I've only seen it from another user on the OpenTSDB mailing list. :-) The hack to write to accumulo using OpenTSDB was to replace the hbase async library. This trace indictes that it is still using the real hbase async library.
You will want to ensure your OpenTSDB doesn't have this library in your path, and just has the fake library that talks to accumulo. -Eric On Tue, Jun 3, 2014 at 7:04 PM, Arshak Navruzyan <[email protected]> wrote: > Eric, > > Curious if you've seen this error before. > > Thanks, > > Arshak > > 2014-06-03 15:30:53,736 INFO [main] ZooKeeper: Initiating client > connection, connectString=accumulo://root:secret@localhost:2181/ecluster > sessionTimeout=5000 watcher=org.hbase.async.HBaseClient$ZKClient@6426f7af > > 2014-06-03 15:30:53,767 INFO [main] TSDB: Flushing compaction queue > > Exception in thread "main" java.lang.RuntimeException: Initialization > failed > > at net.opentsdb.tools.TSDMain.main(TSDMain.java:182) > > Caused by: java.lang.IllegalArgumentException: Invalid path string > "//root:secret@localhost:2181/ecluster" caused by empty node name > specified @1 > > at org.apache.zookeeper.common.PathUtils.validatePath(PathUtils.java:99) > > at org.apache.zookeeper.ClientCnxn.<init>(ClientCnxn.java:363) > > at org.apache.zookeeper.ClientCnxn.<init>(ClientCnxn.java:327) > > at org.apache.zookeeper.ZooKeeper.<init>(ZooKeeper.java:383) > > at org.hbase.async.HBaseClient$ZKClient.connectZK(HBaseClient.java:3004) > > at > org.hbase.async.HBaseClient$ZKClient.getDeferredRoot(HBaseClient.java:2909) > > at org.hbase.async.HBaseClient.locateRegion(HBaseClient.java:1871) > > at org.hbase.async.HBaseClient.sendRpcToRegion(HBaseClient.java:1703) > > at > org.hbase.async.HBaseClient.ensureTableFamilyExists(HBaseClient.java:947) > > at org.hbase.async.HBaseClient.ensureTableExists(HBaseClient.java:967) > > at net.opentsdb.core.TSDB.checkNecessaryTablesExist(TSDB.java:317) > > at net.opentsdb.tools.TSDMain.main(TSDMain.java:147) > > > On Sun, May 11, 2014 at 1:57 PM, Eric Newton <[email protected]> > wrote: > >> It is being maintained. I have tried very hard not to modify the core >> OpenTSDB to support it. But, it would be nice if we could define a >> storage-independent layer to which it could adhere. >> >> I don't believe the OTSDB team is interested, but a basic scalable >> back-end abstraction would be nice. As would a standard java build >> environment, but that doesn't seem to be wanted, either. >> >> We could work towards a common storage abstraction, which would be a >> reasonable request. >> >> Zipkin does a good job of being storage independent. I would work >> towards their model. >> >> -Eric >> >> On May 11, 2014 10:28 AM, "Arshak Navruzyan" <[email protected]> wrote: >> >>> I noticed Eric Newton's opentsdb adapter for Accumulo. Is this still >>> being maintained? >>> >>> https://github.com/ericnewton/accumulo-opentsdb >>> >>> Also wondering if the StumbleUpon folks are willing to merge it in as an >>> alternative to HBase back end. >>> >>> Thanks >>> >>> Arshak >>> >> >
