not sure, but I think it's just to flush the cached index...
Brian Demers wrote:
Why does the web app need to be restarted? are the index files on the
classpath or something? It seem like this is a hack?
On 8/13/07, srampl <[EMAIL PROTECTED]> wrote:
Hi,
Thanks for this valuable information,
I need contionus and latest results in nutch,i have old crawl data "CrawlA"
and latest crawl data "crawlB" . u told after the merge use this command
"touch $tomcat_dir/WEB-INF/web.xml" function in script, that's fine. but at
the time of merging between crawlA & crawlB, we can't able to give the
result. it disply empty page, then only i am asking , how to solve this
problem.
Thanks
Tomislav Poljak wrote:
Hi,
if it helps:
you don't need to restart tomcat to load index changes, it is enough to
restart an individual web application (without restarting the Tomcat
service) by touching the application's web.xml file. This is faster than
restarting tomcat. Add:
touch $tomcat_dir/WEB-INF/web.xml
to the end of your script and this will "tell Tomcat to reload index".
Tomislav
On Fri, 2007-08-10 at 02:50 -0700, srampl wrote:
Hi,
Thanks for ur reply,
I have did that step, but at the time of merging between old index(i.e
curently tomcat running index) and new index or after merging, it not
give
the search result until the tomcat is restart. So we can't able to
produce
contionous search result at the the time of merging and after merging
until
we restart the tomcat.
Plz give idea abt this
Thanks in advance,
Harmesh, V2solutions wrote:
Hi,
The Crawl can be updated by again performing the generate, fetch &
update
cycle step by step
Generate will create new segment and after fetching the documents, the
update cycle will update it
with the older crawl.
Ratnesh,V2Solutions India wrote:
Hi,
Ricardo, Greetings of the day,
We are using nutch and our corporate application is ready but due to
client demand regarding getting refresh crawl data, we are planning to
update our crawldb instead of re-crawling .
So do u have any solution that how to update crawldb which already
have
been crawled and storing some useful information.
It's nice if I find any solutions from u or any of ur colleagues.
With Thanks & Regards,
Ratnesh,V2Solutions India
