AnsweredAssumed Answered

NFS server will not start (V6.0.1)

Question asked by terryhealy on May 30, 2018
Latest reply on Jul 30, 2018 by terryhealy

After upgrading to 6.0.1 and getting through various licensing and MEP issues, I was given a license for our Community Edition cluster. As a result of using the installer as part of these procedures, NFS was installed on ALL the nodes as opposed to just the one node allowed under Community. Some many of the "new" NFS servers are reported up, but I cannot get the "real" one going.

 

Here is what I have tried:

  1. yum remove and re-install;
  2. Start and stop the service with MCS;
  3. Deleted and recreated the nfsserver.log file  (which has been empty ever since, despite open read/write permissions and ownership of mapr:mapr);
  4. Checked everything in  https://maprdocs.mapr.com/home/AdministratorGuide/c_nfs_setup_requirements.html,
  5. changing one line in /opt/mapr/conf/daemon.conf from "mapr.daemon.group=mapr" to: "mapr.daemon.group=mapr.daemon.group" as I understood it to recommend; restarted warden again.

 

So I still have an empty nfsserver.log file. Below are the results of 'grep -i nfs warden.log'

 

Can anyone suggest next steps in troubleshooting? Is it necessary to remove nfs from the "other" nodes? (I would have expected moaning about licenses, and it appears the license I received is a 30 day Enterprise trial rather than a Community)

 

2018-05-30 10:39:41,896 INFO com.mapr.job.mngmnt.hadoop.metrics.WardenRequestBuilder [nfs_monitor]: [e_SERV_RUN, hostName, ma_host, ma_process]
2018-05-30 10:39:41,896 INFO com.mapr.job.mngmnt.hadoop.metrics.WardenRequestBuilder [nfs_monitor]: []
2018-05-30 10:39:42,018 ERROR com.mapr.warden.service.baseservice.Service$ServiceMonitorRun run [nfs_monitor]: Monitor command: [/opt/mapr/initscripts/mapr-nfsserver, status]can not determine if service: nfs is running. Retrying. Retrial #2. Total retries count is: 3
2018-05-30 10:39:42,018 ERROR com.mapr.warden.service.baseservice.Service$ServiceMonitorRun run [nfs_monitor]: nfsserver is stopped
2018-05-30 10:39:42,018 INFO com.mapr.warden.service.baseservice.Service$ServiceRun [nfs_monitor]: Command: [/opt/mapr/initscripts/mapr-nfsserver, start], Directory: /opt/mapr/initscripts
2018-05-30 10:39:42,113 INFO com.mapr.warden.service.baseservice.Service$ServiceRun [nfs_monitor]:
2018-05-30 10:39:52,113 INFO com.mapr.job.mngmnt.hadoop.metrics.WardenRequestBuilder [nfs_monitor]: [e_SERV_RUN, hostName, ma_host, ma_process]
2018-05-30 10:39:52,113 INFO com.mapr.job.mngmnt.hadoop.metrics.WardenRequestBuilder [nfs_monitor]: []
2018-05-30 10:39:52,430 ERROR com.mapr.warden.service.baseservice.Service$ServiceMonitorRun run [nfs_monitor]: Monitor command: [/opt/mapr/initscripts/mapr-nfsserver, status]cannot determine if service: nfs is running. Number of retrials exceeded. Closing Zookeeper
2018-05-30 10:39:52,430 INFO com.mapr.warden.service.baseservice.Service [nfs_monitor]: 96 about to close zk for service: nfs
2018-05-30 10:39:52,433 INFO com.mapr.warden.service.baseservice.Service [main-EventThread]: Process path: /services/nfs/hd10.sec.bnl.local. Event state: SyncConnected. Event type: NodeDeleted
2018-05-30 10:39:52,433 INFO com.mapr.warden.service.baseservice.Service [main-EventThread]: ZK is closed for service: nfs
2018-05-30 10:39:52,437 INFO com.mapr.job.mngmnt.hadoop.metrics.WardenRequestBuilder [nfs_monitor]: [e_SERV_FAIL, hostName, ma_host, ma_process]
2018-05-30 10:39:52,437 INFO com.mapr.job.mngmnt.hadoop.metrics.WardenRequestBuilder [nfs_monitor]: []
2018-05-30 10:39:52,437 INFO com.mapr.warden.service.baseservice.Service [nfs_monitor]: Alarm raising command: [/opt/mapr/bin/maprcli, alarm, raise, -alarm, NODE_ALARM_SERVICE_NFS_DOWN, -entity, hd10.sec.bnl.local, -description, Can not determine if service: nfs is running. Check logs at: /opt/mapr/logs/nfsserver.log]
2018-05-30 10:41:27,175 INFO com.mapr.warden.service.baseservice.Service [Thread-12-EventThread]: Process path: /services/nfs. Event state: SyncConnected. Event type: NodeChildrenChanged

Outcomes