssh
Average Time:  269.75s      33.6s    24.43s
Bandwidth (mbps): 1.69      13.57     18.66

The results are what I expected, at least as far as scp is concerned. It does not do well with large numbers of small files. It copied each file over completely before it started with the next one. Tar of course put the whole thing together and then shipped it off. Rsync read all the files first, then compared them to the server and then shipped them all in one go. Apparently there were some significant I/O savings to be had that way.

One other important item of note is that scp did not handle symlinks the way tar and rsync did. It dereferenced the symlink and copied the contents of that link rather than copying the link itself. That was a problem because I had picked some self-referential directories before I settled on /var/lib/dpkg.

For your reference, here are the commands I ran to test:

for i in 1 2 3 4 5; do time scp -qrp /var/lib/dpkg [server]:/tmp; ssh [server] rm -fr /tmp/dpkg; done
for i in 1 2 3 4 5; do time rsync -ae ssh /var/lib/dpkg [server]:/tmp; ssh [server] rm -fr /tmp/dpkg; done
for i in 1 2 3 4 5; do time tar -cf - /var/lib/dpkg |ssh [server] tar -C /tmp -xf - ; ssh [server] rm -fr /tmp/dpkg; done

For a while, 7UP was running a promotion on their website wherein they would send you a free can of Diet 7UP, Sprite Zero and Sierra Mist Free, so I filled it out and it arrived a few days later. The box came with a cute little instruction card which I found to be a little humorous. Basically they said to chill the drinks, pour them into cups and drink the Diet 7UP last. Well, that's a little less scientific than we do things around here, so we made up our own strategy.

We labeled three cups and filled each with a separate drink, all at room temperature. Then I closed me eyes and mixed them up really good. We tasted each one a few times before making our decisions. Only after we had finished did we look to see which cup was which.

Of the three, we didn't really have a favorite. They were all excessively sweet. Now, let me preface that by saying that neither Karin nor I drink much soda, and in fact sugar products are pretty rare around our house. I'm sure we're not the target audience of these drinks.

The one thing we agreed on was which one we liked the least. Would you believe it was the Diet 7UP? It was the sweetest of the three and the taste was just a bit off balance. It was very hard to discern much difference. The bubbles contributed there. The other two weren't disagreeable, but as I said none of them particularly struck our palates.

After drinking so much pop, I ended up feeling rather sick. I'm not inclined to purchase any of these drinks.

I wish I could title this story "Actiontec Loss" due somehow to my replacing this Actiontec GT701 DSL modem with something better (say a Cisco). But alas, I'm stuck with it for the near future.

You see, my modem started dropping packets today. I can't figure out a rhyme or reason to it. Yesterday I added a pair of servers to my network. If I must be precise, I actually decommissioned one of them today so there's just the one now. It's only a temporary thing, which is good because them suckers is loud. Adding them involved adding two /24 networks to my little DSL modem. Yes, that is heavy over-kill, but it's due to historical reasons. Anyway, it ran just fine all day yesterday without incident but today the modem decided it had had enough.

I started getting packet loss on anything that touched the Actiontec. For example, here's a ping from my Linksys WRT54G which sits right behind the Actiontec.

--- 10.42.41.1 ping statistics ---
45 packets transmitted, 17 packets received, 62% packet loss
round-trip min/avg/max = 1.3/1.4/2.4 ms

Even more curious, from the DSL modem I saw this error.

# ping 10.42.41.2
PING 10.42.41.2 (10.42.41.2): 56 data bytes
ping: sendto: Operation not permitted

I checked for loose cables or ethernet errors, but there were none. I checked system loads and free memory, all fine. I even tried re-arranging the network to eliminate the WRT54G, which incidentally is still how the system is set up, but that didn't help either. In desperation I even unplugged the server and tore the routes out of the Actiontec, but still the packet loss persisted. From the server to the WRT54G was fine. But from the upstream DSL router to the modem, or from my WRT54G to the modem both had large amounts of packet loss.

It finally occurred to me that I need to just reboot the modem, which had been up for about 90 days previously. Now that may seem like an obvious thing to some, and in the Windows world I suppose it is, but in general it's not a good answer and for Linux and Cisco systems most of the time you can find an actual solution. I guess deep down I still have hope that this modem, which is based on Linux, will grow up to be a real boy, er, router. That hope sunk further today as the reboot magically fixed everything.

I guess the question remaining is how long until the modem does it again. Who knows? At least I know what to do now and not to waste time debugging the network. It's so frustrating. It makes me wonder what Actiontec would say about their poor excuse for a router. Probably something like "we have leveraged enterprise technology to bring a workable product to market at a price point that consumers can afford". Maybe there would be more buzz words, I'm not really up on those at the moment. I don't think they would say anything about quality because I doubt they ever cared. Sell, sell, sell and screw the buyer. It's pathetic and the worst part is, they're not the only ones. It's a trend in this country and it makes me sick, especially when I find myself buying into it.

I have upgraded the site from 4.5.8 to 5.1. Really I just had 2 hours in the middle of the night to kill, so why not do a blog upgrade, right? I suppose I should have been sleeping.

The main reason for the upgrade was to install a captcha module to combat the increasing amount of comment spam I've been seeing. It's just annoying to have to delete comments every day. Well, Drupal doesn't have a captcha module built-in so the upgrade wasn't strictly necessary, but I was a few revisions behind so I figured why not.

There are a few things that have changed and I'll have to work on it further, like the missing post category menu from the left. And the archives disappeared. Definitely not cool. But it's late and I need some sleep. Some other time then. Please let me know if you find anything else broken though.

First off, let me up front apologize for not having pictures. My wife took the digital camera on her trip to Oregon and I'm left here with no way to document my life. How will I ever survive?

So, on IRC the other day, somebody was pointing out the vast difference between two things and used the classic phrase, like "apples and oranges", to which another party responded saying that it was more like "apples and tacos". Well, that got me thinking, why not apples and tacos? I mean aside from the fact that they seem so different. Different can be good. I mean bananas and mayonnaise sounds like a horrible combination but turns out to make a delicious sandwich. So it was only fair that I give it a shot.

I complicated matters a little bit because I came up with my own recipe for shredded pork taco meat. Ideally I should have controlled for everything but the apples, but with my smoker out of commission for so many months I've really been craving to slow roast something even without smoke. I purchased a pork roast for just this purpose. (Incidentally whenever possible I but Salmon Creek Farms pork. Yes it really does make a difference, and generally it's actually cheaper. Win, win!)

Well, for this roast I mixed up an unholy combination of most of the spices in my house (ground black pepper, kosher salt, chili powder, cayenne power, paprika, Tabasco sauce, garlic powder, thyme, spicy mustard and worcestershire sauce). That made a nice paste which I smeared all over the pork. Placed it in a dish, covered and roasted at 250° for oh about 3½ hours. Once it cooled I shredded it and stirred it around in a reduction of the sauces that filled the original cooking vessel. So yummy.

But it was late last night when I finished the roast, so I couldn't make the tacos right away. Besides, I was so full of chocolate that I couldn't bear the thought of eating. Instead the pork went into the fridge until today. I diced some onion, sautéed it in some oil with a little ground chipotle, and then added the pork. Once that was warm I tossed in the diced apple and kept cooking until piping hot. I used a Granny Smith which I figured could take the heat while retaining its crispness and the tartness would stand out against all the spice. The final taco was on a soft white flour tortilla with grated mozarella cheese and lettuce.

The apple flavor was definitely noticeable. I think the choice of a tart apple was wise because a mild apple would have been lost in that taco. I would probably do a finer dice (¼ inch or less). My apple chunks were too big and they seems to overpower rather than blend. There were also spots in the taco where there was no apple flavor and that stood out as well.

One concern I had was that the apple would get all mushy and soggy while cooking. It didn't happen. I really should have given the apple more credit. My wife makes and cans apple pie filling. Those things go through a process of cooking, canning and then baking and still come out in one piece. I should have known an apple could sit in a skillet for 5 minutes without melting.

I ate the whole apple so that's probably the last serving of apple tacos for a while, but I think I'll have to try again and see if I can achieve a more balanced flavor.