1) The rundig script as-is will always index from scratch. So if you
reindex and haven't changed any files, it will still index all
files. That's because it calls htdig with the -i flag.
You might want to take a look at the rundig-32.sh script:
2) If you wish to only reindex certain URLs, you may want to consider the
-m flag to htdig (in 3.1.6 or 3.2.0b3 or later):
Williams Students Online
On Thu, 27 Jun 2002, Sankaranarayanan M.G. wrote:
> Dear All,
> Greetings. I am currently using htdig-3.2.0b420020505 version.
> In my /var/www/html/docstosearch/ directory, I have many sub folders (
> for example sf1, sf2, sf3 etc....) wherein I have placed my files
> which are to be indexed. Also from the rundig script I have removed
> the -i option used for htdig, so that the database is not created from
> scratch every time.
> In the htdig.conf file for the start_url I have given
> With this configuration, I am able to index all the files in
> docstosearch directory and its subdirectories. The problem started
> when I add or remove files or directories from docstosearch directory
> or any of its sub-directories.
> 1) It is always taking the same time for the rundig script... even
> when there are no changes in the docstosearch directory or any of its
> 2) Suppose I change only one file in one of the sub-directories, How
> do I re-index that directory alone. I don't want to waste time trying
> to index the other directories. I tried changing the htdig.conf file
> to point to that directory in the start_url and executed the
> runscript. Most of the times its throwing some errors from WordDB
> saying PANIC or CDB__ or WordCompress line number 827... Can someone
> kindly help me out of this..
> Sankaranarayanan M.G.