:: 31-Dec-2002 02:22 GMT (Tuesday) ::

I tried to build live stats for the key-rate data.
They use the same data source as dctievent and the graphics are build
using rrdtool.

* http://www1.distributed.net/~pstadt/rc5-72/
* http://www1.distributed.net/~pstadt/ogr/

They should be updated every 20 minutes.
Any suggestions concerning the colors are welcome ;-)

:: 29-Dec-2002 20:30 GMT (Sunday) ::

The keymaster is going to be resuming distribution of some of the old
OGR-24 stubs that never got enough redundancy on our network. So if
you have clients that are configured to run OGR you’ll notice some of
these stubs eventually coming your way.

:: 29-Dec-2002 10:43 GMT (Sunday) ::

A few more clients are now available on the pre-release page:

* dnetc478-netbsd-sparc
* dnetc478-aix-powerandppc
* dnetc478b-ps2linux-mipsel
* dnetc478-aix4.3-power
* dnetc478-linux-s390x
* dnetc478-linux-s390
* dnetc478-linux-s390-z
* dnetc478-solaris-sparc

And a new build of the personal proxy for a few platforms are also
available on the pre-release page (hopefully this will be the last
build number increment before general availability of the proxy):

* proxyper333-win32-x86
* proxyper333-linux-x86
* proxyper333-linux-x86-glibc21
* proxyper333-freebsd-x86

Remember that the files available in the pre-release directory are
intended for limited use by people who are willing to help us test
them before they are announced for general availability. People who
test them must be willing to watch their machines carefully after
installing them (to notice/report anomalous behavior).

:: 17-Dec-2002 21:17 GMT (Tuesday) ::

Just a quick note to let you know we’re doing stuff:

Joel and I have been working on the OGR verification routines and are
pretty close to the final testing phase. At this point we’ve run into
some problems that require things to be handled manually, which is
obviously a Bad Thing.

Once we get things sorted out, we will have one of the biggest hurdles
to finishing an OGR contest out of the way.

:: 17-Dec-2002 09:36 GMT (Tuesday) ::

Several additional clients have graduated from release-candidate
status to formal release availability on the download page:

* dnetc478-riscos-arm
* dnetc478-irix6.5-mips
* dnetc478-netbsd-arm32
* dnetc478-linux-sparc
* dnetc478a-linux-arm-elf

A few new release candidates have been added to the pre-release page:

* dnetc478-dos-x86
* dnetc478-linux-x86-elf-glibc20
* dnetc478-nto2-arm
* dnetc478-nto2-x86
* dnetc478-ps2linux-mipsel
* dnetc478-qnx4-x86
* dnetc478-win16-x86

Also the proxyinfo monitoring pages have been updated to support a
more modern backend infrastructure, but we have tried to keep the
presentation approximately the same. You can see the new RC5-72
proxyinfo (updated at 5 minute intervals) page at:

* http://n0cgi.distributed.net/rc5-proxyinfo.html
* http://n0cgi.distributed.net/ogr-proxyinfo.html

The random, useless factoids at the top of the proxyinfo page are
indeed outdated and need to be replaced with more interesting quotes
(I’ve merely translated the old 1998 quotes/scripts into modern XSL).
If you have cute ideas for *positive*, general interest quotes, I’m
willing to accept suggestions at bovine@distributed.net. (Preferably
try to keep them project neutral so that they’ll work for RC5-72 keys
or OGR nodes.) For the curious, saved copies of the two input files
that generate the above proxyinfo page are:

* http://www1.distributed.net/~bovine/proxyinfo2.xsl
* http://www1.distributed.net/~bovine/proxyinfo.xml (stale input data)

:: 16-Dec-2002 20:56 GMT (Monday) ::

While I haven’t been helping Decibel much with recovering
from the problems in stats, I think I have an acceptable
excuse…

http://www.toomuchblue.com/gallery/EmmaBirth

Moo!

:: 15-Dec-2002 19:49 GMT (Sunday) ::

Blower is busy catching up on OGR and should be done in a few hours. It’s not
doing RC5 right now because we’re going to rewrite history a bit. :)

Several people have suggested that since very, very few people were able to
submit blocks on Dec. 2 that we should move the Dec. 2 work to make it look
like it happened on Dec. 3. This is exactly what we’ve done, and tomorrow
at around 1600 UTC (10 AM CST) I’ll start re-processing all of RC5-72 from
the beginning. I’m holding off until then so that anyone who wants to watch
themselves climb in the ranks from day 1 can watch stats as the updates happen.

:: 15-Dec-2002 16:39 GMT (Sunday) ::

Good news! Sybase is up and running again. The problem turned out to be a
library issue of some kind… it didn’t like Linux 7.2, but downgrading back to
6.2 solved everything.

I’m taking this opportunity to grab a quick backup; once that’s done I’ll turn
the website back on.

:: 12-Dec-2002 21:19 GMT (Thursday) ::

Just wanted to let people know that we’re indeed having some traffic
capacity issues to the keyservers, so it may take your client several
connection attempts to successfully fetch or flush buffers. Just let
the client retry by itself (don’t manually keep retrying) and it
should eventually connect.

Additionally, we’d like to again request that everyone go around and
check all of their machines on which previous versions of the dnetc
client had been installed and consider upgrading them. In particular,
if you are still running a v2.8010 client, uninstall it or upgrade!!

That version has a known bug that will cause it to rapidly reconnect
to our keyservers if the projects are closed and it is unable to get
work. This is causing a fair amount of donated bandwidth to be wasted
and is causing quite a bit of concern. We’re in the process of
collecting the IP addresses of the most significant offenders and
contacting them individually when possible, but this is a tedious and
partially manual process.

:: 11-Dec-2002 23:48 GMT (Wednesday) ::

Just wanted to keep people updated on stats. In a nutshell, nothing happened
today… bwilson wasn’t around (his wife is due any day, so for all we know
he’s playing with his new baby right now :) ), and Matt and I were busy all
day. Tonight didn’t work out either.

I’m hoping that we can get this taken care of tomorrow evening, but that may
not happen either. Worst case we should be able to get something done this
weekend. (Of course every time we announce a date we usually blow right past it
so I may have just jinx-ed the whole thing…)