@JLo_Hernandez HDP3 uses HBase as a backend for the timeline service. You can either use an external HBase or have a system HBase running on Yarn (the default). With a server crash, your ATSv2 HBase Application data could be corrupted this can impact performance because of timeouts from the ATS To fix that follow: Check for any error in the yarn logs usually in /var/log/hadoop-yarn/yarn/ for anything clear to spot, for instance, not enough yarn memory (and then fix it if relevant) see these 2 files hadoop-yarn-timelineserver-<host_name>.out hadoop-yarn-timelineserver-<host_name>.log Clean up hdfs ATS data as described in Remove ats-HBase before switching between clusters note there are different steps for secure [Kerberized] and unsecure clusters Clean up zookeeper ATS data (the example here is for insecure clusters, you will probably have another znode for kerberised clusters): zookeeper-client rmr /atsv2-hbase-unsecure Log on zookeeper here I am on an secure HDP 3.1 single node cluster # /usr/hdp/3.1.0.0-78/zookeeper/bin/zkCli.sh [zk: localhost:2181(CONNECTED) 0] ls / [cluster, registry, controller, brokers, storm, zookeeper, infra-solr, hbase-unsecure, admin, isr_change_notification, log_dir_event_notification, controller_epoch, hiveserver2, hiveserver2-leader, rmstore, atsv2-hbase-unsecure, consumers, ambari-metrics-cluster, latest_producer_id_block, config] Go for the ATSv2 entry [zk: localhost:2181(CONNECTED) 1] ls /atsv2-hbase-unsecure [rs, splitWAL, backup-masters, table-lock, draining, master-maintenance, table] Delete the entry [zk: localhost:2181(CONNECTED) 2] rmr /atsv2-hbase-unsecure Restart *all* YARN services, Restart ambari server You lose your ATS history with the above hdfs & zookeeper cleaning steps like (ie. job names, timing, logs…), but your actual data is perfectly safe, nothing else will be lost. Please revert !!
... View more
