Thanks for the explanation Jeff. Our release instructions have a section about updating the wiki (just a whitespace update will do) to force the main page to upload. I never understood that until now.
I know we can get RSYNCSITE.sh to run nearly immediately after yours, but the process sounds a bit error prone to me. Is there an easy way to trigger a space export say in a cron job. How useful is having the Nabble RSS feed on the wiki? I never remember to look there (maybe because it's often out of date) so I'm inclined to just link to Nabble and avoid further complicating the cron jobs. Does anyone else feel strongly in favor or against having Nabble embedded in the wiki? -mike On Thu, Jan 15, 2009 at 5:44 PM, Jeff Turner <[email protected]> wrote: > Hi, > > It appears to work as follows: > > - Confluence generates a static HTML dump of pages whenever they are > edited. The static dump is visible at eg. > http://cwiki.apache.org/openjpa/mailing-lists.html. For pages with RSS > content, I think the static content would be from the time of the last > edit. > - At 55 minutes past the hour every hour, the exported files are > rsynced to people.apache.org:/x1/www/confluence-exports, of which > ~jefft/public_html/confluence is a symlink (please update your script). > - RSYNCSITE.sh then rsyncs to the live website directory. > > If you have dynamic RSS content that you want to display on a static > website, the best we could do is trigger a space export just before the > hourly rsync, and then ask mprudhom to ensure RSYNCSITE.sh triggers ASAP > after mine, and then the RSS will only be out of date by up to an hour. > > Or alternatively just link to nabble, and let users see live content > there. :) > > > Cheers, > Jeff > > On Thu, Jan 15, 2009 at 08:23:56AM -0800, Craig L Russell wrote: > > Hi Jeff, > > > > I wonder if there's a better way to keep the mail page current? Or > > should we just abandon the notion? Having a stale page isn't a good > > way to promote our site. > > > > Thanks, > > > > Craig > > > > Begin forwarded message: > > > > >From: Michael Dick <[email protected]> > > >Date: January 15, 2009 7:52:14 AM PST > > >To: [email protected] > > >Subject: Re: Stale web page > > >Reply-To: [email protected] > > > > > >The way the main site is populated from confluence has always been a > > >bit of > > >a mystery to me, so I did a little digging. > > > > > >As I understand it user jefft (Jeff Turner) on people.apache.org has a > > >script which pulls content from confluence. The output is saved in > > >his home > > >directory. We in turn have a script which synchronizes our website > > >from his > > >copy ( /www/openjpa.apache.org/RSYNCSITE.sh on people.apache.org). > > > > > >I'm not sure which of these scripts are running, or how often they > > >rin. I > > >took a quick look at Jeff's directory and it looks like the mailing > > >list > > >page hasn't been updated since summer. > > > > > >If anyone wants to follow up with Jeff and see if we can get the > > >website > > >refreshed more often they're welcome to do so. I'll put it on my > > >list of > > >todos but I might not get to it for a while. > > > > > >-mike > > > > > >On Wed, Jan 14, 2009 at 4:04 PM, Pinaki Poddar <[email protected]> > > >wrote: > > > > > >> > > >>Hi Craig, > > >>This page suffers from refresh problem. Michael Dick and I made few > > >>attempts with advertised 'embedded Nabble' feature without much > > >>luck [1]. > > >> > > >>[1] http://issues.apache.org/jira/browse/OPENJPA-649 > > >> > > >>-- > > >>View this message in context: > > >>http://n2.nabble.com/Stale-web-page-tp2153480p2159394.html > > >>Sent from the OpenJPA Developers mailing list archive at Nabble.com. > > >> > > >> > > > > Craig L Russell > > Architect, Sun Java Enterprise System http://db.apache.org/jdo > > 408 276-5638 mailto:[email protected] > > P.S. A good JDO? O, Gasp! > > >
