Hide Forgot
Document URL: https://access.redhat.com/documentation/en-US/Red_Hat_Storage/3.1/html/Administration_Guide/sect-NFS.html#sect-NFS_Ganesha Section Number and Name: 7.2.4.4.1. Prerequisites to run NFS-Ganesha --> Defining Service Ports Describe the issue: when executing "# systemctl restart nfs-mountd" the local (kernel) nfs-server is started. Since a lot of people will try to configure the Ganesha server also as client machine for debugging - the now started nfs-server will block the ganesha installation to finish successfully. Do we even need to restart the nfs-mountd service? Suggestions for improvement: Find out if we really need to restart this service. If we do - add a Note to stop the NFS-server afterwards. Additional information:
(In reply to Chris Blum from comment #0) > Document URL: > https://access.redhat.com/documentation/en-US/Red_Hat_Storage/3.1/html/ > Administration_Guide/sect-NFS.html#sect-NFS_Ganesha > > Section Number and Name: > 7.2.4.4.1. Prerequisites to run NFS-Ganesha > --> Defining Service Ports > > Describe the issue: > when executing "# systemctl restart nfs-mountd" the local (kernel) > nfs-server is started. Since a lot of people will try to configure the > Ganesha server also as client machine for debugging - the now started > nfs-server will block the ganesha installation to finish successfully. > Do we even need to restart the nfs-mountd service? If MOUNTD port is being configured on the client machine, I dont know if there is any other way to make it effective apart from restarting nfs-mountd service. Request Steve, Ben and others to comment. Also it is recommended to use a separate machine as client as there could be conflicts wrt to lock manager services on the client and server machines. If it is just for testing/debugging firewalld/iptables could be disabled which would mean these ports configuration is not needed. Probably we should add a note that this section need not be considered if firewalld services are disabled. Thanks!
The nfs-client should not need the MOUNTD service. Note that a host that is running Ganesha can NOT use locks via NFS v3 client (NLM which is how NFS v3 locks are implemented requires the client be an NLM server as well as the server. It is impossible to run both nfs-Ganesha and the Linux client NLM servers on the same host).
As per comment#3 by Frank, since we do not need MOUNTD service to be running on the nfs-client machine, I think we can remove those ports configuration from the guide - i.e, modify the section - Defining Service Ports as per below >>>> Execute the following steps on the client machine: Edit '/etc/sysconfig/nfs' using following commands: # sed -i '/STATD_PORT/s/^#//' /etc/sysconfig/nfs # sed -i '/LOCKD_TCPPORT/s/^#//' /etc/sysconfig/nfs # sed -i '/LOCKD_UDPPORT/s/^#//' /etc/sysconfig/nfs Restart the services: For Red Hat Enterprise Linux 6: # service nfslock restart # service nfs restart For Red Hat Enterprise Linux 7: # systemctl restart nfs-config # systemctl restart rpc-statd # systemctl restart nfslock <<<< Change is to remove mountd related lines. @Frank, Request you to verify this section once the corrections are done. Thanks!
Thanks Soumya/Frank, I have gone ahead and made the changes based on comment 5 in section 7.2.4.4.1. Prerequisites to run NFS-Ganesha: http://ccs-jenkins.gsslab.brq.redhat.com:8080/job/doc-Red_Hat_Gluster_Storage-3.2-Administration_Guide-branch-NFS-Ganesha-Bugs/lastSuccessfulBuild/artifact/tmp/en-US/html-single/index.html#sect-NFS_Ganesha Let me know if you have any further comments before I move it to ON_QA.
The changes look good to me. Request Frank to review as well.
You can't run nfslock with Ganesha... Also shouldn't the knfsd also be disabled, not just stopped?
Thanks Frank. You are right. We should be disabling kernel-NFS. We shall document it as part of bug1392000 in pre-requisites. On the server machine after configuring STATD port, we need to restart rpc-statd process. On RHEL 7 systems, we have rpc-statd service file but couldn't find an equivalent init.d service for RHEL6 system. Hence we had documented to restart nfslock service on RHEL6. Is there any other way to restart just rpc-statd service on RHEL6 apart from manually running /usr/sbin/statr_statd? Thanks!
(In reply to Soumya Koduri from comment #9) > Thanks Frank. You are right. We should be disabling kernel-NFS. We shall > document it as part of bug1392000 in pre-requisites. > > On the server machine after configuring STATD port, we need to restart > rpc-statd process. On RHEL 7 systems, we have rpc-statd service file but > couldn't find an equivalent init.d service for RHEL6 system. Hence we had > documented to restart nfslock service on RHEL6. Is there any other way to > restart just rpc-statd service on RHEL6 apart from manually running > /usr/sbin/statr_statd? Thanks! There may not be, the old scripts when we first started on Ganesha had the Ganesha script start rpc.statd.
Thanks Soumya/ Frank. As mentioned in comment 9, documenting disabling kernel-NFS is taken care of as part of bug139000. Moving this bug to On_QA http://ccs-jenkins.gsslab.brq.redhat.com:8080/job/doc-Red_Hat_Gluster_Storage-3.2-Administration_Guide-branch-master/lastSuccessfulBuild/artifact/tmp/en-US/html-single/index.html#sect-NFS_Ganesha
Changes looks good to me. Moving the BZ to verified.
RHGS 3.2.0 GA completed on 23 March 2017