Pages

Sunday, 31 January 2016

Temptation

There's a computer on Ebay with 256 gigabytes of memory, and eight 2.9Ghz Opteron 8389 cpus (each of which is quad core, so that's 32 processors) for £800. Rack mountable, 4U high.

I am *so* tempted!
 

I had a dream

I had a dream last night.

I was standing on a bridge over a river that was flooding, and I saw, in the swirling water below, a bicycle.

So I got down, into the river, and pulled the bike out. While I was doing it, I noticed another bicycle, and another.

Don't ask me how all these bicycles got washed into the river, the dream didn't explain that.

Altogether, I rescued 13 bicycles, then wheeled them to the local police station so that they could be reunited with their owners. The policeman there said they'd keep them for six months, and release a bike to anyone who had reasonable evidence that it was theirs (by the way, I have no idea whether this is actally the law). But I couldn't think how I'd show that a lost or stolen bike actually belonged to me, or if I did lose a bike, why I'd think to go to a particular police station, so I was fairly confident that after six months, most or even all of those bikes would become mine, and I'd be able to sell them on ebay, probably for around £60 each.

A good night's work.

Saturday, 30 January 2016

A bit more fettling

I put end caps on the new cables I added, put the bike on a stand and checked the rear brake. It was binding at one point in the wheel's revolution, and after a bit of fumbling around, I realised that the wheel wasn't true.

A true wheel, when rotating on its axis, and looked at from behind, doesn't appear to wobble from side to side, or up and down. You true a wheel either with a special and rather expensive wheel truing apparatus, or you do it in the forks. The apparatus just shows you where the wheel isn't true; you get exactly the same effect if you rotate it in the forks.

This was the new wheel with the 11-28 gears, and I was surprised to see that some of the spokes were loose. I had assumed that it had been trued before they sent it, but there's no good reason for that assumption. Truing a wheel takes time and a bit of patience, and time's money. So I got out my spoke spanner, and pinged the spokes. You can tell when a spoke is slack, because the ping is either very low pitched, or if it's actually loose, it doesn't ping at all. So I tighened up the ones that were loose, then the ones that were slack, then I worked on truing the wheel.

The Holy Bible for bike maintenance is Sheldon Brown, and he's done a good article on wheel truing, so I won't explain the process; read Sheldon. To be more precise, Sheldon is like Moses, his writings are like the Bible.

Friday, 29 January 2016

Hatley Heart attack, the last of the series

I went out today with my newly fettled bike, to pick up the rest of the series (except the finals).  A rather tiring day, because I went a long distance; this is the first time I've run through both of the 20 AH batteries - fortunately, I also had a 10AH as an emergency spare.

I had just done a Trigtastic, and I was sitting on the bike on the road, looking at the map, working out where to go next, when I noticed a man striding purposefully towareds me. You know that walk? "I'm important, and I'm going to say something important to you."
He was one of a shooting party; they had guns and dogs and those green jackets.

So I put on my "I'm just a stupid peasant" smile, and said "hello".

"Are you cycling around here," he asked. I blinked, looked at my bike, and thought that it would be foolish of me to deny it, what with the bike as a silent but eloquent witness, so I said "Yes. Yes, I am." He pointed north, in a direction where there is no public right of way. "Did you just come down that track?" "No," I said, truthfully, I didn't." He looked at me accusingly. "How come your bike is all muddy., then?"

I looked at the scotch egg he was eating. I didn't think there was much chance of me being given a bite of it, so I didn't ask, but I did the thing that cats do, which is to gaze optimistically at the food and attempt to implant the idea in his head that he should offer me some. "I went along a muddy track," I explained helpfully.

He frowned, trying to absorb this complex information. "Did you come from over there?", and he pointed in the direction my bike was heading towards. So I pointed behind me, "No, I came from over there." At that point, I decided that my attempts to share his scotch egg weren't going to work, so I said "And now I'm going. Cheerio", leaving him standing there none the wiser about who I was, where I'd come from, where I was going, and exactly what criminal activity I might be up to. The game is called confuse-a-muggle, and it's rather fun.

I found 67 caches today, which is a good number, being the same as my years. This is one of the sights I saw.



This was about 30 feet high!

Thursday, 28 January 2016

A bit more bike maintenance

After the major revamp, I took the bike for a test run, out to Amersham and back, doing several caches along the way. There were three problems.

The first was that the front wheel was going bump  ...  bump  ...  bump as I went along. I knew what the problem was - the tire was worn on the sides. It's because the bike gets clogged up with mud, and this rubs against the sidewalls and wears them. Looking at it carefully, I could see that the wear was pretty bad, and I don't want a blow-out when I'm mikes from anywhere. So I replaced the tire, which wasn't a big job.

The second problem was a grinding noise as I pedalled. I tracked that down to one of the pedals. I gave it a shot of oil, and now it runs smoothly; I also put a rubber cap on it to stop water and mud getting in. I might have to replace that pedal eventually.

The third problem was the bike getting stuck in the highest gear, but that seems to be an intermittent problem, and I'll go with it; I'm planning to go out tomorrow.

Something useful I learned - plastic tire levers aren't a good idea; when you put a lot of leverage on it, it snaps.

Computer maintenance

My laser printer has been going grey. I know how it feels.

I have a 20 year old HP Laserjet 6P, it's practically an antique. But it keeps on working. Occasionally, of course, it runs out of toner, and toner cartridges, if you buy them HP-branded, cost an arm and a leg. I think printer manufacturers hope to make their profits on the toner.

But there's a couple of ways around that.

The first is to buy cartidges from third parties. One of the benefits of having a big name brand like HP, is that there's lots of third parties offering cartridges.

A better way is to refill the cartridge with toner, and that's what I just did.

A few years ago, I did this, and it involved using my soldering iron to cut a hole in the plastic cartridge. I poured in a pint of toner powder, then resealed it using duct tape. This time, I peeled off the duct tape, poured in another pint, and resealed it. I'm guessing that this will last me another few years.

I get 230 grams of toner (and that's about a pint) for £13. A cartridge costs several times that.

Bike Maintenance

First, I wanted to get the three front gears working. To do that, I wanted to replace the cable, but I can't see how you do that in the shifter I have. It's a combined gear shifter and brake, and I decided to replace it with two separate units. I already have these on my box of bits.

The inner cable of the shifter looked slightly rusty, so I gave it a good stroking with emery paper, then oiled the inside of the outer, and it ran smoothly. I installed the new shifter, and it worked! I can now change gears on the front sprocket - I don't want to do this very often, but it does help when going up a steep hill.

