nutch fetch issue - empty content

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

nutch fetch issue - empty content

Viral Shah-2-2
Hello --

We are using Nutch to crawl html content for Wikipedia articles. We're using somewhat old nightly build version of nutch. 

We use static list urls as an input. To do this we've injected our list of urls, set db.update.additions.allowed to false, and set the crawl depth to 1.
- We iterate over the output segment files using 'SequenceFile.Reader' and pullout the 'string' as well as 'binary' form of content.
reader = SequenceFile.Reader(filesystem, Path(sys.argv[1]), job)
key = reader.getKeyClass()()
   content = reader.getValueClass()()
     while, content):
content_text = String(content.getContent(), "UTF-8").toString()
                   content_binary = content.getContent()

- I get empty content for some urls but the status in crawldb is set to 'db_fetched'.
The value of content_text is "" and that of content_binary is array('b',[])

- This is completely random in terms of when it happens and the urls involved.

- This failure is completely silent as far as I can tell as nothing can be seen in logs regarding this error.

Again, we are crawling wikipedia which is verifiable for it's content and whether that content is accessible. We have tried manually getting the problem urls and everything looked fine.

Thank you,
Viral Shah