|
| Greg's diary |
| October 2005 |
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Saturday, 1 October 2005 | Echunga | |
| Top of page | ||
| next day | ||
| last day |
Somehow managed to spend far too much time with work matters this morning, something I had sworn not to do. I've barely been on board two weeks, and already I seem to be flat out on various projects.
Today I had at least a partial excuse: I discovered that Monday is Labour Day and thus a public holiday, about which I should have told people. So it looks as if I'll have to do at least a little work over the weekend.
Spent a lot of the day working on the slides for my paper on beer, and made relatively good progress. Gradually my documentation framework is falling into place.
| Sunday, 2 October 2005 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Spring is definitely here today, a wonderful, mild, sunny day, so of course we went out riding. Diane Saunders came along, and we had a terrible time! All the horses were full of the joys of spring, and they behaved themselves terribly, even Darah. Didn't stay out long.
Apart from that, pottered around and did some odds and ends, but nothing of great importance.
| Monday, 3 October 2005 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Today was a public holiday, Labour Day. I always find it amusing that Labour Day is not a work day. Somehow didn't manage to keep away from IRC completely, and was there when a message came from the president of Picturetrail telling us that things had crashed several times “today” (Sunday in San Francisco). sigh.
Spent most of the day working on my slides for AUUG 2005, and made a lot of progress with the groff macros that I have been gradually writing over the last 10 years. One day they might be worth something.
| Tuesday, 4 October 2005 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Many fires to fight today. Picturetrail is still a concern, but hopefully Geert is making progress on that front. Also had a show-stopper bug to look at, and fortunately spoke to Sanja on IRC, who identified the problem and suggested a fix in only a couple of minutes. Finally got my testing environment working and confirmed that yes, indeed, it fixed the bug. That's very reassuring, though I'm pretty sure that I wouldn't have found it by myself.
Chris Yeardley along in the evening. Ate, drank and were merry.
| Wednesday, 5 October 2005 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Somehow lost momentum today. I have been neglecting too many uninteresting (i.e. administrative) issues lately, and today I finally got round to looking at them. Still, the big show-stopper is as good as dead, so today I got another one.
“Birthday” dinner in the evening: mine was last week, Diane Saunders in a couple of weeks' time, and for some reason we had thought that Chris' birthday was also round now. That proved to be wrong (it's in December), but why let a little thing like that get in the way of a good dinner?
| Thursday, 6 October 2005 | Echunga | Images for 6 October 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Another day with little to show. Keeping up with mail has never been easy, and the 30% or so increase in mail since I joined MySQL doesn't help. It's worse than it seems, since much of the mail is more important than the average stuff I'm used to receiving. In addition, I had some other things to attend to in the afternoon, so I had less time to look at anything. And finally, my user ID came to bite me. Emacs said:
Steal the lock on /users/glehey/4.1/sql/sql_yacc.yy:1.394 from grog? (yes or no)
This was on an internal machine, where my ID is glehey. No idea where it got the grog from, but I couldn't get it to go away. Maybe it failed because it couldn't send mail to grog, or maybe it failed because it thought it was talking to an SCCS system, but the result was that I couldn't edit the file with Emacs. Trying vi should have worked better, once you get past the pain of programs that think they know what kind of terminal you're using:
|
|
||||||||
Some of the text is yellow on beige; it's barely recognizable. You can get rid of that with the vi command :syntax off, of course, but I first had to find out how to do that. Even then, for some reason, I was getting incorrect indentation. What a pain!
| Friday, 7 October 2005 | Echunga | Images for 7 October 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Yet another day spent mainly reading and doing administrative things! Yes, I did test the fix to our show-stopper bug (it worked), but most of it was handling mail in some form or another. It's not mail itself, of course, but the work that the mail brought with it. Having the show-stopper behind me is a relief.
Grinding malt again in the evening. Using my big, expensive Bosch pneumatic drill had proved to be less than effective last month, so this time I decided to try it with my cheap battery-driven screwdriver. That worked quite well, it didn't overheat, and I was able to establish that a freshly charged 14.4 V, 1.2 Ah battery is good for about 2 kg malt. Of course they didn't quite manage for the 5.5 kg I wanted to grind, but it did much better than I had expected.
| Saturday, 8 October 2005 | Echunga | Images for 8 October 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Brew day today, and had barely time to look at mail. That was worthwhile, though: during the night Oracle announced that they had bought out Innobase Oy, the maker of InnoDB, which is used only in MySQL. Spent a lot of time discussing what that meant, but at the very least it's clear that it's a recognition of MySQL.
| Sunday, 9 October 2005 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
More work on the presentations for AUUG 2005 today, and finally finished the slides. A good thing too, but at least had a bit of a chance to relax.
