Friday Jun 19, 2009

so long

Today is my last day at Sun Microsystems. It's been a good run. I've
been here for a little over nine years. I've filed about 949 bugs,
closed or fixed 604 (likely more), and I've been involved in a fair
number of projects, including PPP, PPPoE, RIP-2, SolarMAX, Quagga,
Kevlar/Zones, Rampart/TX, Zulu/LU, DHCPv6, NWAM, and RBridges.
I've been the PSARC chair on and off for many years, and even had a
tour of duty on the OGB. I haven't lacked for variety.

But it's time to move on. I have a new job and new challenges ahead
of me, and I'm excited to get started on it.

Many thanks to all those over the years who've given me help, clue,
and support. There are simply far too many of you to list.

Keep up the good work. I'm still an OpenSolaris user -- except for
the family Mac, everything I have runs it -- and I'm looking forward
to what develops in the future.

Thursday Apr 30, 2009

Packets out of the wrong interface

The Problem

A common complaint for Solaris users runs something like this:

I have a Solaris system with two Ethernet interfaces connected to different subnets. Sometimes, I see an IP packet come in on one interface, but the packet goes back out a different one.
This behavior is bad for my network, because I have firewalls that check the packet sources, and they drop these misdirected packets. Why does Solaris do this? And how can I fix it? I've tried disabling routing, but that doesn't seem to help.

Problems like this when reported are usually closed out as "will not fix," as for example CR 4085133.

The Why

The underlying problem here is at last partly a misunderstanding of how TCP/IP works. When a system transmits a packet, it must locate the "best" interface over which to send it. By default, the algorithm for doing that is as described in RFC 1122 section 3.3.1. Note in particular section This requires the system to look at local interfaces first -- all of them -- to try to match the destination address. And once we find the interface by the destination address, we're done.

That alone is enough to make things not work as expected. If you send a packet to the local address on ce0 from some other system, but that other system is best reachable through bge0, then we'll send the reply via bge0. It doesn't go back out through ce0, even if the original request came in that way.

When considering a non-interface route (whether only the "default routes" of RFC 1122 or the more flexible CIDR routes of RFC 1812), the system will look up the route by destination IP address alone, and then use the route to obtain the output interface. This often causes the same sort of confusion when a "default route" ends up causing packets to go to the default router that the administrator thinks don't belong there.

I actually consider this a design feature of TCP/IP, and not a flaw. It's part of the robustness that IP's datagram routing system offers: every node in the network -- hosts and routers alike -- independently determines the best way to send each distinct datagram based solely on the destination IP address. This allows for "healing" of broken networks, as the failure of one interface or router means that you can potentially still use a different (perhaps less preferred) one to send your message.

There are some related bits of confusion in this area. For example, some programmers think that binding to a particular IP address means that the interface with that address is "bound" and all packets will go out that way. That's not correct. The system still uses the destination address to pick the output path for each individual IP packet, even if your socket is bound to an address on some particular interface. And, as long as you don't set the ip_strict_dst_multihoming ndd flag (it's not set by default), binding to an address doesn't mean that packets will only arrive on that corresponding interface. They can arrive on any interface in the system, as long as the IP address matches the one bound.

The Solutions

There are many ways to fix this issue, and the right answer for a given situation likely depends on the details of that situation.

  • The main issue here is the kernel's forwarding table, so putting the right things into the forwarding table is one of the first tasks.

    A common problem is that the administrator has set up a "default router," but that specified router cannot correctly forward to all possible IP destinations. Some packets the system sends end up getting misdirected or lost as a result. The solution is not having that router as a "default router," and instead using more specific routes (perhaps running a listen-only routing protocol to simplify the administrative burden).

  • Some systems have a "route by source address" feature. Solaris isn't one of those, though there is an RFE open on it (see CR 4777670). A better answer, in my opinion, would be to do something similar to what's suggested in CR 4173841. That would be, when we have multiple matching routes, to prefer a route that gives us an output interface in the same subnet as the source address.

