[Nutch-user] Not able to crawl password protected pages using NUTCH 0.9

Stop trolling me...



Rout Biswajit-B16078 yazmış:

Hi Biswajit,

Could you please tell us how you have added the support for
authentication in Nutch 0.9? Nutch 0.9 can not do authentication
properly by default. The authentication feature is buggy in Nutch 0.9
which was fixed with this ticket:
https://issues.apache.org/jira/browse/NUTCH-559

The feature is documented here:
http://wiki.apache.org/nutch/HttpAuthenticationSchemes

The easiest way to use it is to check out the latest version of Nutch
and build it as it contains the authentication feature. If you want to
use it with Nutch 0.9, you have to download the latest patch present
in the ticket page and apply it to the source code and build it. You
might have to resolve some conflicts manually.

I would suggest that you do not send the mail same mail multiple
times. We have received the same mail from you 4 times. It takes
sometime for members to reply to a mail. :-)

Regards,
Susam Pal

On Mon, Sep 15, 2008 at 6:07 PM, Rout Biswajit-B16078
wrote:

Hi Susam,

In order to crawl password protected pages, I am using
nutch-2008-07-10_04-01-48.tar (I have download from
http://hudson.zones.apache.org/hudson/job/Nutch-trunk/ which contains your
patch for HttpAuthentication)

I have modified nutch-site.xml, httpclient-auth.xml.

Please find the attached zip file which contains
nutch-site.xml,httpclient-auth.xml.

Kindly provide me a solution for this.

Best regards,
Biswajit


Susam Pal wrote:
http://www.nabble.com/file/p19492846/Nutch.zip Nutch.zip

The logs show that it is fetching http://localhost:8080/ but you have
set credentials for 10.222.18.113:8080 which is never being fetched.
So, no authentication takes place.

Regards,
Susam Pal

On Mon, Sep 15, 2008 at 1:20 PM, biswajit_rout
wrote:

Hi Susam,

The ip 10.222.18.113 is nothing but the ip address of my machine(localhost).
Now also i changed http://localhost:8080/ to http://10.222.18.113:8080.
However no result, i mean to say still not able to crawl password protected
pages.

Kindly assist me to resolve this issue.

Thanks in advance...

Best regards,
Biswajit.




Susam Pal wrote:
http://www.nabble.com/file/p19507146/hadoop.log hadoop.log

I have also attached hadoop.log for your reference...


biswajit_rout wrote:

Hi Biswajit,

The authscope specifies which IP address or domain-name would the
credentials be used for. If you provide 10.222.18.113 in the
authscope, the credentials would not be used for localhost even though
both represent the same machine.

Please provide logs with DEBUG enabled.

Regards,
Susam Pal

On Tue, Sep 16, 2008 at 1:33 PM, biswajit_rout
wrote:

Hi Susam,

Thanks for your immediate response...
Herewith i am attaching the debug enabled log file(debugenabled_hadoop.log).
Kindly go through the file and let me know what is missing from my end...

Best regards,
Biswajit.


Susam Pal wrote:
http://www.nabble.com/file/p19510820/debugenabled_hadoop.log
debugenabled_hadoop.log

Hi Susam,

Please find the latest log file(latest.log), which shows different error.

2008-09-16 20:46:16,102 DEBUG httpclient.Http - url:
http://10.222.18.113:8080/robots.txt; status code: 404; bytes received: 985;
Content-Length: 985
2008-09-16 20:46:16,384 DEBUG httpclient.Http - url:
http://10.222.18.113:8080/dao/; status code: 200; bytes received: 1941;
Content-Length: 1941

Thanks in advance...

Best regards,
Biswajit.


biswajit_rout wrote:
http://www.nabble.com/file/p19514374/latest.log latest.log

The latest log shows that the page from the URL:
http://10.222.18.113:8080/dao/ has been fetched successfully.

Regards,
Susam Pal

On Tue, Sep 16, 2008 at 3:33 PM, biswajit_rout
wrote:

But still it is not crawling the password protected pages...

Regards,
Biswajit.


Susam Pal wrote:

The log file shows only one fetching line:

2008-09-16 20:46:15,321 INFO fetcher.Fetcher - fetching
http://10.222.18.113:8080/dao/

This has been fetched successfully. There is no other page being
fetched. Have you set up Nutch properly so that it can fetch all the
pages you need? If it tries to fetch a page but fails due to
authentication, then it is a problem with authentication.

In this case, it is not even attempting to fetch those pages. So, the
problem lies elsewhere. You need to first find out why it is fetching
only one page and not others.