Next, I wanted the new rear gears, which go as high as 11 teeth (previously, I could go as high as 14). First, I taped the inside of the wheel. Then I put the tire on, on one side. I put a bit of air into one of my very thick puncture-resistant inner tubes, and pushed that onto the wheel. Then I added the gel insert, that sits between the tire and the inner. All this is to avoid a thorn giving me a puncture, becuase I ride on rough grass, and I'm often going over brambles and blackthorn. Finally, I put the tire on fully, and I was surprised how easy that was - no tools required, and not much thumb pressure. I inflated it up to 45 psi, and now it was ready to go on the bike.

I turned the bike upside down, and took off the back wheel. This is the wheel that iuntil recently, had been my spare, and it was going to become a spare again. The new wheel went on easily. I've never used a wheel with quick-release skewers, but it was obvious how to do it. One benefit of this - there's a spanner I dn't need to carry with me in future, saving a few ounces of weight.

The back gears needed a very slight adjustment, ditto the back brakes. I oiled the chain, and it all looks good.

So now I have an electric bike with a two-speed motor, I can use an 11-toothed rear gear for when I'm going fast along tarmac, and I have access to very low gears for when I'm struggling up a steep hill. The brakes both work, the bell goes "ding" and I even have a clock/thermometer attachment!

Tuesday, 26 January 2016

Hartley Heart Attack, part 11

I parked in the same place as last time, and the first thing I did, was to finish off the Quizzical Smile series, including the bonus. Then I continued on with the HHA,  A total of 51 caches done today.

I got back to the car at 4:30 while it was just still light, had my (late) lunch, then drove to [ick up something I thought I'd need.

This is about the bonus bonus. The cache page said that the coords were in cache 424. I've done that cache, but didn't pick up the coords, so I decided to go back there - a special trip.

The cache wasn't there.

I've found it before, so I know where it ought to be, and it wasn't there.

So I phoned Marcus. There was a howling gale going on at the tome, which made it hard to have a conversation, but the gist of what I learned, is that I'd trekked out there for no real reason, because the coords for the extra final are not in this cache now. It's a numbering error.

So it wasn't that I failed to notice the coords, it's that they weren't there when I did that cache.

Still, it wasn't entirely a wasted journey. I replaced the cache smile emoticon.


One track led me across a claggy-mud field. And as happens in this situation, moving the bike along got harder and harder, and when I was about 30 yards from the field edge, it got nearly imposible, and each pull at the bike was moving it mere inches. I was very glad when I got to the grassy verge!

Friday, 22 January 2016

You can't have it both ways.

There's a fixed (and diminishing) number of police working on actual crimes (that's apart from people claiming to be insulted or offended). Scotland Yard has a shortage of detectives; they're 800 light.

The disappearance of Madeleine McCann, was obviously tragic. But does it really make sense to continue to deploy police resources nine years later?

The historic sex abuse cases; popular stars groping teenagers, or worse, 40 years ago - certainly this is a disgrace, but what about us. Now?

The Rotherham child sex scandal that came to light a couple of years ago, was first looked into 17 years back. If the police deployed on the investigation into Jimmy Saville (I'm not defending him, but he is dead, and so cannot be prosecuted, and is also thereby guaranteed not to repeat any crimes in future) had been deployed in Rotherham, is it not likely that some of the 1400 abused children would have been saved from their awful experiences. To that, add Rochdale, Derby, Oxford, Bristol, Telford, Banbury, Aylesbury and Keighley.

You can't have it both ways. If you want historic crimes (that happened 40 years ago) investigated, there will be less police available to investigate today's offences.

This is important. The murder rate in the UK, for example, is up by 14%. If we can't afford the cost of more police (and the police budget is being cut), then it's necessary to prioritise.

And I think that crime today, is a lot more important than the investigation of Saville.

Haram

In a place that I cannot disclose for obvious reasons, in a country far away, a furtive group of criminals meets on Wednesdays. They meet in heavy diguise, for fear that the Religious Police might infiltrate their group and betray them - a flogging would be the best they could hope for. Because the purpose for which they meet, is haram. Not Kosher. Their activity spreads enmity and hatred; it's a gateway to gambling and other vices. It's a waste of time.

They sit, mostly quietly, in chairs, in pairs. They are probably mostly men, but it's impossible to know under their burka disguises.

They do not use the dread equipment commonly in use for this purpose, for that would be a dead giveaway of their foul actions. No, it's all in the mind, but for the occasional murmured phrase. Hark! Listen carefully, and you'll hear their characteristic sounds.

P-Q4.


More bike maintenance

I used the special tool to get the gears off the wheel. The special tool was turned by a socket wrench, and I added a four foot pipe to the end of that, for leverage. And that was all needed, it was really stiff.

Once that was off, the freewheel could be taken apart using a hammer and screwdriver, and it was caked inside with either mud, or maybe congealed oil. So I'm leaving it for a while to soak in degreaser, and maybe it will work when that's done. And there's ball bearings; 19 at one end, and 37 at the other, and I think I have them all.

With that done, I thought about the other gear cassette that stopped working in exactly the same way a few years ago. So I dismantled that too, scraped out the gunk from the inside, and it's also sitting overnight in degreaser.

With any luck, I'll have two working gear cassettes. There's no hurry though, because my spare rear wheel is looking good.

I also took the plunge and ordered a replacement back wheel. This uses a freehub instead of a freewheel; the difference is that a freewheel screws onto the wheel with a thread; the freehub is part of the hub, and the cassette slots onto it. The big reason for ordering this one, is that A) it will give me gears from 11 to 28 teeth (now I've got 14 to 28). That means that the same rotation of pedals will get me 30% further. So when I'm whizzing along tarmac and I can't pedal because I can't turn the pedals fast enough, that will help. And B) it uses a quick-release instead of nuts. I'm not expecting to have to use that much, but it does mean that I won't need the very large spanner that I've been carrying just in case I need to fix a puncture problem. Slight weight reduction!

I had another go at the front gears on the bike; I can change from high gear to low, but changing from low to high needs an amount of pressure on the gear changer that's absurd. I oiled the cable, and it does move freely in the outer, but that hasn't fixed the problem. This is one for future thought. I've been doing fine without the front gear changer up till now, so it won't be a great loss. This is because when you'd need the lower gears for going uphill in a normal bike, the motor on my electric bike means that I don't really need such low gears.

Thanks to Jason for helping me dismantling those gears, and with all the other help on the bike!

 ... later ...

Maybe I need a new front derailleur mechanism? I've ordered one on Ebay.

Thursday, 21 January 2016

Bike maintenance

After yesterday's bike problems, a big heap of maintenance was needed.

First, the back wheel problem. I removed the wheel. The inner tube is done for, but the tire and gel insert still have lots of life, so I washed the mud off them and stored them. I had a quick look at the wheel, but it wasn't looking good. So I doused the workings with penetrating oil, and set it aside for later.