    It's a simple tweak, and would at least fix the folks who have problems default route selection. It would not fix the problems people with interfaces on separate subnets have, though.

  • Applications that care about interface selection can use IP_BOUND_IF or IP_PKTINFO to select the specific interface desired.

    See the ip(7P) man page on your system for details.

  • If all else fails, you can use IP Filter's fastroute/to keyword on an output interface to put packets right where you want them. You should be aware that when you do this, you're circumventing IP's routing features, which means that if there's an interface or path failure, you may cause connections to fail that didn't need to fail.

Tuesday Mar 17, 2009

first instrument approach

It's been a while since I had any dual (instruction) time -- since
August -- so I went up on Saturday with my primary instructor, Tim.
The plans were a little vague; just brush up on everything.

He had me do a couple of steep turns first. On the first one, I held
altitude, but then forgot to pull out the extra power on rolling out.
On the second, I didn't get the power in soon enough. It's obvious
that I'm a little rusty there, but it wasn't too bad.

Next was some engine-out practice into Plum Island. On the first
attempt, I managed to get myself abeam the numbers at 1000 feet and
really pretty well set up. But then I bungled the rest of the
landing, as I turned base too soon, didn't slip enough, and couldn't
bleed off enough altitude to make a good landing. Rather than
overshoot, I went around. Someone had just taxied off the runway and
he thought I was aborting the landing because of him, so Tim had to
explain that we'd blown it ourselves and he wasn't to blame.

I tried again. I came in a little better, but still too high. Tim
took the controls and put us right down to maybe 50 feet; perhaps
less. With full flaps, we still ended up with a lot of speed, but I
was able to make a decent landing from there. Lesson learned: don't
worry so much about a stable approach; worry instead about getting on
the ground at the best possible spot.

On taking off out of Plum Island, he had me put on the hood and head
for the Isles of Shoals. He called up Boston Approach and got us a
practice ILS 34 into PSM. My first instrument approach! They
vectored me for intercepting the localizer at 3000 feet, and I managed
to do that part without too much incident. I even started a stable
descent without trouble.

But staying within the lateral bounds is much harder indeed. We
wandered back and forth, and once or twice Boston called us and said
"we're showing you to the left of course" or "we're showing you to the
right of course." There really ought to be a "student driver" button
on the transponder ... they must have thought we were drunk. In any
event, I was able to wander in towards the runway, and at about 500
feet, Tim had me take the hood off. Like magic, the runway was right
in front of us. I was surprised and disoriented enough that I had
trouble maintaining my previously reasonable approach, bounced twice,
forgot about the crosswind, and ultimately made the right choice by
rejecting the landing.

Still, a really good experience.

We then went back to LWM under the hood, and he had me do the VOR
approach for 23. This involved a lot more work than the ILS, with a
couple of intermediate altitudes to fly and hold while waiting for
certain positions. After I was at the MDA for a bit, he had me take
off the hood. The runway was a bit to my right, but I made a half-way
decent landing.

It was about 1.5 hours of work, and a lot of good practice.

Friday Sep 19, 2008

"bug buddy" is no friend of mine

In academia, when the name of a subject matter contains the word "science," it's a fair bet that there's little actual science involved, if any at all. It looks like a similar principle is involved with software: if it calls itself someone's "buddy," it's a fair bet that it's unfriendly.

This certainly applies to GNOME's "bug buddy." This utility intervenes on crash of a GNOME application, and tries its level best to format up a problem report. Unfortunately, its best just isn't anywhere near good enough in most cases, and by design it prevents a core dump from happening. That latter attribute means that it actually \*prevents\* someone from fixing the problem, as core dumps contain far more information than just the stack trace that "bug buddy" would have logged on its best day, and thus it annoys me to no end.

After much tinkering, I've found a way to turn this bit of nonsense off in Solaris when logging in via dtlogin. The environment for GNOME is set by /usr/dt/config/Xinitrc.jds. Fortunately, like all good X applications, it looks at /etc/dt/config first, so we can use this to set a new variable.

