Why does Riak log "unavailable channel closed" message for ERROR message

I'm not a Riak expert, but it looks like the message inavailable channel closed

should be an INFO level, not an ERROR level log message. We open some Riak connections when our server boots up to fetch user data and then they timeout (which should be fine, right?) When they exit the message box inavailable channel closed

.

Are we using Riak incorrectly? Is there a REAL bug? Or is it just an INFO level because it is part of the normal Riak lifecycle?

http://grepcode.com/file/repo1.maven.org/maven2/com.basho.riak/riak-client/2.0.0/com/basho/riak/client/core/RiakNode.java

13:13:51.533 [nioEventLoopGroup-2-1] DEBUG s.r.s.i.n.u.i.JavassistTypeParameterMatcherGenerator - Generated: scifinder.riak.shaded.io.netty.util.internal.__matchers__.com.basho.riak.client.core.RiakMessageMatcher
13:13:51.541 [nioEventLoopGroup-2-1] DEBUG s.r.s.i.n.u.i.JavassistTypeParameterMatcherGenerator - Generated: scifinder.riak.shaded.io.netty.util.internal.__matchers__.com.basho.riak.client.core.FutureOperationMatcher
13:13:51.552 [main] INFO  com.basho.riak.client.core.RiakNode - RiakNode started; riak.address.org:8087
13:13:51.553 [main] INFO  c.basho.riak.client.core.RiakCluster - RiakCluster is starting.
13:13:51.865 [main] INFO  com.basho.riak.client.core.RiakNode - RiakNode started; riak.address.org:8087
13:13:51.865 [main] INFO  c.basho.riak.client.core.RiakCluster - RiakCluster is starting.
13:13:52.174 [main] INFO  com.basho.riak.client.core.RiakNode - RiakNode started; riak.address.org:8087
13:13:52.174 [main] INFO  c.basho.riak.client.core.RiakCluster - RiakCluster is starting.
13:13:52.206 [main] DEBUG com.basho.riak.client.core.RiakNode - Operation being executed on RiakNode riak.address.org:8087
13:13:52.220 [nioEventLoopGroup-4-1] DEBUG s.r.s.i.n.util.ResourceLeakDetector - -Dscifinder.riak.shaded.io.netty.leakDetectionLevel: simple
13:13:52.233 [nioEventLoopGroup-4-1] DEBUG com.basho.riak.client.core.RiakNode - Operation onSuccess() channel: id:-486690393 riak.address.org:8087
13:13:52.234 [nioEventLoopGroup-4-1] DEBUG c.basho.riak.client.core.RiakCluster - operation complete; remaining retries: 2
13:13:52.235 [nioEventLoopGroup-4-1] DEBUG com.basho.riak.client.core.RiakNode - Channel id:-486690393 returned to pool
13:13:52.235 [nioEventLoopGroup-4-1] DEBUG com.basho.riak.client.core.RiakNode - Released pool permit

      

Error logs

14:13:31.708 [nioEventLoopGroup-2-1] ERROR com.basho.riak.client.core.RiakNode - inAvailable channel closed; id:601172895 riak.address.org:8087
14:13:32.036 [nioEventLoopGroup-3-1] ERROR com.basho.riak.client.core.RiakNode - inAvailable channel closed; id:-319701844 riak.address.org:8087
14:13:32.410 [nioEventLoopGroup-4-1] ERROR com.basho.riak.client.core.RiakNode - inAvailable channel closed; id:-486690393 riak.address.org:8087

      

+3


source to share


1 answer


It looks like the developers noticed the same thing because this bug was filed in June last year and the post was reduced to INFO in this PR a month later.



If you still see them as errors, please update your java-riak client to v2.0.7 or newer.

+1


source







All Articles