I have a spare back wheel - I got it from Ebay a while back, so I checked that the tire is Kevlar, the gel insert is in place and the inner tube is one of the thich puncture-resistant ones, and put that on. Then I had to adjust the gear cable.

In doing that, I found that the cable wasn't moving freely in the outer, and changing  gear was impossibly difficult. I pulled off the outer, and saw where the problem was - the last couple of feet of the inner was rusty. I could see a break in the outer, and water would have been getting in. So I decided - new inner, new outer.

I partially dismantled the gear changer to replace the cable, but I discovered that A) the cable has a terminator that is different frmo the ones I have in stock, and B) I couldn't get the cable out of the gear changer. I was thinking, this probably means I have to replace the whole gear changer, which isn't as bad as it sounds, they're only about £25 (I have a combined gear changer and brake lever). But then I had an inspiration.

I cut the rusty part of the inner off, and cut a new length of outer to replace the outer that was letting water in. I took a length of inner from my box-of-bits, so that gave me a good outer, and a good inner, except for one little problem - the inner was in two pieces. And that's where the cunning plan came in. I used an electrical connector, the sort where you have a brass tube and a screw at each end to hold the wire in place. I put both the ends of the inners into that, overlapping, so that each of the two screws hold both wires. I've tested it, and it seems to hold fine!

Then I decided that the rear brake wasn't running smoothly enough, and again that was a cable problem. I replaced the outer and the inner, and now that's fine. I also replaced the outer and inner of the front brake.

One slight problem - the rear brake was now fouling the back rack. I fixed that with a longer retaining bolt and half a dozen spacer washers.

I also spent a while on Ebay, looking at what's available. I thought about having a rear disc brake, but the bike doesn't have a lug for fixing it to, so I think that's out. Then I thought about using a hydraulic brake instead of the cable inner-and-outer, because the problem of corrosion is pretty inevitable. But a hydraulic brake system would need such a major reworking, and can I get a combined brake/gear shifter? The fact is, I don't use my brakes much while caching, because I'm not on the road much. So I'll stay with what I've got. My road bike does have hydraulic front brakes, and it has discs front and back. On the road, being able to stop in a hurry can be really important.

And then I thought about replacing the spare back wheel. First, I tried to dismantle the one I just took off the bike, but I couldn't get the gears off. And anyway, I'll need a new freewheel, for sure.

But I looked through Ebay, and I can get a back wheel with a seven gear cassette that offers a range of gears from 11 teeth to 28. Currently, I'm on 14 to 28. If I switch to 11, that means that when I'm going at a bit of speed along tarmac, I won't have to be cranking the pedals as fast as I do now - sometimes, especially on a bit of a downhill, I can't crank fast enough to keep up with the bike! And when I'm on rough ground, the top gear is irrelevant, because I'm usually in third (out of seven).

The one I'm looking at is £29 (plus £8 postage). No tube or tire, but that's fine because I'd want to use the ones I prefer. I haven't had a puncture for several years now. I've had valve failure twice (that's what happenedd on Tuesday) but considering that I'm mostly riding over rough ground, and there's brambles, and blackthorn, that's pretty good!

I've also removed the thermometer that monitors the motor temperature, because in the winter, that's not going to be an issue - I might put it back in summer.

And the thing I had that gave me a brake light at the rear, and a musical horn, and a turn indicator? I've hardly ever used any of that, and I had to grab the bike as it tilted over, and I grabbed that thing, and it broke. So I've removed the whole thing - not much of a loss.

Also, I took one of the pedals off because it wasn't rotating as freely as it should, gave it a shot of oil and now it's fine.

So I think I'm ready to roll again!





Tuesday, 19 January 2016

A day of a dozen disasters

Today was another bite at the Hatley Heart Attack. But things soon started to go wrong.

1. A traffic jam on the M25 meant that I was an hour late in starting.

2. When I got the bike ready, I fiound I was getting no power. I tried various things before finally giving up; I was planning to walk the route, which I probably can't do, because of my foot problem. But just as I was taking things apart, I noticed that a connector had come adrift. When I fixed that, everything worked!

3. So I set off down the bridleway. The ground was very soft and soggy, making for a difficult ride. There was a hedge, and I could have gone down either side. Of course, I chose the wrong side, and had to back track.

4. I forgot that I needed to collect letter values from the "Quizzical smile" caches, so I missed the first few. I went back and got the first one, and I think I might now have enough to work out the final.

5. After a few caches, I noticed that my back tire was almost completely deflated. Luckily, I carry a pump (it isn't luck), so I was able to pump it up. I suspect that the problem is the tire pressure valve cover, it's letting air out.

6. I was doing a series, "Quizzical smile". I got one of the puzzles wrong, which took me across a muddy field, caking the bike in clingy mud. And in the middle of the field, I realised that this wasn't working, had another look at the puzzle, re-solved, correctly this time, and had to backtrack about a mile to collect the cache.

7. By now, it had turned really cold. I had six layers of clothing on top, three layers on my legs and two pair of socks. But my feet were still getting cold.

8. I took a picture of an unusual notice. I ate horses no dogs



Then I saw the view from there, and tried to take that, but the iPhone hung, which is a potential problem, because that's  my backup plan for if I break a leg, or my GPS stops working, or if I just want to get the most up-to-date infor from the geocaching web site. When I plugged it in back home, the iPhone rebooted, and it's behaving itself now.

9. Then the freewheel on the bike started playing up. The effect, when it was fully non-operable, was that I couldn't pedal; turning the pedals didn't transmit any power to the back wheel. So I was going along purely on battery power. Luckily (it isn't luck), I kew that I had enough to get me round the circuit back to the car.

10. And then the back tire blew. The inner tube valve came off the inner tube, leading to a sudden and total deflation. The only fix for this would be to replace the inner tube. Luckily (it isn't luck) I carry a spare inner tube and all the tools I need to change it - I've had to do this once before. But I weighed up the pros and cons. Against changing the tube, there was A) the back wheel and tire were caked in slimy mud, which meant I'd get mud insire the tire, and I'd have to redo it when I got home. And I was only two miles from the car. So I walked the bike for about 800 meters, to where the bridleway met the road, and chaing it up to a tree. Then I continued on foot bak to the car, and then drove round to pick up the bike.

11. And somehow, while doing all that, I lost my left glove, which isn't too big a problem, because it was pretty much worn out anywy (holes have appeared).

12. When I got home, I needed to power wash the bike to get all the clingy mud off, so that I could fix the freewheel and tire on a clean bike. But the external water tap was frozen up. I had to use the other external tap (near the kitchen), which meant wheeling my wounded bike all the way round the back.

51 caches done, no DNFs. A great day out!