The beer slides posed an unexpected problem: apart from the format I have on the web site, I also have a script that creates a bullet-by-bullet PDF file, so that I can display the individual bullets in a bullet list by paging down. Unfortunately, it also includes multiple copies of the the BSD daemon logo, which is almost exactly 1 MB in size. The results are some enormous PostScript files, which fortunately shrink again when converted to PDF:
=== grog@wantadilla (/dev/ttyp6) /home/Book/Beer 385 -> ls -l paper/pdf paper/ps
paper/pdf:
total 7
-rw-r--r-- 1 grog lemis 4521506 Oct 9 15:39 mslides.pdf
-rw-r--r-- 1 grog lemis 490977 Oct 2 11:35 paper.pdf
-rw-r--r-- 1 grog lemis 1705753 Oct 9 15:47 slides.pdf
paper/ps:
total 298
-rw-r--r-- 1 grog lemis 211784437 Oct 9 15:39 mslides.ps
-rw-r--r-- 2 grog lemis 16150025 Oct 2 11:32 paper.ps
-rw-r--r-- 1 grog lemis 83640538 Oct 9 15:47 slides.ps
The problem is that I postprocess my PostScript files with some Emacs scripts, and mslides.ps overflowed the maximum Emacs buffer size, so I had to rewrite the scripts in C as a filter, which was surprisingly simple.
| Monday, 10 October 2005 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
My user ID has been changed! Now I'm grog on MySQL machines as well. Spent some time tidying up some straggling stuff there, and also a lot on the phone. The events of the weekend are keeping us busy. Looks like I'll be travelling in the near future, and so some of my priorities are changing.
In the afternoon, finally got round to ordering a new laptop. I had tried earlier, but the Dell web site was such a pain that I had given up. Today it wasn't much better. It continually gave me error messages (“Document contains no data”), and of course the security settings were invalid:
Of course, they won't deliver to a post office, (and in the past they refused to deliver to my house, only 40 km from the middle of Adelaide), and they insisted on a “Shipping Address” and a suburb. But I'm used to that. In less than two hours I had ordered the laptop, of course to be called back by Ramya (I think that's the name) some time later to tell me that she couldn't locate the “suburb”. Sigh. Less than 1% of Australian land mass has suburbs.
Out to Belair nursery to buy some more plants; that takes a surprising amount of time. On the way back, a phone call from Ramya of Dell wanting to know where “no suburb” was. They refuse to deliver to PO boxes for “security reasons”, but as we've seen in the past, their couriers also refuse to deliver to us (on the way to the Post Office!), and dump the stuff (without a signature) at the General Store instead. We're in for more fun, I fear.
| Tuesday, 11 October 2005 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Still no order confirmation from Dell. On investigation, I discovered it had arrived, but had been classified as spam:
Spam detection software, running on the system "wantadilla.lemis.com", has
identified this incoming email as possible spam. The original message
has been attached to this so you can view it (if it isn't spam) or label
similar future email. If you have any questions, see
the administrator of that system for details.
Content preview: Thank you for choosing Dell, This email is to let you
know we have received your order. Because we let you decide what you
want, we take extra care to make sure we can deliver! [...]
Content analysis details: (3.3 points, 3.0 required)
pts rule name description
---- ---------------------- --------------------------------------------------
0.4 FROM_NO_LOWER From address has no lower-case characters
3.0 BAYES_95 BODY: Bayesian spam probability is 95 to 99%
[score: 0.9869]
-0.1 AWL AWL: From: address is in the auto white-list
The original message was not completely plain text, and may be unsafe to
open with some email clients; in particular, it may contain a virus,
or confirm that your address can receive spam. If you wish to view
it, it may be safer to save it to a file and open it with an editor.