Create /etc/dt/config/Xinitrc.jds with this for contents:

exec /usr/dt/config/Xinitrc.jds "$@"
Mark the file as executable (chmod +x), and the problem is solved.

For gdm, it's sufficient to add this to /etc/profile:

Unlike dtlogin, gdm reads /etc/profile by default.

Friday Aug 29, 2008


I suppose the title of this entry gives too much away.

This morning at around 8:30AM, I went to Eagle East. Tim and I looked through the required documentation, the plane's records, and did the checklist to make sure everything was in order. By 9, there wasn't anything left to do but to fly to Portsmouth.

I headed out to PSM at 2500. The airport was busy, with a jet taking off as I was inbound, and an Arrow to follow me. I landed on 34, and taxied off Alpha to November and to the former Pan Am Services to park. I was getting nervous, and I almost held up the Arrow departing the runway. We both parked, and it turned out that was Don.

We talked first about how things would go, and what he was expecting out of me. His last words before going into "test mode" were that I was already a pilot, and all I had to do here was keep my certificate.

The oral exam was easy. We started with the paperwork issues, what you need to be legal, what the airplane must have, and so on. He had a couple of odd questions (such as tire pressure), and then we were on our way.

On the ground, he's your friend. In the air, he's all business. We started off with a power-on stall under the hood, then tracking a VOR. We then went to the East a bit and did a 720 steep turn. Then a few power-off stalls at 1500 feet. Then we descended to 1000 feet and did a couple turns about a point.

Next, it was off to Skyhaven. This is where I started getting in trouble. I was all over the place on my altitude, and he was taking it personally. "That'll be a violation." I just did what I could to hold myself in place, but it was tough. I think 80% of it was nerves and maybe 20% weather (the high pressure made for unstable air and a lot of thermals). We went into the pattern and did some short-field and soft-field take-offs and landings. On that part, he was all over me for being too willing to talk on the radio -- it's aviate first, then navigate, and way after that, communicate.

After a few passes of that, we headed back to Portsmouth. I was right on 1100 feet as we came in. In landing, I made my base too short (and too high), and I was left of the centerline. He was all over me on \*that\*. "There's no excuse." And he's right. As he got out of the plane, he asked if I had a cell phone with me. I said I did. He said, "call your wife and tell her you're a pilot; you didn't lose your certificate."

I'm wiped out. I know what I need to practice from now on in order to get better. But I'm now officially a Private Pilot, ASEL (Airplane Single-Engine Land). In the picture below are Beth, me, Benjamin, and Madeline. And, of course, 61976.

Wednesday Aug 13, 2008

good practice

Since my last posting, I've been doing some more dual test prep work. I went up with Gary once as a practice checkride (some areas good, some not so much), then twice with Tim. Today was my first solo in about a month. I did a soft-field take-off (partly because I was head of landing traffic, and it'd take less time) and headed out to the practice area, and did a couple of clearing turns. Then I tried a 720 degree steep turn to the right. Not so good -- I had trouble keeping altitude, but my roll-out was right on. I did a couple more clearing turns, and then did three 360 degree steep turns -- two to the left and one to the right. On each one, I rolled out on my heading, right at my entry altitude, and I hit my own wake turbulence each time. It was incredible! The first time it happened, I was momentarily surprised (did I just run over a cat?), but when I realized it, I yelled. I then went down to 1000 feet and practiced several turns about a point using a blue water tower that's just west of Plum Island. I tried to keep it in tight (as Tim has been demanding), hold altitude, scan for traffic, and keep track of fields near the tower that could be used in an emergency. After a few turns, I rolled out to the southeast, and called Lawrence. Straight in for 23, I did a short-field landing that probably could have been better. I went back up into closed traffic, doing a forward slip and a soft-field landing. I was to the left of the centerline, and a bit slow on touchdown (which made the flare less substantial), but not bad. I parked and called it a day. I'm planning to go up with Sean some time soon (perhaps Saturday), and get some more checkride simulation. It won't be too long until I'm headed to Portsmouth for the test.

