[jira] Commented: (NUTCH-269) CrawlDbReducer: OOME because no upper-bound on inlinks count

2010-01-08 Thread Julien Nioche (JIRA)

[ 
https://issues.apache.org/jira/browse/NUTCH-269?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12797990#action_12797990
 ] 

Julien Nioche commented on NUTCH-269:
-

I will shortly commit a variant of this approach whereby the inlinks are stored 
in a priority queue in order to keep the best scoring ones. The size of the 
queue is determined by the parameter db.update.max.inlinks which has a default 
value of 1.

 CrawlDbReducer: OOME because no upper-bound on inlinks count
 

 Key: NUTCH-269
 URL: https://issues.apache.org/jira/browse/NUTCH-269
 Project: Nutch
  Issue Type: Bug
Reporter: stack
Assignee: Julien Nioche
Priority: Trivial
 Attachments: too-many-links.patch, too-many-links2.patch


 A CrawlDB update repeatedly OOME'd because an URL had hundreds of thousands 
 of inlinks (The british foriegn office likes putting a clear.gif multiple 
 times into each page: 
 http://www.fco.gov.uk/Xcelerate/graphics/images/fcomain/clear.gif).

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.



[jira] Commented: (NUTCH-269) CrawlDbReducer: OOME because no upper-bound on inlinks count

2010-01-08 Thread Hudson (JIRA)

[ 
https://issues.apache.org/jira/browse/NUTCH-269?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12798305#action_12798305
 ] 

Hudson commented on NUTCH-269:
--

Integrated in Nutch-trunk #1034 (See 
[http://hudson.zones.apache.org/hudson/job/Nutch-trunk/1034/])
 : OOME because no upper-bound on inlinks count (stack + jnioche)


 CrawlDbReducer: OOME because no upper-bound on inlinks count
 

 Key: NUTCH-269
 URL: https://issues.apache.org/jira/browse/NUTCH-269
 Project: Nutch
  Issue Type: Bug
Reporter: stack
Assignee: Julien Nioche
Priority: Trivial
 Fix For: 1.1

 Attachments: too-many-links.patch, too-many-links2.patch


 A CrawlDB update repeatedly OOME'd because an URL had hundreds of thousands 
 of inlinks (The british foriegn office likes putting a clear.gif multiple 
 times into each page: 
 http://www.fco.gov.uk/Xcelerate/graphics/images/fcomain/clear.gif).

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.