Monday, 18 January 2016

Memories of the Brain virus

Brain was the first computer virus I saw.

It was 1987, and we were selling a range of utility programs, and had a good business in data recovery. People came to us having lost all their data, and we would get it back for them, with a no-fix, no-fee guarantee, and a 95% success rate. Hard drives were 10 or 20 megabytes, and we also did floppy disks.

So when a lady at the University of Bradford saw that the volume label "(c) Brain" was appearing on several floppy disks for no reason, she phoned us up.

Ladysolly took the call, because in those days, I had a day job. She persuaded the woman to send us the floppy, because she knew I was interested in computer viruses. There was a lot of talk on this topic at the time, but no-one seemed to actually know anything. And there was a major confusion between trojans and viruses.

I'd written a story. It was pure fiction, and I called it "The Doomsday Virus". I wrote for a lot of magazines at the time, and I sent it to PCW. The editor said that he liked the story, but didn't want to publish it in case it gave someone ideas.

So the diskette arrived. And I did nothing for several days. Ladysolly was quite disappointed, and reminded me a few times. But eventually I cleared my workload sufficiently that I could tackle it.

I took it seriously. I used a spare room that we had. I took a computer into that room that had no hard disk, and several floppy disks, and set about trying to see what was going on. Sure enough, I found that if you tried to boot from an infected floppy, even though it wasn't a boot diskette, the next disk you put in the floppy drive would be infected, and get the volume label "(c) Brain". But how was it doing this?

Because I was into data recovery, I knew that on the boot sector of the floppy disk, there's a tiny program that looks for the Dos operating system and loads it. So I looked at the boot sector of an infected floppy disk using one of my data recovery tools - it looked perfectly normal.

So I tried all sorts of other things, and after a while had passed, I'd managed to infect several of the disks in that room (except the write protected ones, which was a confirmation that you can't over-ride the write protection). Eventually, I used a program that is part of the standard Dos distribution called chkdsk, which looks for FAT chains that don't have a directory pointing to them, and various other inconsistencies, and chkdsk reported 3kb of bad clusters.

That was the clue I needed. On a 360kb floppy disk, if Dos format finds a bad sector, it marks the whole track as bad. So you can have zero bad sectors, or 5kb bad sectors, but you can't have 3kb of bad sectors. Something fishy must be going on.

So I looked at those bad sectors, which weren't actually unreadable, and one if them I recognised as a standard boot sector. So what's that doing further down the disk? I looked at sector zero again (side zero, track zero, sector one) and that looked exactly the same.

And then the penny dropped. It was fooling me. It had trapped interrupt 13h, and when I tried to look at (0, 0, 1) it was showing me the boot sector that it had parked further down the disk. At the time, we didn't have a word for this trick - later on, we called it "stealth".

So what I had to do, was boot from a clean Dos disk before looking at an infected disk. I did that, and saw on the boot sector:


Welcome to the  Dungeon
(c) 1986 Brain & Amjads (pvt) Ltd
VIRUS_SHOE  RECORD   v9.0
Dedicated to the dynamic memories
of millions of virus who are no longer with us
today - Thanks GOODNESS!!
BEWARE OF THE er..VIRUS  : \this program is catching
program follows after these messeges..... $#@%$@!!


This is one of a number of variants - another variant has three Lahore phone numbers.

So I captured the sectors of virus into a file, and worked on that, disassembling it to see how it worked - you can use debug (part of the Dos distribution) for that.

Then I wrote an article for one of the magazines I wrote for. Then other magazines asked me to write articles, because I had the advantage of being able to write about something I'd actually seen, whereas up till now, everything people had written had been speculation.

Pretty much every computer magazone you read, there was an article by me about viruses in general, and the Brain virus in particular. My recommendation was to be choosy about who you got floppy disks from, and where you got software.

I also got interviewed by various journalists, and there's where the idea came from that the first virus seen in the UK was a university in the Midlands. I wasn't actually trying to mislead, I really am that bad at geography.

As an interesting side effect, I developed a slight skin problem on my hands. Ladysolly said it was because I'd greatly inceased the frequency with which I washed them. I switched to a gentler cleanser, and the pproblem cleared up. But why was I washing my hands more often? I'm not so stupid as to think I could catch anything from a computer! But maybe something deeper inside my head didn't know that.

Brain virus spreads only on 360 kb floppy disks. I heard rumours about a version that also infects hard drives, but if you know anything about how the FAT works, you'd know that this wouldn't be a minor variant, it would need a major reprogram. No such variant was ever seen.

Since I now had an actual virus, I could test what antivirus software there was around. And it turned out to be pretty useless. One program was designed to  check programs that you ran (Brain was a boot sector), one program checksummed your files looking for changes (Brain isn't a file!) and one program was specific to Brain, it claimed that it would detect it even if you were infected, but when I tried it out, it didn't. I think the reason why all the antivirus software at that time was so awful, is that the people who wrote it, hadn't actually seen a virus.

Back in 1987, a lot of PCs were twin-floppy, which would be a good environment for Brain. But it would also spread on computers with a hard disk - if you left an infected floppy in drive A when you powered up, the boot would fail, you'd take out the floppy and continue the boot from the hard drive, but the virus had gone memory resident by then, and would infect any write-enabled diskette that you put into drive A.

I think Brain was always very rare. I only encountered a few people who got it. When I say very rare, I mean *very* rare. It wasn't until we saw Stoned (which will infect a hard drive) and the memory-resident file viruses that we began to see them in any number.

Today, viruses are pretty much irrelevant. They can't spread on floppy disks, because when was the last time you even saw a floppy disk? Most computers today don't even have a floppy disk drive. And the other means of spreading, via shared executables, doesn't happen either, because you can't just copy an exe file and expect it to run. It has to go through a whole install procedure. Today, the threat is trojans. I get dozens of these sent to me per week via email, and "antivirus" programs mostly don't flag them. The other means of travel is when you access a web site that has malvertising.

30 years ago, when I saw Brain, I predicted that there would be lots more viruses, and I was right! By six months after I saw Brain, I'd seen six viruses, and I realised that what was needed was some sort of kit of tools for dealing with them. I called mine, "Dr Solomon's Antivirus Toolkit".

And that's how it started.

Sunday, 17 January 2016

Duh!

One of my servers wasn't sending mail. This server is supposed to report by email on its usage each day; it's also supposed to send alerts on various bad situations, should they develop. But it wasn't sending email. Why?

This wasn't an urgent situation, so rather than dive in and try everything possible, I mulled it over. Today, I came up with a possible answer, and when I checked it, I was right.

Sendmail wasn't running. That's why it wasn't sending mail. Duh.

