One solution would be for someone to do a clean checkout of the site into the /www/avalon.apache.org directory.
Sounds like a job for someone young enough to do this with no fear - and smart enough not to screw it up. The excludes me on two counts. Leo - what are you up this afternoon!
well, up to my neck into Analysis of Linear Systems, actually. Something ado with some guys named Cauchy, Lipschitz and Schwarz. Barring spelling mistakes.
Just did:
ssh [EMAIL PROTECTED] cd /www/avalon.apache.org mkdir ~/sitebackup mv *download.* ~/sitebackup ~/bin/avalon-site-up.sh
(ie remove the files that caused conflicts then ran an update)
the log of that goes to http://avalon.apache.org/cvsuplog.txt, which no longer reports any conflict. However, that hasn't exactly improved things: download.cgi now redirects the user to www.apache.org/dist/. Therefore, I did a
ssh [EMAIL PROTECTED] cd /www/avalon.apache.org cp ~/sitebackup/* .
(ie copy back over those files that caused conflicts). I then tried mkdir downloadtest cd downloadtest cp -r ../CVS . cvs up
but that won't work because of the way the download script works (the mirror script being referred to has an absolute path programmed to /www/avalon.apache.org/download.html and I don't have karma for changing that).
Someone or something broke the download script(s) and committed faulty / missing versions into CVS. The CVS log for avalon-site says:
---
mcconnell 2003/09/27 16:30:17
Removed: site bindownload.html download.html srcdownload.html Log: Remove generated content (sill experimenting to get this sorted out) --- and some other stuff. Steve, what've you been up to?
AIUI, what is needed is for someone to commit a valid download.html file into avalon-site/site/, then
ssh cvs.apache.org umask 002 cd /www/avalon.apache.org mv download.html download.bak cvs up *download*
gotta run!
- LSD
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
