[
https://issues.apache.org/jira/browse/METRON-734?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15880624#comment-15880624
]
ASF GitHub Bot commented on METRON-734:
---------------------------------------
Github user merrimanr commented on the issue:
https://github.com/apache/incubator-metron/pull/462
+1 on committing based on Justin's Travis results
> Builds failing because of MaxMind DB transitive dependency
> ----------------------------------------------------------
>
> Key: METRON-734
> URL: https://issues.apache.org/jira/browse/METRON-734
> Project: Metron
> Issue Type: Bug
> Reporter: Justin Leet
> Assignee: Justin Leet
>
> Original Error
> {code}
> [ERROR] Failed to execute goal on project metron-enrichment: Could not
> resolve dependencies for project
> org.apache.metron:metron-enrichment:jar:0.3.1: Failed to collect dependencies
> at com.maxmind.geoip2:geoip2:jar:2.8.0 -> com.maxmind.db:maxmind-db:jar:1.2.1
> -> com.fasterxml.jackson.core:jackson-databind:jar:2.9.0-SNAPSHOT: Failed to
> read artifact descriptor for
> com.fasterxml.jackson.core:jackson-databind:jar:2.9.0-SNAPSHOT: Failure to
> find com.fasterxml:oss-parent:pom:28 in http://clojars.org/repo was cached in
> the local repository, resolution will not be reattempted until the update
> interval of clojars.org has elapsed or updates are forced -> [Help 1]
> {code}
> Discussion on the lib
> https://github.com/maxmind/GeoIP2-java/issues/77
> The geoip2 lib from MaxMind has a transitive dependency on open ended version
> range of jackson-databind. This causes it to try to use 2.9.0-SNAPSHOT,
> rather than a distinct version.
> The easy answer is to just exclude jackson from being pulled in by geoip2
> lib, and use our own. The catch is that the version geoip declares in its
> POM is higher than the version we use and I don't know if that causes any
> problems (I'm guessing not, since it's going from 2.7.x to 2.8.x, but I
> haven't tested it).
> If there's a way to skip snapshots that might also contribute to solving our
> problem, but I still don't like the unrepeatability of the build if they
> depend on an open ended range.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