I dimly remember shutting it down temporarily, and I don't remember restarting it. So I restarted it, and I was instantly rewarded with a flood of 100 emails that it's been keeping in its out-tray for the last couple of weeks.

Duh!


Winter wonderland

We've had nealy half an inch of snow, it's magical! The land is blanketed in a veil of white.

So we've canceled our projected trip to London, because A) it might be the Wrong Sort of Snow and B) they're closing parts of London for the light show, so between that and the snow, it's likely to be chaotic there.



Saturday, 16 January 2016

A tempting link

From: apoirier.conseil@gmail.com
Subject: give back


Unable to show full message.
You can view it by clicking here

drsolly error code: 8723 (Sat Jan 16 12:15:38 ART 2016)


And that's all there is. The link goes to

3dd440c998f2fe3b3c00.3343367.33ecb07.imsgview.xyz

and I haven't gone there to look, but I'd guess it's a page that's supposed to do something bad.

So what else could you do but click on it?

Daffodils

I've been seeing daffodils flowering by the roadside for a few weeks now, but today, the daffs in my back garden have started to flower. Only two so far, but the others aren't far behind.

This is so early! It's almost as if things are warmer than they used to be.

Friday, 15 January 2016

Stuff happens

I was woken too early, after a late night playing Civilization, with a report of a funny noise coming from the infrastructure. I dressed and had a gander.

The firewall I use is a Cisco Pix 506E, which is a totally obsolete piece of kit, but it gives up to 100 mbps of throughput, and since I have a 2 mbps line, it isn't going to feel the strain.

The noise was coming from that, and it was a gentle rurrr-rurrr-rurrr, which I diagnosed as a cooling fan getting close to the end of its life.

The Pix 506E has an Intel Celeron running at 300 mhz, which shows you how old it is. And that needs a heat sink, and the heat sink needs a fan. Actually, I'm not at all sure that such a low speed chip needs a fan, but it can't hurt.

Problem is, if I take the Pix out of service to look inside, all my comms are down. Solution is, another Pix 506E. So I configured my spare Pix to do the duty of the failing one, put it in place, took out the faulty one, opened it up and sure enough, the fan wasn't spinning as merrily as it should. This calls for a 50 cm fan, 1 cm thick. I rummaged through my box of fans, and found a 60 cm fan, 1 cm thick. A bit of surgery on the plastic plug, soon had it in place, and it spun nicely. I reassembled the Pix, and put it back, and everything worked fine.

Then I had a look at the spare, or at least I tried to. I couldn't plug the power into the box! The Pix 506E comes with a little power brick and a peculiar 8-way connector, and I just could not get the connector into the socket. I opened up the Pix, had a look, and saw that the socket had come adrift, and was half an centimeter inside the box. When I pushed it out, I was able to connect, and then I made sure that the configuration was exactly as it should be. I labelled it as "Spare Pix for data shed" and put it in a box.

In that box, I have another Pix 506E. That's intended to be a spare for the pix at Cheltenham, and it's configured and labelled as such. So if my Cheltenham Pix fails (Pixes dont fail, in my experience, but see below) all I have to do it a two hour drive, swap a box, and I'm good.

So Pixes don't fail, huh? Well, the fan does. Fans are a weak link on many computers. They are a cheap mechanical device, and they have a mean time between failure of maybe 20 years. But if you have 100 of them spinning happily, you won't need to wait 20 years until the first one fails.

The other part that fails, is the power supply, and I can tell that this is common by the number of Pixes on Ebay sold without power supply. I'm guessing that what's happening here, is a much loved Pix 506E power supply fails, and the sad owner decides to replace it with an ASA, Cisco's current firewall. The ASA is much more powerful than the Pix, and if you compare prices when new, much better value for money. Except that you can buy a Pix 506E on Ebay for typically £20, and you can roll your own power supply out of any used PC power supply. Here's how, part 1 and part 2.

In the middle of all this, I decided that one of those Pixes wasn't going to work (the one I couldn't plug the power in to), so I went on to Ebay and bid for a Pix, including power supply, for £20. I'll probably get it, but I'm not bothered if I don't, because I do actually have two spare Pixes.
 

$951,000,000

From: GUINNESS FORTUNE <guinness.fortune@yandex.com>
Reply-To: guinness.fortune@gmail.com
Subject: GUINNESS FORTUNE

Congratulation!!! Your email address has won You Sum of $951,000,000 in the Guinness Fortune
Spain.Your email address was chosen
based on an Internet random selection exercise,You have been chosen by the Guinness Fortune,to
receive $951,000,000
(nine hundred and fifty one thousand united state dollars)on the Guinness Fortune Award.

However,your email was attached to ticket number ( 83GP-79 ) and drew the lucky number (
411367023132049 ) Batch N:: GN-250/2015
Ref N:: AGF/017-BAR-15/ESP,

Contact your claims agent with the following email: (fortune.guinness@gmail.com)


Sincerely ,

Brr. Luis Jose.


No, that's 951 million dollars, nearly a billion. It's the best offer I've had this week.

Thursday, 14 January 2016

An email to the New Stateman magazine

To: sbrasher@newstatesman.co.uk
Subject: Ad blocking.

You might be wondering why many people are blocking the adverts on your web site.

Here's why.

https://plus.google.com/+GrahamCluleyCom/posts/ZvcwW68SFMG

And then it happened again, also to Forbes:

http://www.networkworld.com/article/3021113/security/forbes-malware-ad-blocker-advertisements.html


It isn't only Forbes, of course, I give them as an aexample of a very respectable journal. This keeps happening, again and again.

http://blog.drsolly.com/2016/01/more-on-malware.html

Until you find a way to avoid serving up malvertising, people like me, and the people who listen to me, aren't willing
to take the risk.

You need to guarantee that if I visit your web site, I don't get my computer taken over by an advert that didn't actually come from you, but the effect on me is no less for that.

Wednesday, 13 January 2016

Hartley Heart Attack, part 9

It was quite cold today, so I bundled up with lots of layers, wore furry gloves, and tackled the HHA again.

The HHA has been growing, and that meant that I had a couple of routes to run, one long, one short. I was also hoping to clear up my one remaining DNF, which had been a DNF because the cache was missing when I went for it. Unfortunately, that one was in the middle of nowhere.

The highlight of the day was this:


Let's just look more closely at that tree ...


I didn't make use of it.

At the start of the day, I went through a track covered in deep water, so I got a wet foot. Again. At least it wasn't squelchy-wet this time.

40 caches found today, no DNFs.

Memories of a hacker

This is another one from International Wool. We got a new computer, an HP 3000. It was very different from the PDP 11/44 we had before; with the PDP you submitted a pack of punched cards, and eventually you got back a printout. Or you didn't.