Looking at the attachments, I discovered:
I 1 <no description> [text/plain, 8bit, us-ascii, 1.2K] A 2 original message before SpamAssassin [message/rfc822, 8bit, 81K] I 3 |-><no description> [multipa/mixed, 7bit, 80K] I 4 |-><no description> [text/plain, 7bit, utf-8, 2.5K] A 5 -->THANKU_QT_391_10540891_101005_152402.HTM [application/octet-stream, base64, us-ascii, 77K]
What a mess! Why do they attach HTML files as application/octet-stream? It's still surprising, though, that it had such an effect on they Bayesian analysis.
Apart from that, much administrative work. Looks like I'm going to have to miss the AUUG 2005 conference and go to Helsinki instead—and that after preparing a paper and a tutorial. We're living in interesting times.
| Wednesday, 12 October 2005 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Things don't stop! Got into the office this morning and discovered that I've got to be in Athens for a meeting from 9-11 November. That's four dates in Europe in 7 weeks, but with two lots of two weeks in between. Spent quite some time making flight arrangements. It's amazing how much time all this is taking up; got almost nothing else done today.
Dilip from Dell called about the shipping address. Connected me to Ramya again, who of course could still not find my “suburb”. Finally connected me to Kirin (?) Kumar, who promised that he would get a courier who would deliver here. If he does that, he'll be better than most Dell people.
| Thursday, 13 October 2005 | Echunga | Images for 13 October 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
At least there was no new change of direction this morning, but I can see that my email load is increasing:
On 13 October 2005 you received 2314 mail messages and sent 31 mail messages.
So is the graph of my mail queue:
The place to look at here is the bottom of the curves: that shows how many mail messages I have left over without time to look at them.
Round noon, Peter Hansteen of Bergen showed up, just in time for lunch before I left for a meeting of the ICT Council for South Australia. We now have a new executive director, Claire Baker, who looks like she is going to be pretty forceful.
Then to the LinuxSA meeting, where Yvonne brought along Peter to talk about their implementation of RFC 1149:
|
|||||||||||||||||||||||
He's in the process of writing up a document on the subject.
After that to the Oostende beer cafe. I summed it up in the review I left there (anonymously, it seems; I had no opportunity to leave my name):
Wonderful ambience, good beers, but the food could be more authentic. I had a confit de canard that had never been near the confit process. It tasted good, but it wasn't what was advertised.
Belgian food has the reputation of being so plentiful that you can't finish it. This was not the case here; it was definitely not enough, and I left the place still hungry. They should review their portions.
Back home, had a bite to eat and far too late to bed. I'm amazed that Peter seemed to hold out better than me.
| Friday, 14 October 2005 | Echunga | Images for 14 October 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
More rotation today; spent just about the whole day on the phone, mainly to travel agents. My schedule as of this morning had me leaving for Helsinki on 17 October and getting back here on 30 October, then leaving for Athens on 6 November and getting back on 16 November, then leaving for Frankfurt on 20 November and getting back on 8 December. By the end of the day I had modified that to leaving for Helsinki on 17 October and getting back here on 27 October, then leaving for Athens on 7 November and getting back on 7 December. Still means a month in Europe, which I've found some ideas about how to spend. It also means a ticket refund, which will be expensive, but not as expensive as actually making the flight.
Still no invoice from Dell. Checked the web site with the order number they gave me, and it didn't find it; it looked for all the world as if they had forgotten the order. On the phone things looked better: the order is being processed, but their broken web site doesn't know how to find it. It should arrive almost exactly when I get back from Helsinki. And the missing invoice? “Oh, we don't give an invoice until we deliver the goods”. The fact that they deducted the price from my credit card on Tuesday didn't make them feel any obligation to issue an invoice.
| Saturday, 15 October 2005 | Echunga | Images for 15 October 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Thank God for the weekend! It's been a pretty busy week, and it's good to wind down just a little. Off with Peter in the morning to show him around the countryside, then back for a barbecue. While preparing for that, got another phone call from Qantas about my flight changes, during which it eventuated that I might have trouble with my flight because the name I specified (Greg) did not exactly match the name in my passport (Gregory). In the olden days the booking would just read LEHEY/GMR, but now it has to be exact. The upshot was that they re-issued the ticket, with the interesting side effect that the whole thing ended up $140 cheaper. I asked James (the agent) to do it again, but wasn't able to.