Wednesday Jul 16, 2008

last required item

Last night, I spent some quality time with xfig and drew up some cross-country flight logs for use with my trip to Keene, NH (KEEN). They fold nicely for my kneeboard and have all the information I need.

My plan was to challenge myself a bit more. I'd check the wind en route and see what I could do with checkpoints. That was the plan, at least, and I'm happy to say that I learned something, although not what I was planning to learn.

First off, the weather wasn't cooperating. It's beautiful here -- perfectly clear skies, almost unlimited visibility (as much as I'd expect in summer), and no wind. At all. All the way up to 9000 feet, BOS was forecasting 9900.

So much for calculating wind, which is really just as well. I had no trouble on the flight out staying on the EEN VOR radial and pulling checkpoints from GDM, but getting an accurate enough time source was problematic. The plane's clock is near-unreadable except for short timings (good for timing out a turn under the hood, but not good for checkpoints a few minutes apart), and neither my watch nor cell phone (set to "flight," which turns off the radio) has sweep-second.

Getting magnetic heading and temperature (the latter for TAS) was no problem. Working the whiz wheel while scanning for traffic -- not so good. It's much easier at the kitchen table. I think that if I were in a real bind, I could probably manage it, but I'd probably be better off getting the raw numbers and then landing early to replan the rest of the trip. Assuming I'm doing this for fuel calculation purposes anyway when dealing with an unexpected wind, setting down and getting more fuel and perhaps calling an FSS fixes all problems at once.

So, it looks like I'll probably want to get some kind of fancy computer for the usual case. At least now I know what sorts of features I need for it to be useful.

The other interesting note was about comfort. It was 32+C on the surface this morning (90+F), and rising, but as soon as I got up above 3000 feet, it cooled way down, to about 18C. It occurred to me that this is just a feature of a high-pressure system: descending air is going to bring with it cooler temperatures from above, and that means a more rapid than normal lapse rate on ascending. I don't remember reading that in any of the books (or in ground school), but it seems quite logical, and it's nice to know.

At EEN, I met up with what appeared to be a local pilot talking with a Cessna aircraft salesman. The latter was showing off a really nice looking Skyhawk SP with G1000 screens. I asked the pilot to sign my log book, and a few other folks (the pilot's wife and some other guy) showed up to introduce themselves. We talked for a couple of minutes. The pilot told me that they're moving the terminal to the other side of the field soon. I thanked them, and went back to go home.

The trip back home was mostly uneventful. I couldn't pick up Boston Approach on the outer frequency for EEN -- probably because they were extremely busy -- so rather than pester them I waited until I was in range for 124.9, and I stayed with them until I was about over Lowell. I was a little distracted by staying clear of Pepperell (jumping possibly active), and I had to reorient myself once on the LWM VOR, but I had the field in sight (and told Boston so) when they kissed me goodbye.

Total time was 2.0. That's probably about 1.2 in the air, and as much as 0.8 taxiing around and doing run-up. In any event, I've got the last piece of the requirements puzzle, and I now have to talk to Tim about doing practice checkrides and then scheduling the real thing.

Saturday Jul 12, 2008

brushing up

It's been a few weeks since I last flew, so I was expecting this solo practice to be a bit rough. I started out with some minor problems during pre-flight: the fuel hadn't been topped off since the last flight, and the pilot-side window was sticking. I ordered some fuel and found a way to maneuver the window into closing.

I had listened to the scanner before driving in, and it sounded busy. It turned out that there was a fly-in at EAA, so as soon as I switched from getting ATIS (all in one shot this time; my first one-shot copy ever) to ground, I heard a lot of directions and chatter. I waited for my moment and then called "Lawrence Ground, Cessna 61976, at Eagle East, with Information Oscar." The someone else made another call, and the tower answered them before getting to me. "Taxi to runway 5 via Alpha, winds are 110 at 8, caution: mower adjacent to taxiway."