With the HP3000, we all had monitors on our desks, and we could use the computer interactively. It was at that point that I discovered the Colossal Cave Adventure, which was probably the first ever adventure game in the world. I played it a lot. I mean A LOT. Other people did too, and we compared how far we'd got into the cave, it was a lot of fun. Meanwhile, I was also reading all the manuals for the computer, learning what else I could do with it. At the time, I was an economist, or at least that's what my job description said, but actually I was a programmer, always have been, using the computer to do the work that I was supposed to be doing on an ancient desk calculator.

So all this playing of Colossal Cave came to the attention of The Powers That Be, and a fatwa came down, forbidding the game. Fair enough, this was a working computer, it was work time, we shouldn't have been playing games. So naturally, I obeyed the fatwa ...NOT. Fatwas glance off me like water off a duck's back. But this fatwa was enforced by using the computer security system to make the game unavailable.

Oh well. So instead of playing the game, I explored the computer, using the knowledge that I'd gotten from the manuals. And one day, I found a file that looked interesting. I hexdumped it, and found inside the system manager's password. It was "rubella". I checked it out, and it worked. Well, that's interesting. What can I do with that, I wondered.

Here's what you can do. You can tell the HP3000 to make a list of all the authorised users, and their passwords (and that was very bad, that should *NOT* be possible, it should have stored a one-way hash of the passwords, not the passwords themselves). And I printed that out on the line printer.

I left it on the desk of Bill, the senior manager in charge of the computer department for him to find the next morning.

I got in early next day, sat at my desk looking innocent, and waited for the explosion. When it came, it was satisfyingly huge. Even better, Bill came straight to my office. "Was it you?" he demanded. I grinned. It was indeed.

I don't think he was angry. Well, maybe a bit. But I hadn't done any damage, I'd just demonstrated a monstrous hole in our computer security, and I explained to him how I'd done it, so he could fix it.

But what I didn't tell him, was that while I was logged on as system manager, I gave myself access to Colossal Cave.

You can access it here or on your Android phone here.

Memories of a hacker.

This memory comes from when I was working at International Wool. I was curious (I'm a devil for curiosity) about how everyone else's salaries compared to mine. But the culture was, you don't tell people your salary.

Each month, we'd get a wages slip, in a sealed envelope. It was printed, inside the envelope, but some printer, and there was a dazzle pattern on the outside so you couldn't see what was inside.

One month, I examined mine carefully, and I saw that if you tore the envelope open and looked at the inside, you could see my salary.

You can probably guess the rest.

I stayed a bit late one evening, on the day that the salary slips came round. I harvested all the ones I could find (which was pretty much all of them) from the wastepaper bins in everyone's office. I tore them open, looked in side, and made myself a handy database of what everyone was paid. THis meant that for a long time to come, I knew whether my boss was telling the truth or not when he told me how well I was being paid.


Tuesday, 12 January 2016

The importance of backups

Daughter.1 can't find her laptop. She's searched the whole flat, a few times, and it's missing. She can only suspect that a guest might have walked off with it. It's a problem, for several reasons.

Your browser remembers where you've been, unless you tell it to forget. It also remembers your passwords, a handy feature. So now she has to go onto everything she uses, and change her passwords. I also suggested that she check her bank and card statements carefully for a while, just in case.

She hasn't lost any data, because she didn't keep data on it. But it shows you the importance of backups - it isn't only the possibility of a disk crashing, it's also the possibility of losing your computer, either by forgetfulness, or theft.

She asked me if it could be traced. No, it can't.

Here's how things work. When your computer contacts your ISP to get onto the internet, your ISP gives it a temporary loan of an IP address, which looks like 123.65.231.12. The ISP has a record of the date and time that this happened, and it knows which customer it's dealing with.

So if the user of that computer were to download a copyrighted file, and if the aggrieved company got a court order and knew when the download happened and from where, it could request that the ISP look up its logs to fnid out which customer had the IP address that downloaded the file at the time it happened, and so they can take legal action against the downloader, because now they have the customer's name and address.

A defence sometimes used, is that it wasn't the customer, it was someone else using that computer. That might, or might not, work.

The police can go the same route, and find out the name and address of the customer who is suspected of doing something illegal online.

Unfortunately, that doesn't help daughter.1, because if her computer was stolen and is being used, it won't be using her ISP, because the computer doesn't have the password for the ISP, the router does that.

... later ...

She found the laptop!

The Pix crashed

I use three Pix firewalls. One at my colocation, one here, and another one to give additional security to servers such as my mail server. The third one just crashed.

I found out because I had 500 emails, which triggered my email monitor that tells me when there's more than 100. Each email was six emails from my monitoring server, telling me that it couldn't contact six of my servers.

I went down to the Data Shed. First I looked at the servers that were out of contact. They were all running fine. Then I  looked at the switch, that was OK. And then I looked at the Pix firewall, and it had no lights. Power off. And then I saw it was switching itself on and off, on and off. So I did the first thing one does in this situation, because it's easy to do and it sometimes fixes the problem. I powered it off, waited five seconds, then powered it on. It came back on, and stayed on, and the six servers are back online.

This power cycle cure is embedded in anyone who deals with computers a lot, so much so that it's the first thing we try when *anything* stops working. It's worked for my car, several times, for example.

Pixes never crash, or so I believed. This belief is based on more than ten years of using them. Now the belief is "Pixes hardly ever crash". I'll be putting that Pix onto my power controller, so that if need be, I can cycle the power remotely.

 ... later ...

It crashed again. So I've put another Pix in it's place.

I opened it up, I can't see anything wrong. I'll leave it running for a while, to see if this is a long-term problem, or can I use it as a spare?


Monday, 11 January 2016

More on malware

I've been quiet on malware for a while. That isn't because I've stopped receiving it - it's because I also want to talk about other stuff. But today, I got a large number of "invoices". They were all doc files.


SHA256 DE9DCC1777A3B26124442739B6E8FB30E7E142660697F14C09809B4FCAFC0C59

Jotti: Eset, Kaspersky and Quickheal flagged them, 18 other products didn't.
Metascan: All 43 products passed it as clean
VirusTotal: Arcabit, Mcafee and Trend flagged it. 51 products passed it as clean. My upload was the first time Virustotal had seen it.
Payload Security: Malicious. That's based on them sending it to VirusTotal.

So it's the same story. Malware is emailed out, probably in large numbers, nearly all products don't spot it (and even those that do, there's some doubt) and your only defence is that you're not silly enough to open a doc file from a supplier you know.

And, of course, suppliers continue to send out invoices as doc files, acclimatising people to the trick.



In related news, Forbes has been malvertising again. I like that word! It mean, malware delivered to your browser via advertising.

