Bleak PCMCIA news...

In Linux, it turns out that kernel modules contain version information that is checked against the current kernel when a module is loaded. Quoting from the pcmcia HOWTO:

The intent was for this to make modules less version-dependent, because the checksums would only change if a kernel interface changed, and would generally stay the same across minor kernel updates. In practice, the checksums have turned out to be even more restrictive, because many kernel interfaces depend on compile-time kernel option settings. Also, the checksums turned out to be an excessively pessimistic judge of compatibility.

The practical upshot of this is that kernel modules are closely tied to both the kernel version, and the setting of many kernel configuration options. Generally, a set of modules compiled for one 2.0.31 kernel will not load against some other 2.0.31 kernel unless special care is taken to ensure that the two were built with similar configurations. This makes distribution of precompiled kernel modules a tricky business.

Since the version of the pcmcia software I removed is different from the version that's on my distribution CDs (which implies the compiled kernel is different from what is on my distribution), and since I could not find the version I removed on the net as a binary, and since even if I did, the HOWTO indicates that it wouldn't work anyway...

I guess I'll just copy my files onto floppy and reinstall. On the one hand, it was a bitch getting X to work on the thing; on the other, if I have the wireless card installed in the other slot, maybe it will be automagically configured when I reinstall...

In other Linux news, I am going to track down a copy of SmallLinux and try to install it on my IBM ThinkPad... maybe.

And oh, yes! There are those translations! (I finished the tables last night in time for me and Galina to watch Jane Fonda and Donald Sutherland in Klute, which I thought was a surprisingly good movie, if you can overlook the attempt to be too hip, and the fact that Fonda stars in it, both of which I decided to do last night.)

Cheers...

Good news, and other news...

Well, thanks to an almost unconscionable level of boilerplating, I'm on page 17 of 19 for the day. Once I finish these last 3 pages, I think I'll hang up my translation spurs for the day.

On the Linux front(s), I was able to get SmallLinux to boot on the ThinkPad, and even managed to run fdisk to partition the 1-GB hard drive into a 48-MB swap partition and an "everything else" Linux partition that's bootable. That's as far as I got. I'm not exactly sure, off the top of my head, how to go about formatting the partitions (or even if that's required); in any event, it would be nice to transfer the boot info to the hard drive, so that I don't have to keep a floppy in place (the ThinkPad only has room for a floppy drive or the CD-ROM drive, and I'd prefer to keep the CD-ROM drive in place, if only because a small piece of plastic that keeps the floppy in place has been misplaced).

The files I wanted to save from the crippled eSlate all fit onto 3 floppies, so that was not that big of a deal. When it came to reinstalling the Mandrake 7.1 (the only distro where I have a fighting chance to run X), the process seemed a lot clunkier today than it did way back when. Most certainly no support was installed for either PCMCIA card (my Linksys NP100 or the WaveLAN wireless card), at least any to the level where the hardware was recognized as a network card. The install process didn't ask me for network configuration parameters unless I specified a "custom" install, and when doing a "custom" install (one of three installs done today), the list of network drivers I'm asked to select from doesn't seem to have anything I care to specify.

So right now, the eSlate is sitting there with a "fresh" Mandrake 7.1 on it, incapable of running X (I remember having to finagle with it for a while before it'd cooperate), with no network support, etc., etc. I've initiated a download of the latest/greatest RedHat ISO disk images for their 7.2 version, but unless things speed up overnight, it'll take 18 hours to download each of the two images necessary to install RedHat 7.2. We'll see though; it's not as though I have a metered connection, have any pressing deadline, or anything like that.

At any rate, in the final tally, today's Linux-related activities are running below average, but that's not bad, considering how little attention I've devoted to them.

* * *Drew went up to Wolf Creek today and just got back a few minutes ago. He looks tired, but happy. Shannon didn't go with him, at least not directly. She stayed back a couple of hours and went with Huntur to a friend's house, after which she and baby caught up with Drew on the mountain.

Galina listened to some Robert Kiyosaki tapes while making the most delicious borshcht (the first such soup I've had in way too many a moon). Yum!

Back to the subject of translations: I am feeling a little bette about the reference documents I've been sent for the job. If anything, they show that some level of inconsistency has already crept into the translated documents (e.g., calling a "building" a "facility"), so I'm not going to grind myself unnecessarily to find the one-and-only, consistently used term.

Cheers...

The sole purpose of spam...

...is to let me know that my mail is working. Since nothing has arrived in any mailbox since yesterday afternoon, I started getting suspicious.

I looked at my 'maillog' file, and it is apparent that something happened between 14:33 and 14:37 yesterday, because at 14:36:53 (according to 'maillog'), 'postfix' starts misbehaving.

The log shows that postfix/pickup throws a fatal error, saying (I think) that the 'maildrop' directory is open, and exits. Over the next several seconds, 'postfix-script' warns that the 'active', 'bounce', 'corrupt', 'defer', 'deferred', 'incoming', 'private', 'public' and 'saved' scripts are not owned by 'postscript' (checking the eSlate shows that all these directories are owned by owner 'postfix' and group 'root' in the fresh-out-of-the-box config, and that the last character in the permissions string for 'maildrop' is 't', the so-called sticky-bit, and not 'T', which I am unfamiliar with [I thought the range was 'rwxXstugo'; maybe this is peculiar to RedHat?]).

There is also a log entry to the effect that '/var/spool/postfix/etc/passwd' and '/etc/passwd' differ from one another. This worries me, until I look and see only that some user accounts that were added after installing 'postfix' are not reflected in the copy of 'passwd' in the /var tree.

I begin to suspect that I have yet again been rooted, but that's a gut response, as what happened could, conceivably, have been caused by a failure to open or close a file properly. For someone to walk in and their cover tracks in the system logs but to leave all of 'postfix' lying out in the rain doesn't make much sense.

Cheers...

Postfix fixed...

...at least now I seem to be getting (and more important, sending) mail.

Also, in just over 11 hours of ftp connection, I've downloaded almost 340 MB of the first RedHat 7.2 install image. This puts me at just over 50% done, which - assuming things don't get any faster overnight - will complete the download sometime tomorrow around 10 am.

While I'm on the subject - and still conscious - the SmallLinux install is going fairly well on the ThinkPad. I've been consumed with the problem on 'onegin' (my main Linux box) for most of the evening (starting after Galina and I finished watching The Spanish Prisoner, which was pretty entertaining), so haven't done much there, but I've formatted the native Linux partition and the swap partition and have copied the files from the root floppy to the hard disk. I don't recall what the exact next steps are, but they seemed fairly straightforward when I reviewed them a few minutes ago.

I'm definitely starting to babble. Time to go to sleep...

Cheers...