This is my first clue that something is up: the winds are 60 degrees off of the runway, and they're not switching. They have a line of people coming in, one after another, and 5 is the longer runway. I start my taxi and wave to the guy on the mower so he knows I see him. I start my run-up, and I listen as some guys come in to land making position calls as though they were at an untowered field. I chuckle to myself and imagine what the controller must think. "Uh, yeah, Piper, I can \*see\* that you're on short final." It's good to know that it works both ways, and my discomfort with untowered airports is similar to theirs.

I announce that I'm ready, eastbound, and get my clearance. As I taxi out to the runway, I wrestle that window closed. I'm going to need to say something about that. I do one last glance at the oil gauge, then put in the power and, after a short roll, check to see that airspeed is alive. I then remember the crosswind and put the wheel most of the way to the right. I lift off into what has to be my best take-off ever. I'm right on the runway heading the whole way up.

I then headed out to the practice area at 2000 feet. I do two clearing turns, slow down a bit for maneuvering, and I tune in the CTAF. There are a few folks coming in and out at Plum; it's busier than I've ever seen. I head out towards the coast and start doing a steep turn to the right. I go around 720 degrees and roll out on my heading, but I've lost 200 feet -- twice the standard.

I remind myself to put in power during the turn. I then try 720 degrees to the left. This time, I lose 100 feet, but then end up climbing out as I forget to pull the power back out.

Well, this is why we practice. So I try something harder. I do another clearing turn, and I spot traffic above and to the south, traveling west. It doesn't seem a factor. I roll into a steep right turn and apply about 300RPM. After 360 degrees, I roll left into a steep turn. There's a brief moment when I'm less than coordinated, but it turns out well. I do another 360 degrees and I'm about 50 feet off of my original altitude. This is better.

I then do some slow flight and maneuvering with partial flaps. I climb up to 2500 feet and do one convincing power-off stall before heading back to home.

They give me a right downwind for runway 5. I descend early enough that I'm well set up for pattern altitude and airspeed before the time I get there. As usual, though, I end up a bit too tight. The tower gives me clearance before I can even make the midfield call. The wind is now 140 at 8 -- a direct crosswind. I pull the power early, before the numbers. I know that with no headwind, I'm going to be high. My base is almost non-existent and I end up a bit to the left of the runway and high. I pull all the power, put in full flaps and bring it right down. I'm at 70 over the numbers, and I land cleanly, though to the left of the centerline (why am I \*always\* left, no matter where the wind is?). I finish off the crosswind by turning the wheel to the right while slowing to taxi, and I need a bit of power to get off at Delta.

I taxi off, and we're busy enough to be like a real airport. The controller gives me a frequency for ground, and I have to contact them to go park. I clean up, get clearance, and park right on the centerline.

All that was just 0.9 on the Hobbs. After I settle the rental, I let them know about the sticky window. I'm getting a bit more confidence that I can get to the practical test soon. I have to get the sequencing down for those performance maneuvers so that I'm well within the requirements, though.

On Wednesday, I'm planning to go to KENE (Keene, NH) for a solo cross-country. I need another 0.7 worth of solo cross-country time, and that ought to do it. After that, it's probably some practice checkrides with Tim and maybe another of Eagle's CFIs.

Friday Jul 11, 2008

one test down

I've been attending the ground school at Eagle East over the past 14 weeks, and it's an intense effort, especially at the end. There are a lot of random things to memorize (why 10 days to report an accident?), some skills to practice (finding wind correction from course and wind data, finding wind speed and direction from measured course progress), and quite a few things to learn in depth (such as weather).

It paid off, though. I got my sign-off on Wednesday by taking a practice test, and this morning, I went to Eagle East's CATS center, and finished the real FAA test. I missed four questions, giving me a 93%. I wish I'd done better (of course), but I'm quite satisfied with both the results and the help that Eagle's ground school (Sean and Don) gave me. I'm sure I would not have done nearly as well just by reading the books alone.