What happens is, advertising is sold my middlemen agencies; they take your money to display your ad, and they pay money to various web sites to display your ad. So you only have to deal with one agency, instead of dozens of web sites. And a web site only has to deal with one agency, instead of dozens of advertisers. A good idea, obviously.

Enter the Bad Person. Bad Person pays the ad agency to display their ad, but the ad includes malware that will install somethnig bad on the user's computer. Obviously, the ad agencies should be checking the ads they accept carefully. Obviously, they don't. Until they do, the only prudent course for users is to install ad blockers.

Forbes has form, they've done this before , last September. But this time they went one better. Forbes doesn't like ad blocking, because it hits their revenue. So they recently started blocking access to their site if you use an ad blocker.

So Forbes ask you to disable your ad blocker. But if you do, you get hit by malware.

Something has to change. I just went to forbes.com, and I was greeted by "Hi again. Looks like you’re still using an ad blocker. Please turn it off in order to continue into Forbes’ ad-light experience".

No chance. I'll forgo the Forbes' ad-light experience, in the interest of not having my computer taken over.

But whose fault is this? The ad agency, or Forbes? Well, of course, it's the fault of the Bad Person who put up the malware on the ad agency's site. But I think that the contract between an ad agency and the web sites who pay them, has to include a guarantee, with penalties, that there's no malware in the ads. Because you can't ask the Bad Person to take on that job!

More on Network Solutions

So this morning I phoned them about their problem, their problem being that I can't access their web site, which means I can't make changes to my domain name details.

The techie suggested that I clear my browser's cache. I did that, and it made no difference, because why should it? Then he suggested I use a different browser, so I used Chrome, and that made no difference, because why should it.

So then he said I was the only person reporting a problem, so it must be with my ISP. So I told him that a bunch of other people were reporting the same problem on their Facebook page.

He told me that the problem would be resolved within two business days. And we parted.

I got a call soon after from them, asking me to do a traceroute, and email it to them, which I did. I also did a ping test (my own idea) which indicates 90% packet loss. That means there's severe congestion somewhere on the route between them and me, and fully explains the reason I'm having problems accessing their web site. The traceroute fails at f5defense-ic-308524-ffm-b1.c.telia.net. Telia is a Swedish telecoms company, I don't know why my access to them is going via Sweden, my guess is that they have a server in Europe that I'm directed to.

I also tried pinging from Cheltenham. That gives me zero packet loss; if I tried to access 
Network Solutions from there, it would work fine. Unfortunately, I can't use a browser from there.

The call from  Network Solutions ended with me again being assured that they'd have this fixed within two days. I asked if they could email me when it is fixed, and he said he would.

If I were currently in the position of needing to make changes to my domain name details, I'd be USC without a paddle.

Trying to log on to Network Solutions.

Their web site is running in treacle. It's so slow, it takes several minutes just for the home page to come up. And when I try to log in to manage my account, it sits there doing nothing for a long time, and eventually times out.

I don't know what their problem is. I've gone to Facebook to get info (see, there *is* a use for Facebook!) and it seems I'm not the only one to have experienced this problem.

One of the services that they offer is web hosting. If I'd been thinking of using that service (I wasn't) I'd be looking elsewhere.

What Network Solutions need right now, is a solution for their network.

On a related note, I got some spam from a UK company offering hosting. Obviously I'm not likely to sign up with anyone who thinks that spam is a good idea, but I did email them to find out about costs, so I could compare with what I'm currently paying.

They are quite a lot cheaper.

For 100 mbit, 13 IP addresses, half a rack and 4 amps, I'm quoted £565 per month. I'll put that to my current suppliers, see what they say.

Saturday, 9 January 2016

A cunning scam?

I got an email, apparently from Network Solutions.

Date: 9 Jan 2016 00:16:05 -0500
From: donotreply@networksolutions.com
To: drsolly@xxxxxxx.com
Subject: Alert: Keep Your Domain Active


[logo.png] [www.networksolutions.com]
Action Required
Dear xxxx xxxxxxx,

This is in regards to the following account:

Email Address: drsolly@xxxxx.com
Phone Number: xxxxxxxxxxxxxxxxx
Address: xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx


Network Solutions is now required by ICANN (the regulating body for domain registrations) to have all domain owners confirm their email
address contact information or their domains will be deactivated. If your domains are deactivated you will still own the domains but you
will not be able to have live websites until you verify your contact information. If you wish to view the list of domains subject to
verification, please login to ACCOUNT MANAGER [www.networksolutions.com]

To ensure your domains remain active, please click the CONFIRM button below to confirm the email address we have for you is accurate.





                                     Confirm Email Address [whoisaccuracy-portal.networksolutions.com]


If you have any questions, feel free to contact customer service at 1-888-642-0209.


I've xxxed out the details, but they are all accurate. Giving your correct name and address is is good sign that the email is not a scam. But in this case, is it actually a scam? I suspect this because I looked at the email header, which tells me that the email actually came from jax4mhco02.myregisteredsite.com. So how did the scammers get the name and address? Because that's in the whois database, which is publicly accessable.

So I went to Network Solutions facebook page.

Here's what they say: "This is a legitimate email requesting confirmation of contact information in your account. This email is mandated by ICANN."

 So maybe it isn't a scam.

Why the hell do so many organisations (HMRC, for example) send out emails that have all the hallmarks of a scam? Don't they realise that they're setting people up to fall for real scams?



Friday, 8 January 2016

Caught on camera

I was coming back from a wedding. I hadn't imbibed, of course. But I was driving an unfamiiar route, in the dark, late at night.

I strayed into a bus lane. The bus line was about 10 yards long. I was caught on camera.

So I'm donating £65 to Harrow Council. I hope they spend it wisely.

VAT Moss nightmare

VAT Moss is a nightmare.

It came about because the powers that be, realised that when a sale is made electronically from the UK to, for example, to Germany, the VAT rate paid should be the VAT rate in Germany, and it should be paid to the German tax authorities. Hurrah for the European Community. And HMRC, realising that people like me having to make 27 different payments in 27 different ways to 27 different tax authorities was not going to fly, set up the VAT Mini One Stop Shop, so I can do all 27 in one place, which is nice.

Then they didn't bother to tell anyone about the change. I found out two days before it was due to come in, because ladysolly's bridge club had to suspend accepting payments for a while because of this. I was lucky - I just needed to modify my software to collect the data, and writing software is one of the few things I'm able to do.

The first time I tried to do the VAT Moss dance, I made several mistakes, which we eventually untangled. The next two times I got it pretty much right.

I collect the necessary data each time I do a billing, making note of the country and amount. Then, at the end of each quarter, I do a spreadsheet to work out  the total for each of the 27 EC countries (but not the UK, that's done in a different payment).

