Hi Shigeki, The hopcount mode has nothing to do with the situation as you describe it.
Remember that MCF is an incremental crawler. When it encounters a document that is no longer present, it removes it from the index. It tries to figure out if the missing document is a transient condition or not, of course, but if it decides that the document is permanently gone it will remove it. Karl On Wed, May 22, 2013 at 12:32 AM, Shigeki Kobayashi < [email protected]> wrote: > Hello, guys. > > > > I have a question about Web crawling with setting Hop count mode. > > In Hop count mode, you can choose “Keep unreachable documents forever”. > > With that setting, first crawling was fine. But when the web service that > is to be crawled is down, the second time crawling deletes all index. > > Doesn't this setting mean MCF does not delete index? “Keep unreachable > documents forever” does not actually keep index. Is this suppose be the > designated behavior? > > > > I use MCF1.1.1 running in MySQL 5.5. > > > > Regards, > > Shigeki >