At the barbecue, had a few new faces: David Newall was there, but with Harry, a friend from Indonesia. In addition Dan and Angela Carosone were over from Melbourne, and this time they stayed longer than they did last year. Jack Burton rounded off the list; it was his first time here. A larger proportion of old farts was present than normal, and somehow ended up talking about the good old days, dragging out bits of 30 year old computers:
|
|||||||||||||||||||||||
This one shows a 384 kB ECC memory board for a Tandem/16 (later called NonStop 1). More on today's photo page. Also dragged out a program listing (underneath the board in the photo) of the UNIVAC 494 drum handler (i.e. device driver) which I had modified on 15 December 1976: the FH 432 drums on the 494 were so fast for the day that two of them transferring in parallel would saturate the memory bandwidth of the machine and cause “late acknowledge” (i.e. overrun) errors, which were fatal. My fix (6 instructions; we were frugal in those days) made them retryable.
| Sunday, 16 October 2005 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Off into town to take Peter to the airport, somewhat confused by the fact that there's a new terminal; it wasn't clear if it was open yet, though the number of cars parked in front suggested that it was. Peter was flying Virgin Blue, and we at least confirmed that they're still in the old terminal.
Didn't do much in the afternoon; mainly pre-flight preparations, of which there seem to be more and more. It's at a time like this that the deficiencies of my system upgrade procedures become most obvious. Managed to completely delete my mail in the process, but fortunately the recovery was trivial.
| Monday, 17 October 2005 | Echunga –> Singapore –> Bangkok –> | Images for 17 October 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Off to the airport far too early this morning; with the current paranoia, I thought it prudent to really arrive the 90 minutes in advance that they ask for. That was probably just as well: discovered the rest of the mystery of the terminals that we had wondered about yesterday. The old International Terminal still sports the sign, and the new terminal (next to it) does not have anything. It seems, though, that they're breaking with tradition and using the new terminal both for international and domestic traffic, and currently only the international stuff is there. It's been open since Saturday, and lots of things don't work. The check-in staff were helpful, but obviously under some stress. One tried to print out a tax receipt for me and was unable: “Those incredibly intelligent people over there tell me that I won't be able to print anything until Wednesday”. The Qantas Club is also pretty rudimentary; no computers, and even the (sliding) door had to be opened manually.
The plane to Singapore was an A330 with a new digital entertainment system, including video on demand. Unfortunately, the user interface is terrible, and in addition the film I wanted to watch was so dark as to be almost completely unrecognizable. The staff tried hard, including a “repair” in Darwin which, as far as I could see, achieved nothing. About the only interesting thing was when they rebooted the surprisingly big box under the seat in front of me:
|
|||||||||||||||||||||||
Faded rapidly after Singapore, and from the stop in Bangkok (where we were allowed to stay on the plane) slept nearly all the rest of the way.
| Tuesday, 18 October 2005 | –> Helsinki | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Arrived in Helsinki far too early, as planned, and got through the airport surprisingly quickly, despite particularly inquisitive immigration officials. Monty was still in bed, and spent an hour talking with Anjuta and descaling their remarkably complicated coffee machine.
Gradually Monty and Jonas Orelund showed up, the latter here from Stockholm for a couple of days. Spent the day talking about database storage formats and other internals, notably ndb, on which Jonas is working. It's surprising how familiar all this stuff is.
Monty evidently believes in all-round people, and in the evening he gave us and Jani “Jamppa” Tolonen the task of assembling a Weber barbecue that he had just bought. Quite a bit different from the kind of barbecue that we have in Australia, and it's a good thing that Jonas has one himself, so he knew how to put it together.
| Wednesday, 19 October 2005 | Helsinki | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
On with our discussions today, at quite a pace. It's amazing how much I have learnt on so short a time. Jonas left in the evening, so maybe I'll have time to catch my breath tomorrow.
For some reason, eucla, the “old” laptop (Dell Inspiron 9100, with a Pentium IV HTT desktop processor) that I still have on loan from Rocksoft, has been crashing on me repeatedly. In the evening investigated and discovered that the cooling fins at the back were clogged with dust. Given the heat that the things generate, that's enough to explain the crashing. Fortunately the fans are easily removable, so spent some time cleaning the fins; hopefully that'll stop the crashes.
| Thursday, 20 October 2005 | Helsinki | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Up early this morning in the mistaken assumption that Monty was there before me; he didn't show for some time, and when I found him he was in the process of removing a titmouse pair from the lounge room.
