Hi everyone) Namenode heap usage has increased dramatically in the last 2 months, while more than half of the productive load has been transferred to another database management system. How to find out why it happened? ... View more

Hello, dear community. There is a non production cluster HDP 2.3.4 with active kerberos. There was a failure as a result of technical work on network equipment in the data center, which is why the cluster ended up in free space on the HDFS, all node managers became unhealthy, and hbase was not working. It was not possible to add free space at that time, so the files of one of the tables in HDFS were deleted, resulting in enough space on HDFS and all the components in the cluster were able to run, but after that, as expected, the number of Regions in Transition - 6. Regions with the status DN1705000000: Timings, 4 \ xB2 \ x00 \ x00 \ x01i \ xDB \ xB3 \ xC8, \ x00CIp, 1556195004165.45dea4afb84fd338a2ac703193a8f2c3. state = FAILED_OPEN, ts = Fri Apr 26 20:25:09 EEST 2019 (326894s ago), server = vm-hdp-data7.bdata.ab.loc, 16020,1556284336561 How to fix this problem? hbck with different keys can not help. I tried using zookeeper-client to run rmr / hbase-secure / table / DN1705000000: Timings, which as a result issues Authentication is not valid: / hbase-secure / table / DN1705000000: Timings. Help solve the problem. Thanks a lot in advance. ... View more

Hello. Please help in the next question. We always add free disk space to datanodes data mountpoint on the go and its always correctly seen increased disk capacity in ambari. After the last attempt disks capacity physically increased, but ambari saw it on only one datanode. How we can to fix it. ... View more

Hello everybody. I have a problem with access to hdfs via HUE file browser, when active namenode, where HUE installed, switching to standby. Is it possible to keep access to hdfs in this case? StandbyException: Operation category READ is not supported in state standby (error 403) ... View more