Between ground school, vacation, and other things, I've been away from flying for several weeks now. I plan to go up tomorrow morning and file off some of the rust before completing my requirements for the practical test.

Friday Jun 13, 2008

more hood work

Today was a bit cooler than Monday, around 80F instead of 90+. He had me do a short-field take-off, and I think I was up in about 700 feet or so. The one thing he caught me at was that I wasn't checking the gauges before releasing the brakes.

Tim is starting to transition from "instructor" to "examiner" on these trips. For this one, he had me put on the hood at 500 feet, and we headed out to the practice area. We practiced holding a heading and controlling airspeed. He then had me take off my hood, and he said, "you're out of the clouds, but your day just got worse: your engine is out," and he pulled the power.

I trimmed for best glide, and looked around. In front and to the left was Hampton, which is a grass strip. I had no idea where I was, so I'm glad that Tim took over making the calls as we approached, though, really, I do need the real-world practice in making those position reports. I was way too high, so I made a standard-rate turn to the right, and came out exactly in position to land. I was on a decent approach (as best I could tell), but Tim "helped" the landing.

Tim did criticize my maneuver. I shouldn't be turning away from the field, even if I can make it come out ok. Instead, I should head right for it, and then circle my way down.

We taxied around, with a Cub landing behind us, and got into a position to take off again, this time with a real soft field. I made my calls, and he had me go through the pattern once. The first time we came around, I was way too high, so I just went around again. The second time, I was much better, though perhaps a little low. I landed cleanly, though.

Then I taxied back and took off again, and the hood went back on. We went east, and then south, and practiced some stalls. One started falling towards a spin, but I got the nose down, kicked the opposite rudder hard, and got back out. Then the hood came off again, and he said, "engine out." After trimming to best glide and running through a slightly better (though still flawed) mock restart procedure, I looked around, and I said, "I know where I am here; that's Plum Island up ahead." I headed towards it, and started slipping to lose the altitude I needed to lose.

Tim didn't quite like that, and I was perhaps still a bit too high. He took the controls and did a 45-to-60 degree bank in a circle to lose a bunch of altitude. He gave it back to me, and I landed without much trouble. I made the calls, taxied back, and did a short-field take-off. This one turned out quite nice.

Then the hood went back on, and we headed back to Lawrence. On the way, he had me do unusual attitude recovery. He said that the traditional way to do this is to have the student close his eyes, then the instructor does something crazy, and the student tries to fix it afterwards. He said that this is just unrealistic; nobody's going to do that to you in real life. In real life, you're going to get distracted while you're in the clouds -- maybe talking to ATC -- and _you_ are going to mess up the plane's attitude.

So the procedure was that I closed my eyes, and he called out a series of maneuvers for me to perform. First a standard-rate turn to the right. Then a turn to the left. Then a climbing turn to the left. Then leveling out. He had me open my eyes, and I looked at the instruments. The airspeed was stable, and the wings were tilted a bit. I fixed that, and then said, "uh, it looks like I'm done." He said he was a little disappointed, because I didn't get nearly as messed up during the eyes-out part as I should have. Oh well!

We went back to Lawrence under the hood, and he had me take the hood off in the downwind. He wanted me to do a power-off accuracy landing, and hit the 500' mark. I did ok on making the approach without power, but I was long on the landing because I tried to keep it too tight. I don't have that judgment yet -- I haven't practiced power-off much.

I'm on the schedule for next Friday. I plan to do a solo cross-country if the weather holds. Right now, they're calling for poor weather, but that always changes. This is the last required item on the list. I need to do a good bit of practice, and maybe some more prep time, but the requirements are just about done.

Sunday Jun 08, 2008

the "f" in "I'm safe"

Tim and I went up to knock off some of that simulated instrument time I still need, and to get some flight test prep time, as I think I'm getting close to taking the test.