Spent most of the day reading code; even well-documented sources are not easy to understand if you don't have an overall picture. It looks as if I'll soon have to resort to running the server under gdb to find out what it really does.
Still problems with eucla locking up. I wonder if it's related to disabled ACPI; the processor always runs at 100%, conceivably even when it means that it's overheating. On the other hand, this is the kernel I've been running for over a year, and it's only recently become a problem.
| Friday, 21 October 2005 | Helsinki | Images for 21 October 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Didn't sleep too well, and in the morning had difficulty getting my eyes open. More work on the documentation, somewhat hampered by the unusual environment (only a laptop instead of my normal multi-monitor environment). Still, I'm making progress.
Monty down round midday with some different things to do. Told him about the issue with only a single monitor, and showed him my home page with the photo of my desktop. He was sufficiently impressed to drop what he was doing and start setting up his machine as a multi-head system.
As usual, that wasn't as simple as it should have been. Monty uses SuSE 9.3, and for some reason X -configure failed consistently with multiply defined symbols. yast2 and friends didn't detect a second head on the card, and we gradually came to the conclusion that they were right, so looked for a second card.
That wasn't made any easier by the lack of an AGP port on the machine (brand new AMD64-based machine; interesting to note that the Linux amd64 kernels don't work too well on it either, so he's running an i386 kernel), and I didn't recognize the slot in which the existing card fitted. Still, there were a couple of PCI slots, and Monty thought he might be able to find a suitable card at a local shop. He wanted to go shopping anyway, so we all went together. Found no PCI cards, of course, but my surprise discovered that the card that was in the machine was a PCI Express card—I had always thought that they had a particularly small slot, but this was larger than a normal 32 bit PCI slot. There was a second PCI Express slot on the motherboard, so Monty bought another card, almost the same, and we installed that when we got back. yast2 didn't seem to know how to configure it, but I was able to set it up manually without too much difficulty. Unexpectedly, it worked first time:
|
|||||||||||||||||||||||
This photo was taken a couple of days later.
| Saturday, 22 October 2005 | Helsinki | Images for 22 October 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Finally got a bit more sleep, and up relatively late. Monty and family were going into Helsinki today, so arranged to meet up with some BSD people whom I know from IRC: Atte Peltomaki (“Koston”) and Jari Kirma. Took a while to find the, including getting lost some distance from where we had agreed to meet, but finally made it. First to Teerenpeli, where I had my first experience of Sahti, a beer made apparently by spontaneous fermentation, but with a clove aroma which leaves any Weizenbier behind. After that to Kaisla, who had a wide choice of beers, including Coopers. Tried a Finnish beer which confirmed Jari's explanation about why people in Finland tend to drink imported beers.
Back home by train and taxi; late to bed.
| Sunday, 23 October 2005 | Helsinki | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Into the office this morning to find that eucla had crashed—nothing new recently. But this time it wouldn't reboot: it asked for the name of the root file system, something that it should have found in /etc/fstab, and then fsck SIGSEGVd when it was run. Further investigation showed that at 1:01 this morning, the contents of /etc/fstab were replaced with the contents of /etc/crontab, presumably during the nightly cron job that checks in the RCS-controlled files. /etc/crontab, like the end of /etc/fstab, was filled with binary zeroes.
How do you recover from this situation? I was able to boot single-user, but fsck wouldn't run. Even when I told it the name of the partition to check, and despite what the man page says, it read /etc/fstab and SIGSEGVd. But without running fsck, how do you modify anything on the disk? The only other option is fsdb, and it only manipulates metadata, not data.
On further investigation, I discovered a file /etc/fstab.old, which at least contained the correct definition for the root file system. With a bit of playing around (chname command of fsdb) I was able to rename fstab to FSTAB and fstab.old to fstab, after which I could run fsck. I'm not sure what I would have done otherwise. Clearly there's a good reason to keep a backup fstab (with a name that's at least as long: fsdb manipulates the directory entries in place, so it needs the space.