I've recently synchronised the two; my VAT used to be September to November while VAT Moss was October to December, but I've now set it up so that both are done in October to December. Of course, that means that for one time only, I pay the VAT Moss for three months, and the VAT for four. Horrible calculations.

So this week, I wanted to do my tax payments for the fourth quarter, 2015. I won't say I enjoy paying tax, but I can see the necessity; how else could MP's expenses be funded?

The first step was to calculate the VAT Moss for October-December, and I did that, and then I tried to feed that to the HMRC web site. In the past, I've done this country by country, but now I see that I can just submit a spreadsheet in a format they give me, so I can do it as one lump. Hurrah! It means I can automate the creation of that spreadsheet; less work for me, more work for the computer. Ideal. So I tried that, and ran into two problems.

The first was that it wanted all numbers to be to two decimal places (e.g. £123.45). But because we're working with non-round numbers, mine are to umpteen decimal places. I know how to fix that, it's called rounding. But then I ran into a worse problem.

For each line of the VAT submission, you have to give your VAT number.  I tried that, and it rejected it. Then I noticed that it says "Enter that VAT Registration Number of your fixed establishment (including country prefix eg DE for Germany). So instead of giving my VAT reg code as 123456789 I gave it as UK123456789. Rejected. 123456789UK. Rejected. So I googled a bit, and found that I should use GB, so I tried GB123456789. Rejected. 123456789GB. Rejected.

Wouldn't it have been nice if they gave an example of what I should be putting here? And if the example was for *this* country, not for Germany?

So at that point, I gave up and did it the old way, by hand. It only takes an hour or so, but that's an hour I could have been spending playing Civilization. I've asked the nice HMRC people how this should be done, but they've referred the question (without actually knowing what the question is yet) to their tech people, and I'm not expecting an answer any time soon.

So after tediously entering in all the data, it gave me a total to pay, which sounded right, and that's when I blundered.

I paid it.

By debit card.

What I'd forgotten, is that although you can pay pretty much any HMRC tax bill by credit or debit card, you can't pay VAT Moss that way. There's a whole complicated procedure for paying, and it doesn't involve plastic cards.

So the payment went through, but heaven knows where it went. I had to call HMRC to ask nicely to have it back, because I've paid them a bunch of money and they must be wondering what it's for.

Now I have to do my VAT (as distinct from VAT Moss) and pay that.

I'm hoping that it will be easier for me next quarter, because they'll be synchronised, and because I've put a note in my procedure "DO NOT PAY BY CARD".

And maybe I can even do it by uploading a spreadsheet?


Wednesday, 6 January 2016

Hartley Heart attack, part 8

Another trip today, and this was to do a part of this series that has just appeared!

After the first few caches, I was faced with this.


I was on the grassy side of the ditch, the cache was on the other side, the water was about 15 inches deep. I leaned over to hold on to the tree, intending to hold it while I bent over for the cache. But the ground was steeply sloped, muddy and slippery, and I started a slow but inevitable slide downwards. I realised that wetness was inevitable, so I quite deliberately put my left foot into the ditch; the water was sufficiently deep to overtop my boot, so I spent the rest of the day with a wet foot.

I got the cache, though!

And then I had a lot of trouble getting out of the ditch, but eventually I managed to get myself up without getting my right foot wet.

Several caches later on, I had a similar problem. But this time it was easy, because I didn't care about putting my left foot into the water - it couldn't get any wetter!

Later on in the series, I crossed with another cacher - we didn't realise it when we crossed, but he stopped at the cache that I'd just done, and I stopped at the cache he'd just done. So we waved to each other from a distance.

I finished the day by picking up two caches that I'd DNFed before, for a total of 48 finds.

Tuesday, 5 January 2016

The TomTom died

Just as I was going round the M25 on the way home from Center Parcs, the TomTom XXL made a noise it's never made before, and died.

I took it apart to see if there was anything I could do, but inside is a PCB, and a component of that has failed, I think, and I can't repair those. So I've ordered a replacement, second hand, just inder £45. Ebay! The battery probably isn't very strong, but I don't mind that. And now I know how to dismantle a TomTom to put in a new battery.

Meanwhile, I have TomTom software on a Mio Digiwalker, so I'll use that until the replacement arrives.

The TomTom XXL is better, because it can handle 60,000 POIs; the Mio can only manage half that.  I carry the Mio in the car as part of standard caching equipment; it's a backup to the TomTom.

I'm planning to go caching tomorrow  for the first time since before Christmas.

Monday, 4 January 2016

Filtered out!

At about 7pm yesterday, one of my servers in Cheltenham stopped responding to the once-per-minute check. And I was unable to log on to that server from here. But if I logged in to another server in Cheltenham, I could log on to that server.

Which means that it's some sort of routing problem. But what caused it, and how to deal with it?

First, I rebooted the server, because that's easy to do, and often fixes problems. That didn't help. Then I tried to set up the route table, but that didn't help either. Eventually, I had the idea of doing a traceroute to that server, which bombed out before it reached my firewall, which meant that the problem was outside my control. But I had a customer-facing server not working!

I reported the problem to the hosting company, so they could start to work on it. Meanwhile, I used the NAT (network address translation) capability of the firewall, to move the external IP address to something else. That also meant that I had to edit the DNS files and re-propagate them, so that external computers would be able to find the server on the new address.

And that all worked.

Today, I found out what the problem was. The hosting company monitors the packet volume to each IP address, and that IP address was getting a lot. I mean, a *LOT*. They interpreted that as some sort of attack, and killed the routing to the server. Actually, it wasn't an attack, it was a rather greedy customer.

I can understand why they killed the routing, but (as I told them today) they should also have notified me that they'd done so. That would have saved me some hours in working out what the problem was, and I'd guess that some people running servers wouldn't be able to work it out at all.

Everything is OK now. And they've told me that they'll disable the route-killer for my IP addresses.

Weight report

After Christmas, four days at Center Parcs and the previous 12 months, I@m back at 16 stone 11 pounds.

The diet starts today!

Friday, 1 January 2016

A good deed

I was at a new years party dinner, and I was chatting with the guy next to me. He told me he worked for a haulage company, I told him I messed about with computers. So he guessed that my family relied on me for tech support. I explained that I mostly mess around with servers, and can't help people running iPads or Windows. So he told me that they have a server where he works, and he's responsible for it.

So I said, as I always do, and with a smile, "I hope you make backups".

No, he didn't. "But we'll be moving soon, I'll see about it after then."

So I explained to him that the third most stressful thing you can do to a server is switch it off, the second most stressful is to switch it on, and the most stressful is to move it. And he's about to do all three.

He looked thoughtful, got out his iPhone and sent an email, there and then. He's going to do a backup before he moves the server.

So I've done my bit for computer security for 2016!