We started with a soft-field take-off from Lawrence, which was a bit shaky. I haven't practiced this in a while.

The temperature in the morning was pushing 90, and it just got hotter as the day went. We headed out to Portsmouth first, putting on the hood at 500 feet AGL. Tim had me do the approach and descent to about 500 feet before I could take the hood off again. A quick side-step, and I landed without trouble -- not too hard to do with an 11000 foot runway.

We taxied off, and Tim pointed out where I'll go for my practical test, then we sweltered in a run-up area while I set the radios up for a short trip to an untowered airport for practice.

I sounded like a rank beginner when I got there. I think it was mostly the heat -- and fatigue -- that caused this. I keyed the microphone, let out a huge "uuuhhh" and proceeded to give one position report after another that left big question marks in the sky. Sigh. An uneventful landing, then another soft-field, but a little better.

Then more hood work on the way back to Lawrence. Somewhere near Portsmouth (I can always tell where that is by the turbulence), he had me do a stall and recovery. It's nerve-wracking under the hood, because I can't see where the nose is going, so I have more worry about spinning. In recovery, I forgot the carb heat, and Tim let me know it.

I then used the VOR to get back to Lawrence, and took off the hood at around 1000 feet. We were at 1.9 hours total time (around 1.2 under the hood), and I was drenched in sweat. I'm feeling a little more confident, though.

Friday Jun 06, 2008

raining again

Another rainy day here in North Andover. I went to Eagle East with the hope of getting a picture of Cleo (Tim's dog) for my son's class project, but neither was there. I just updated the schedule to go back on Monday morning, and I hope we'll get some good time in. The forecast calls for a high of 94, but maybe the morning won't be too bad.

Thursday May 29, 2008

catching up

I'm posting this a bit late, as I've had a couple of busy weeks.

I mostly needed to get rid of some of the rust that had accumulated since doing my night work, so I headed out to the practice area by myself for a little practice, taking off from runway 14.

I started with two clearing turns, then did one standard-rate 360 degree turn to the right, and then one to the left. I had planned to do some steep turns, but it was getting a bit bumpy, and I was unsure of myself, so I skipped that.

I then headed down to 1000 feet AGL, just west of Plum Island, while listening to the CTAF. There's a nice blue water tower that's by itself, and I did several turns about a point with the tower as a reference. There was some wind from the southeast, so this was good practice.

I then went south to do some S-turns above I95. Nothing special there, but I did what I could to make my transitions between turns smooth and on-time.

It was time to head back to LWM. The tower gave me a left downwind to 14, so I entered the pattern near the cat-in-the-hat water tank. My approach was less than perfect, and I was a bit high on final. I put in 40 flaps and just brought it down.

The whole time I was out there, I just felt a bit off. Partly, it was the time since the last flight. Another big part was the time of day -- afternoon rather than morning. Still another was the heat; it was stifling. But it was just one of those days.

I need to get some time with Tim, do some more hood work, and practice up for the test.

Tuesday May 13, 2008

night moves

The last night flying I did was quite a while ago. On that trip, I got just two take-offs and landings to a full stop. So, per the regulations, I needed another 8 in my log book.

I scheduled time with Tim tonight in 61976 (my usual plane), starting at 8:30PM, and I got to the airport at about 8:20. Civil Twilight today started at 8:31. If I'd been thinking about it, I probably would have scheduled for 8PM. It always takes at least a half hour more than I expect to do the pre-flight, run through the checklist, get clearance, do the run-up, and get ready to roll. By the time we headed out of runway 5, it was almost 9PM, and the tower was closing down the pattern.

So, we headed off to Beverly. I flubbed the navigation a bit -- the top VOR with the integrated controls sometimes doesn't like me -- but with the airport only 13 miles away, it was easy enough to find the beacon visually. I made a straight-in approach to runway 16, then taxied back to go up again.