I've always wondered why, in some cases, web browsers such as firefox won't restart after a crash, claiming that the “profile” is in use. Today it happened again. I knew that this information seems to be stored somewhere in the directory ~/.mozilla, but I couldn't remember where. find didn't show up anything much:
=== grog@eucla (/dev/ttyp2) ~/.mozilla 26 -> find . -type f -mmin -360
./firefox/profiles.ini
./firefox/oiq6n8jm.default/localstore.rdf
./firefox/oiq6n8jm.default/cookies.txt
./firefox/oiq6n8jm.default/history.dat
./firefox/oiq6n8jm.default/Cache/_CACHE_MAP_
./firefox/oiq6n8jm.default/Cache/CD194CD4d01
(many more of same)
Looking at the directory (in this case with emacs, sorted by date) was more useful:
/home/grog/.mozilla/firefox/oiq6n8jm.default: used 2 available 967 -rw-r--r-- 1 grog wheel 115312 Oct 23 11:34 history.dat drwxr-xr-x 2 grog wheel 512 Oct 23 11:33 Cache drwx------ 6 grog wheel 1024 Oct 23 11:22 . -rw-r--r-- 1 grog wheel 14671 Oct 23 11:22 cookies.txt -rw-r--r-- 1 grog wheel 7468 Oct 23 11:04 localstore.rdf drwxr-xr-x 2 grog wheel 512 Oct 23 11:03 Cache.Trash lrwxr-xr-x 1 grog wheel 19 Oct 23 11:03 lock -> 192.109.197.145:950 -rw-r--r-- 1 grog wheel 1740 Oct 21 10:14 downloads.rdf
That file lock looked suspicious, and from past recollection I know that removing would be enough to work around the problem. But why should it be a problem? 950 looked like a process, so:
=== grog@eucla (/dev/ttyp2) ~/.mozilla 27 -> ps p950
PID TT STAT TIME COMMAND
950 v0 S 0:02.37 /usr/local/bin/emacs -fn 6x13 -display :0 -geometry 110x70-53+0 -font 6x13
Coincidentally that was the same emacs that showed the previous display. After stopping it, firefox started with no problem. So the real issue is that firefox doesn't do adequate checking before refusing to start. The date of the symlink was earlier than the boot time, and the process wasn't another firefox.
| Monday, 24 October 2005 | Helsinki | Images for 24 October 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Pretty busy day today, and didn't get much out of the ordinary done. I'll be glad to get home and get some sleep.
| Tuesday, 25 October 2005 | Helsinki | Images for 25 October 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Another busy day. I managed to finish the work I had planned for this trip, but my original intention was to have it ready yesterday. Still, better now than at the end of the week. Late to bed.
| Wednesday, 26 October 2005 | Helsinki –> London –> | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Up in the middle of the night for an 8 am flight, to discover that about 5 cm snow had fallen overnight. Apart from having to step through slush at the airport, that didn't make much difference to me, but the plane to London was delayed by an hour while they cleaned the snow off it, something they had apparently not planned for:
|
|||||||||||||||||||||||
How I hate Heathrow airport! The transfer between terminal 1 and terminal 4 involved going through a security screening. That wouldn't have been notable except that they are obviously not adequately equipped:
|
|||||||||||||||||||||||
This was just the tip of the iceberg, the people waiting outside because the main hall was too full (I'd guess about 300 people). I suspect that I would have been in trouble if I had taken a photo in there. The delay must have been at least 30 minutes, maybe much more. I later heard from one person that he had repeatedly missed connections as a result. Again there was a fast (well, somewhat shorter delay) lane “by invitation only”. I had to ask twice before I was “invited” to take the lane—an additional advantage of having received a business class upgrade on the flight to Sydney. As a result, it only took about 10 minutes to get through.
The joy continued in the British Airways lounge. You'd think they had been paid to be bureaucratic. They require reading (“swiping”) the boarding pass on entry, but everybody was busy attending to other things, and when I walked past only showing the boarding pass, I was followed by an irate attendant who grabbed my card and marched off. Why do they have to be so unfriendly when, in the end, their system isn't working?
