About sbn

sbn · ‎02-19-2021

hi, when i upgraded to cdh 6.2.0, the namenodes stopped exposing metrics on 9070. do anyone know where the metrics are exposed now?

Kezia · ‎02-11-2021

Yes, you can set up a local mirror for offline installation. It is possible with CDP version and with older CDH 5 and 6 version. But even theses legacy versions are behing paywall now.

ask_bill_brooks · ‎02-10-2021

Hi @sbn I think you will be able to find a relevant, previously-posted answer to a question very similar to yours in this thread: https://archive.cloudera.com/cm6/6.2.0/ubuntu1604/ displaying 404 error Hope this helps.

sbn · ‎12-08-2020

i have a long running job, that has survived some upgrades (os and older cdh versions) the command executed: /usr/bin/hadoop jar /opt/cloudera/parcels/CDH/lib/hbase/hbase-server.jar importtsv the exception this produces: Exception in thread "main" java.lang.ClassNotFoundException: importtsv it seems that importtsv is not there? however i have not been able to locate the class responsible for this.

GangWar · ‎04-29-2020

@sbn NO, there is no any such public facing doc which can show you the future road map. The only matrix you can see about those versions is available publicly. https://docs.cloudera.com/documentation/enterprise/6/release-notes/topics/rg_os_requirements.html#c63_supported_os https://docs.cloudera.com/cdpdc/7.0/release-guide/topics/cdpdc-os-requirements.html

sbn · ‎03-11-2020

CDH: 5.13.0 HIVE version: VER_ID | SCHEMA_VERSION | VERSION_COMMENT | SCHEMA_VERSION_V2 --------+----------------+----------------------------+------------------- 1 | 1.1.0 | Hive release version 1.1.0 | 1.1.0-cdh5.12.0 when trying to run the 'Upgrade Hive Metastore Database Schema' command through the GUI (CDH) it reports the following: (STDOUT) Starting upgrade metastore schema from version 1.1.0-cdh5.12.0 to 1.1.0-cdh5.13.0 schemaTool completed Exit code: 0 However (STDERR) reports sort of the opposite: INFO metastore.CDHMetaStoreSchemaInfo: Current version is higher than or equal to 1.1.0-cdh5.12.0 Skipping file 1.1.0-to-1.1.0-cdh5.12.0 i want to be on cdh5.13.0 but hive is thinking it's in earlier version. did i miss a migration/upgrade step somewhere to cause this? does anybody know how to force an upgrade to 5.13.0?

sbn · ‎10-24-2019

thanks, i suppose one could just move that to wherever you want the home to be. this was part of a hack deployment, so ultimately not needed.

sbn · ‎08-14-2019

on this guide: https://www.cloudera.com/documentation/enterprise/5-3-x/topics/search_hdfsfindtool.html there is a reference in the -mtime option: Evaluates as true if the file modification time subtracted from the start time is n days what does the "start time" in here reference? it does not seem to work like normal bash find. i want functionality like: if mtime is more than 30 days ago (from current time), print the folder/file.

sbn · ‎06-20-2019

thanks for your answer, do i need the .meta and .meta.tmp files?

jmanel · ‎11-09-2018

It is a fluentd issue, have you checked fluentd documentation ? Take a look to https://github.com/fluent/fluent-plugin-webhdfs : <match access.**> @type webhdfs namenode master1.your.cluster.local:50070 standby_namenode master2.your.cluster.local:50070 path /path/on/hdfs/access.log.%Y%m%d_%H.log </match>

saranvisa · ‎06-28-2018

@sbn yes it requires a cluster restart. But not only this option, whatever the option that you are going to try, it requires a cluster restart, so that namenode can understand the recent configuration change for the journal node. Regarding copying journal node edit dir, was it mentioned anywhere in the link that i've provided? if not, you can ignore that. if so, you can follow the below link (but this is just for reference, i think you don't need to do this) https://www.cloudera.com/documentation/enterprise/5-13-x/topics/cm_mc_jn.html#concept_cqn_dxp_rs

sbn · ‎06-28-2018

thanks, this is what worked in the end, coupled with kill -9 for the really resilient procs.

sbn · ‎04-25-2018

UPDATE: did some experimentations on my own, and i deleted the pkg_resources, in /usr/lib/python2.7/dist-packages this apparantly fixed the issue as hue will now run, and allow me to log in. i am not sure if an update would also work...

sbn · ‎04-05-2017

i want to put a 1:1 elastic cluster inside my hadoop cluster, 1 elastic node on each hadoop datanode. to not interfere too much with the hadoop cluster, i would like to run the elastic nodes on a disk of it's own. the setup: CDH 5.6.0 datanode disk layout(in mounted dirs): /data/disk1 . . . /data/disk10 20+ data nodes say i wish to remove disk10 from each datanode, how do i do that without data loss? removing the disk on a decommissioned datanote, and later recomissioning it, takes too long time. - any hint on making this process faster? can i use the rebalancer? ( i saw there is internal datanode balancer in CDH5.8)

Online	Offline
Last Visited	‎02-19-2021 09:34 AM

Date Registered	‎02-16-2017 04:34 AM
Last Visited	‎02-19-2021 09:34 AM
Total Messages Posted	20
Total Kudos Received	1

namenode port 9070/metrics moved in 6.2 where is i...

Re: can i mirror archive.cloudera.com for cdh 5 an...

Re: archive-primary.cloudera.com/cdh5/cdh/5 is not...

importtsv not working after upgrade to cdh 6.2.0

Re: OS targets for new releases (focal)

Upgrade Hive Metastore Database Schema Command not...

Re: Set SPARK_HOME for spark2

hdfs file "start time" in org.apache.solr.hadoop.H...

Re: spooldir channel error - too many files. - how...

Re: webhdfs way to always post to ACTIVE namenode

Re: moving a journalnode or commision a new, and d...

Re: how to move cloudera procs to new supervisord

Re: import error (python) starting hue after upgra...

Removing a disk on every datanode - for a elastics...