Hi Millebii,
Thanks for your valuable inputs.
As per our requirements we need to run multiple nutch instances with each instance pointing to their own conf dir and crawlDB.
crawl -urlfilter.txt is different in both conf folder. But in our case both nutch instances picking same conf dir instead of picking their own conf dir.
So both crawlDB have same data. [Actually we have separate filter in both conf, so data in both crawlDB should be different]
Have you tried such scenario? Since
Thanks
-Pravin
-----Original Message-----
From: MilleBii [mailto:millebii@gmail.com]
Sent: Tuesday, March 09, 2010 1:06 PM
To: nutch-user@lucene.apache.org
Subject: Re: Two Nutch parallel crawl with two conf folder.
Yes it should work, I personnaly run some tests crawl on the same
hardware, even on the same nutch directory thus I share the conf
directory.
But If you don't want that I would use two nutch directory and of
course two different crawl directory because with hadoop they will
end-up on the same hdfs: (assuming you run in distribued or pseudo)
2010/3/9, Pravin Karne <pravin_karne@persistent.co.in>:
Can we share Hadoop cluster between two nutch instance.
So there will be two nutch instance and they will point to same Hadoop
cluster.
This way I am able to share my hardware bandwidth. I know that Hadoop in
distributed mode serializes jobs.
But I will not affect my flow. I just want to share my hardware resource.
I tried with two nutch setup , but somehow second instance overriding the
first one's configuration.
Any pointers ?????
Thanks
-Pravin
-----Original Message-----
From: MilleBii [mailto:millebii@gmail.com]
Sent: Monday, March 08, 2010 8:02 PM
To: nutch-user@lucene.apache.org
Subject: Re: Two Nutch parallel crawl with two conf folder.
How parallel is parallel in your case ?
Don't forget Hadoop in distributed mode will serialize your jobs anyhow.
For the rest why don't you create two Nutch directories and run things
totally independently
2010/3/8, Pravin Karne <pravin_karne@persistent.co.in>:
Hi guys any pointer on following.
Your help will highly appreciated .
Thanks
-Pravin
-----Original Message-----
From: Pravin Karne
Sent: Friday, March 05, 2010 12:57 PM
To: nutch-user@lucene.apache.org
Subject: Two Nutch parallel crawl with two conf folder.
Hi,
I want to do two Nutch parallel crawl with two conf folder.
I am using crawl command to do this. I have two separate conf folders,
all
files from conf are same except crawl-urlfilter.txt . In this file we
have
different filters(domain filters).
e.g . 1 st conf have -
+.^http://([a-z0-9]*\.)*abc.com/
2nd conf have -
+.^http://([a-z0-9]*\.)*xyz.com/
I am starting two crawl with above configuration and on separate
console.(one followed by other)
I am using following crawl commands -
bin/nutch --nutch_conf_dir=/home/conf1 crawl urls -dir test1 -depth
1
bin/nutch --nutch_conf_dir=/home/conf2 crawl urls -dir test2 -depth
1
[Note: We have modified nutch.sh for '--nutch_conf_dir']
urls file have following entries-
http://www.abc.com
http://www.xyz.com
http://www.pqr.com
Expected Result:
CrawlDB test1 should contains abc.com's data and CrawlDB test2
should
contains xyz.com's data.
Actual Results:
url filter of first run is overridden by url filter of second run.
So Both CrawlDB have xyz.com's data.
Please provide pointer regarding this.
Thanks in advance.
-Pravin
DISCLAIMER
==========
This e-mail may contain privileged and confidential information which is
the
property of Persistent Systems Ltd. It is intended only for the use of
the
individual or entity to which it is addressed. If you are not the
intended
recipient, you are not authorized to read, retain, copy, print,
distribute
or use this message. If you have received this communication in error,
please notify the sender and delete all copies of this message.
Persistent
Systems Ltd. does not accept any liability for virus infected mails.
--
-MilleBii-
DISCLAIMER
==========
This e-mail may contain privileged and confidential information which is the
property of Persistent Systems Ltd. It is intended only for the use of the
individual or entity to which it is addressed. If you are not the intended
recipient, you are not authorized to read, retain, copy, print, distribute
or use this message. If you have received this communication in error,
please notify the sender and delete all copies of this message. Persistent
Systems Ltd. does not accept any liability for virus infected mails.
--
-MilleBii-
DISCLAIMER
==========
This e-mail may contain privileged and confidential information which is the property of Persistent Systems Ltd. It is intended only for the use of the individual or entity to which it is addressed. If you are not the intended recipient, you are not authorized to read, retain, copy, print, distribute or use this message. If you have received this communication in error, please notify the sender and delete all copies of this message. Persistent Systems Ltd. does not accept any liability for virus infected mails.
