[
https://issues.apache.org/jira/browse/NUTCH-2000?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14589860#comment-14589860
]
Sebastian Nagel commented on NUTCH-2000:
----------------------------------------
This could be caused by a bug introduced with NUTCH-1913 :( - a lock file is
left if both db.ignore.internal.links and db.ignore.external.links are true.
The next call of "invertlinks" will then fail with the mentioned error. Patch
attached.
> Link inversion fails with .locked already exists.
> -------------------------------------------------
>
> Key: NUTCH-2000
> URL: https://issues.apache.org/jira/browse/NUTCH-2000
> Project: Nutch
> Issue Type: Bug
> Affects Versions: 1.9
> Reporter: Julien Nioche
> Fix For: 1.11
>
>
> using standard crawl script with a brand new test dir in local mode I am
> getting
> Link inversion
> /data/BLABLABLA/runtime/local/bin/nutch invertlinks
> /data/BLABLABLA/testCrawl2//linkdb
> /data/BLABLABLA/testCrawl2//segments/20150423114335
> LinkDb: java.io.IOException: lock file
> /data/BLABLABLA/testCrawl2/linkdb/.locked already exists.
> PS: 2000!
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