It was a nice, cool, clear, and calm night. We had had strong gusting winds all day, but they were predicting that they'd die down by evening, and they were quite right. At least I didn't have to correct for wind too much in an unfamiliar pattern at night -- with the PCL timer cutting out on me every now and then just to make things interesting.

We went around the pattern five times, taxing out to the other end each time to take off in the other direction; a luxury afforded by the lack of wind, and probably a good thing for the neighbors. On the last landing, he had me land with lights out to simulate an alternator failure. It's hard to judge the ground like that, and I came down a bit firmly.

We headed back to Lawrence about two minutes to 10. They close the tower at 10, so we came in doing position calls. I picked runway 32 (approximately straight in) for my first approach. My reasoning was that I could \*see\* the runway in front of me, and I was already low enough to make a good flaps-and-lights-out approach. Tim pointed out that it was the short and narrow runway, and that I should have gone in the pattern. I came in, made a sloppy sort of slip, and landed hard again.

We taxied out to the far end and turned around to take off on 14. I went up and turned left into the downwind for 23. I made my position calls, then turned base. I was a touch high, and Tim said to slip it. I eased into the best slip I've ever done. The nose pivoted to the right, I stayed right on the track I wanted, and I got at least 1200 fpm descent, right to the runway. I let out the slip, flared, and landed firmly (again without lights), but pretty much where I wanted.

I taxied off at Alpha near the approach end of 5, after taxing the long way down the runway. Tim said I did the best I've done so far, and I told him about Sean's advice regarding trim.

That's it; my full 10 landings are done. On to the next item to check off.

Monday May 05, 2008

more practice

A clear day with no wind: a good time to practice those landings. What I've discovered is that if there's a headwind, it'll slow your actual ground speed, and being a little fast on approach won't make much difference for the ground roll. Every landing looks good. But on a calm day, you have to be as slow as you can make it, or you'll have trouble. Plus, I tend to be a little anxious to turn base as soon as I'm able, and that can easily set me up with too much energy for the rest of the landing. So, as contrary as it seems, it's the calm days, not the windy ones, that are the tougher. (A tailwind is far harder still.)

I headed out to the practice area, did a couple of slow and deliberate clearing turns while practicing my scanning, and then entered slow flight. First with 10 degrees of flaps, then 20, and trimmed down to 60 MPH. I did a few turns in slow flight, and one power-off stall, and then headed back.

The tower gave me a right downwind entry for runway 5, so I made my descent towards pattern altitude and came in. I intentionally kept it a bit slow -- 90 MPH -- so that I had time to think about how close I was (spacing out that downwind is tough if you don't have the rhythm and mental picture of the turn when the numbers hit 45 degrees) and line up my ground track. I then proceeded to make a somewhat high but doable approach using full flaps. I ended up getting permission from the tower to turn off at the runway intersection and taxi back. Obviously, I landed longer than I wanted.

I taxied back to the start, and tried a short-field takeoff. I set the breaks, applied full power, and then rotated up as soon as I could. Then nose down a little in order to get to a normal Vy. I was off before the touchdown zone, so I think I did fairly well. On approach, I tried to do a short-field landing as well. I got the airspeed right and was able to get off at Delta, but I touched down longer than I'd want.

I taxied back and tried a soft-field takeoff. I set 10 degrees of flaps while waiting at the hold short line. Then I taxied out and, without stopping or slowing, made a turn to the runway centerline and brought up full power. I tried to hold the elevator back a bit so that I was keeping pressure off the front wheel. Just as with Plumb Island long ago, I had trouble keeping it in ground effect as I was supposed to do. It just wants to fly. I got up to a couple of hundred feet, accelerated to 85, and retracted flaps.

For this last landing, I tried short-field again. I set full flaps and aimed for the numbers. I set down shortly after, and hit the brakes. I had to bring power back in order to make it to the first turnoff at Delta. Success! I taxied back and tied off.

I talked with Tim, and we'll be doing some night work next week (I need 8 more takeoffs and landings at night), and my plans for ground school and the tests coming up.




« December 2016