Off on the flight to Sydney, delayed again by an hour. More problems with the in-flight equipment: they have power for laptops, with the slight problem that it turns off after about 10 seconds. Qantas really need to work on their equipment.
| Thursday, 27 October 2005 | –> Singapore –> Sydney –> Adelaide | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Uneventful trip back to Australia. In Sydney it took nearly an hour to get through immigration (instantaneous), collect baggage (priority baggage doesn't seem to work on journeys where you change planes) and quarantine (which took longer than it should have done). Another trip over, nearly 35,000 km:
| From | To | km | ||
| ADL | SIN | 5410 | ||
| SIN | BKK | 1440 | ||
| bkk | hel | 7870 | ||
| HEL | LHR | 1850 | ||
| LHR | SIN | 10900 | ||
| SIN | SYD | 6290 | ||
| SYD | ADL | 1160 | ||
| Total: | 34920 |
Back home, lots of things awaiting me: a number of database books and my new Dell Inspiron 6000, complete with scratched cover (don't they have any quality control?). Turned the latter on and was presented with a screen asking me whether I accepted the EULA. The problem was that there was no way to enter “no”. Pressed n and was accepted anyway. This is nonsense.
Late to bed.
| Friday, 28 October 2005 | Echunga | Images for 28 October 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Slow day. Somehow I managed to sleep until nearly midday, and didn't do much for the rest of the day.
I've been away less than 2 weeks, but the garden has changed almost beyond recognition. Strangely, half a wattle tree died:
|
|||||||||||||||||||||||
This was a spontaneously planted tree that grew up a few years back. Unusually, it has two trunks, and in a very short space of time one of them seems to have died, while the other looks completely healthy. There's no evidence of any external influence. I'm baffled as to how it could happen, and how it could only affect one half of the tree.
More fun with Dell: you'd think that something as obvious as a scratch on the lid of a laptop would be easy to handle, but when I spoke to “Teh” (I wish these people wouldn't mumble their names), she was unable to help: first the technical team needed to “troubleshoot” the issue. But they weren't reachable. She promised a call back, something that Dell usually manages, but it didn't happen by evening.
| Saturday, 29 October 2005 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Another quiet day. With my new Dell toy, of course, I had to play around and install software. This is the first laptop where I have really completely wiped the Microsoft operating system: I have far too many already, and on this one I wanted to run FreeBSD, Linux and NetBSD, so even on a 60 GB space is at a premium.
Somehow installing Linux is always a pain. After my previous experience installing Ubuntu, decided this time to do the default installation, which worked just fine. If you like the defaults. It then downloaded something like 450 MB of updates, leaving me wondering if I shouldn't have first got a more recent CD.
But, try as I might, I can't make friends with GNOME. I suspect KDE would do no better, but I didn't have time to try. I've written a separate rant about the subject. Tried installing fvwm2, which brought its own series of problems: I couldn't find it anywhere with apt-get, apt-cache and friends. When I finally did find a .deb (at the site above), it wouldn't install because of broken dependencies. With a bit of messing around, found the right incantations to sort out the dependencies. But then apt-get came back and said “Ha ha, only kidding, here's your fvwm2”, thus causing another conflict. At least it ran after that, though I first had to work out how to start the server, which brought a number of error messages I've never seen before. I think I'll run three servers for a while: one each with KDE, GNOME and fvwm2. I suspect it'll be a while before I give up the last one.
| Sunday, 30 October 2005 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
More tidying up today, and also work on getting the new eucla up to speed (I've decided to keep the same name, though this may be sub-optimal). Installing software is easy (“I've done it a thousand times”), but keeping the configuration up to date is not. My upgrade system probably needs yet another conceptual overhaul.
UPSs have doubtless saved me multiple outages, but today it was the opposite: I have to return three UPSs to Rocksoft tomorrow, and to do so I had to shut down both wantadilla and echunga. And I'll have to do it again later in the week when we finally get the new UPS installed and integrated into the house power system.
| Monday, 31 October 2005 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Still trying to catch up, not helped by several mailing list threads from hell that kept me busy most of the morning.
In the afternoon to a planning meeting of the ICT Council for South Australia. Had quite a lively discussion; it looks as if we're rethinking our course once again.
Do you have a comment about something I have written? This is a diary, not a “blog”, and there is deliberately no provision for directly adding comments. It's also not a vehicle for third-party content. But I welcome feedback and try to reply to all messages I receive. See the diary overview for more details. If you do send me a message relating to something I have written, please indicate whether you'd prefer me not to mention your name. Otherwise I'll assume that it's OK to do so.
| Top of page | Previous month | Greg's home page | Today's diary entry | Next month | Greg's photos | Copyright information |
