:: 05-Feb-1999 14:57 (Friday) ::

Well, I spent most of yesterday massaging the data in the master RC5-64
table on statsbox-ii. When I re-installed sybase onto the new drives,
I forgot to change the default sort order. As a result, all the
recent participants I added as I’ve been testing the logfile import
were added improperly. Normal stats operation is to ignore casing on
email addresses, but since I forgot to change the sort order we were
getting duplicate email addresses in the participants table. Like:

nugget@Slacker.com
nugget@slacker.com
Nugget@slacker.com

being three entries instead of just one. Easy fix, yeah, but it took a
while to hand-correct the participants table and re-link to the master
table.

statsbox-i is pretty much a wash at this point. I may bang on it today
some, but I’m really leaning towards just focusing on the new machine
to get it into production as soon as possible.

Today’s goals involve tidying up the rough edges on the daily-process
script and fiddling some with apache/php.

:: 04-Feb-1999 20:56 (Thursday) ::

The following clients have been updated

rc5des435-dos-x86-cli.zip
rc5des435-win16-x86-cli.zip
rc5des435-win32-x86-cli.zip
rc5des435-aix4-power.tgz

Enjoy!

:: 03-Feb-1999 22:07 (Wednesday) ::

Hi everyone !

I updated the public source code in http://www.distributed.net/cores/.
Here’s a short “what’s new” :
– Bruce magic MMX DES bitslicer ;
– a new Alpha DES core from Christoph Dworzak ;
– updates in most of the other cores ;
– egcs+alpha patch in logstuff.cpp ;
– some fixes in the ./configure script ;
– lots of portability fixes in the C++ code, borrowed from
the full v434 client source.

Alas I don’t have time right now to test this source on a Win32
machine. So if you have patches for this platform, send them and I
will upload a new set of sources.

:: 03-Feb-1999 19:31 (Wednesday) ::

Well, there’s good news and bad news…

The bad news is that statsbox-i is trying really, really hard to die. I’m
looking at some database corruption that’s preventing a dbcc checkdb from
properly completing on the stats database. I’ve been banging on it since
last night, but I’m still not certain exactly where the problem is. This
is why stats did not update last night and is also why stats are not
currently online. I’ll keep you posted as the situation develops.

The good news is, there’s a VERY temporary bit of uselessness running on
the new statsbox. I’ve got solid data through 17-Jan and the emrankings
code is completed. So, you can fiddle with the 17-Jan emails rankings
if you want. This url will *not* be around for very long, so if it
404’s on you, sorry, I’ve just pulled it offline. Also, it’s quite likely
to not work at all if you look at it while I’m fiddling with stuff.

Still, here’s an unimpressive little teaser:

http://tally.distributed.net/demo.php3?low=1&limit=20

:: 02-Feb-1999 03:11 (Tuesday) ::

Well, yesterday and today were pretty much a wash. I spent the day breaking
down and re-building the database devices on statsbox-ii. Sunday morning
I started getting i/o errors on the IBM 9gb drive, so I decided to re-run
the surface analysis and badblock and all that. Turned up 25 bad sectors
that hadn’t been flagged before. *shrug*

So, it’s been frustrating the past few days, looking at:
Feb 1 16:02:53 tally kernel: scsi0: MEDIUM ERROR on channel 0, id 1, lun 0, CDB: Read (10) 00 00 32 94 57 00 00 02 00
Feb 1 16:02:53 tally kernel: Current error sd08:12: sense key Medium Error
Feb 1 16:02:53 tally kernel: Additional sense indicates Unrecovered read error – recommend reassignment
Feb 1 16:02:53 tally kernel: scsidisk I/O error: dev 08:12, sector 3041670

That repair wrapped up late this afternoon and I’m back to the business of
grooming statsbox-ii for production. I did get some neat php stuff coded
while I waited for the various surface analyses to run.

Also, here’s a snippet to make you drool:

CPU states: 190.7% user, 9.2% system, 108.4% nice, 0.0% idle
Mem: 516596K av, 513496K used, 3100K free, 229032K shrd, 85852K buff
Swap: 128484K av, 756K used, 127728K free 296772K cached

This box is going to fly.

Anyway, if you’ve mailed me at all during the past four days, odds are I haven’t
even looked at the email yet. I’ve got ~200 “are stats down?!?!” emails and
a dozen more “You’re a loon, IDE doesn’t suck” messages. Frankly, I’ve been
more concerned with keeping statsbox running (well, walking) and putting
time into statsbox-ii. I’ll take a stab at sifting through my inbox tomorrow,
so if you’ve been waiting for a reply I apologize.