Regards,
Susam Pal

On Tue, Sep 16, 2008 at 5:24 PM, biswajit_rout
wrote:

Hi,

There is nothing to crawl in the home page of
http://10.222.18.113:8080/dao/.

So this time i have crawled another site. I have successfully crawled all
the public pages but not able to crawl private pages.
I have attached a log file(new.log). Can you please check and let me know
what needs to be done from my end???

Best regards,
Biswajit.


Susam Pal wrote:
http://www.nabble.com/file/p19552519/new.txt new.txt

Hi Susam,

Please give a look into new.txt and suggest a solution for this. This time i
have crawled another site. I am able to crawl all the public pages but
password protected pages crawling is not happening...

Best regards,
Biswajit.


biswajit_rout wrote:

Hi Susam,

Please give a look into the attached file (new.txt) and suggest a solution
for this. This time i have crawled another site. I am able to crawl all the
public pages but password protected pages crawling is not happening...

Best regards,
Biswajit.


biswajit_rout wrote:

Hi Biswajit,

I don't find a single error caused due to authentication problem in
the 'new.txt' file you have attached in some mail before.. Most of
them are HTTP 404 or HTTP 302 errors, which means either the page is
not available or the page has been moved to another location, which
the crawler would try to fetch. There's nothing I can do to help you
in this matter. You have access to the network and you can analyze
better why this is happening. Please do not send the same mail
multiple time. As, I have told you before, it takes time for members
to respond as they do so only in their free time.

Regards,
Susam Pal

On Fri, Sep 19, 2008 at 5:38 AM, biswajit_rout
wrote:

Hi Susam,

I saw,
http://www.nabble.com/Problems-testing-Authentication-td13991771.html#a13995888

Where your patch is used for web server (Tomcat) manager authentication.

However my requirement is different…

I am trying to crawl sites which are password protected just like gmail.
That means I have to pass correct user name and password, after that I will
be able to see all the pages / modules. So for this, is there any different
configuration mechanism is there?

Could you please let me know???

Best regards,
Biswajit.



Susam Pal wrote:

Replies inline.

On Mon, Sep 22, 2008 at 1:40 PM, biswajit_rout
wrote:
My patch is used for Basic, Digest or NTLM authentication schemes
only. It works with Tomcat probably because it uses Basic
authentication to login to the manager. I hope you have read this:
http://wiki.apache.org/nutch/HttpAuthenticationSchemes.

POST based authentication is not done. You might want to read this:
http://wiki.apache.org/nutch/HttpPostAuthentication

Regards,
Susam Pal

Hi,

It's regarding the indexing which NUTCH is doing...
question is what is the maximum value of this indexing? Can we control this
value to some limit?

2008-09-24 13:13:23,390 INFO indexer.Indexer - Optimizing index.
2008-09-24 13:13:23,406 INFO indexer.Indexer - merging segments _ram_1e (1
docs) _ram_1f (1 docs) _ram_1g (1 docs) _ram_1h (1 docs) _ram_1i (1 docs)
_ram_1j (1 docs) _ram_1k (1 docs) _ram_1l (1 docs) _ram_1m (1 docs) _ram_1n
(1 docs) _ram_1o (1 docs) _ram_1p (1 docs) _ram_1q (1 docs) _ram_1r (1 docs)
_ram_1s (1 docs) _ram_1t (1 docs) _ram_1u (1 docs) _ram_1v (1 docs) _ram_1w
(1 docs) _ram_1x (1 docs) _ram_1y (1 docs) _ram_1z (1 docs) _ram_20 (1 docs)
_ram_21 (1 docs) _ram_22 (1 docs) into _1 (25 docs)
2008-09-24 13:13:23,437 INFO indexer.Indexer - merging segments _0 (50
docs) _1 (25 docs) into _2 (75 docs)
2008-09-24 13:13:24,216 INFO indexer.Indexer - Indexer: done
2008-09-24 13:13:24,216 INFO indexer.DeleteDuplicates - Dedup: starting
2008-09-24 13:13:24,232 INFO indexer.DeleteDuplicates - Dedup: adding
indexes in: crawl/indexes
2008-09-24 13:13:27,723 INFO indexer.DeleteDuplicates - Dedup: done
2008-09-24 13:13:27,723 INFO indexer.IndexMerger - merging indexes to:
crawl/index
2008-09-24 13:13:27,738 INFO indexer.IndexMerger - Adding
crawl/indexes/part-00000
2008-09-24 13:13:27,816 INFO indexer.IndexMerger - done merging
2008-09-24 13:13:27,832 INFO crawl.Crawl - crawl finished: crawl

Best regards,
Biswajit.


Susam Pal wrote: