Topic
  • 11 replies
  • Latest Post - ‏2013-05-13T20:59:11Z by StorageTeamGenomeCenter
StorageTeamGenomeCenter
31 Posts

Pinned topic packages after nfs-utils-1.2.3-15.el6 on rhel breaks CNFS GPFS version 3.5

‏2013-03-15T19:34:43Z |
If a more current version nfs-utils is used than nfs-utils-1.2.3-15.el6. The client gets denied access when trying to mount the volume over nfs. The server shows "getfh failed: Operation not permitted" in the logs.

If nfs is stopped and started manually using rhel's init scripts it works fine. But when nfs is started by GPFS/CNFS it does not work correctly.

We started seeing this starting in RHEL 6.2. We have opened up tickets with Redhat and IBM. Since NFS functions properly without GPFS redhat points to GPFS and of course IBM wants to point at Redhat. So we are stuck using this old version that has bugs.

Mar 1 10:56:46 linuscs109 rpc.mountd18917: Version 1.2.3 starting
Mar 1 10:56:47 linuscs109 kernel: NFSD: Using /var/lib/nfs/v4recovery as the NFSv4 state recovery directory
Mar 1 10:56:47 linuscs109 kernel: NFSD: starting 90-second grace period
Mar 1 10:56:52 linuscs109 CNFS: NFS clients of node 10.0.28.59 notified to reclaim NLM locks
Mar 1 10:56:52 linuscs109 sm-notify19535: Version 1.2.3 starting
Mar 1 10:56:52 linuscs109 sm-notify19535: Already notifying clients; Exiting!
Mar 1 10:56:52 linuscs109 CNFS: Monitor has started pid=19570
Mar 1 10:58:36 linuscs109 rpc.mountd18917: authenticated mount request from 10.0.2.225:1010 for /vol/aggr0 (/vol/aggr0)
Mar 1 10:58:36 linuscs109 rpc.mountd18917: internal: no supported addresses in nfs_client
Mar 1 10:58:36 linuscs109 rpc.mountd18917: getfh failed: Operation not permitted
Mar 1 10:58:36 linuscs109 rpc.mountd18917: getfh failed: Operation not permitted
root@linuscs109 /# yum list --showduplicates nfs-utils-1.2.3
Loaded plugins: product-id, rhnplugin, security, subscription-manager
This system is not registered to Red Hat Subscription Management. You can use subscription-manager to register.
This system is receiving updates from RHN Classic or RHN Satellite.
Installed Packages
nfs-utils.x86_64 1:1.2.3-15.el6 @rhel-x86_64-server-6
Available Packages
nfs-utils.x86_64 1:1.2.3-7.el6 rhel-x86_64-server-6
nfs-utils.x86_64 1:1.2.3-7.el6_1.1 rhel-x86_64-server-6
nfs-utils.x86_64 1:1.2.3-15.el6 rhel-x86_64-server-6
nfs-utils.x86_64 1:1.2.3-15.el6_2.1 rhel-x86_64-server-6
nfs-utils.x86_64 1:1.2.3-26.el6 rhel-x86_64-server-6
nfs-utils.x86_64 1:1.2.3-36.el6 rhel-x86_64-server-6
Updated on 2013-03-19T15:23:27Z at 2013-03-19T15:23:27Z by StorageTeamGenomeCenter
  • truongv
    truongv
    81 Posts

    Re: packages after nfs-utils-1.2.3-15.el6 on rhel breaks CNFS GPFS version 3.5

    ‏2013-03-15T21:29:18Z  
    We don't see this problem. I have a cNFS node that is running RHEL6.3 2.6.32-279.el6.ppc64 and nfs-utils-1.2.3-26.el6.ppc64. I was able to mount the a few difference distro clients. Have you try to mount with nfs3 option on clients?
  • tomerperry
    tomerperry
    18 Posts

    Re: packages after nfs-utils-1.2.3-15.el6 on rhel breaks CNFS GPFS version 3.5

    ‏2013-03-17T09:17:41Z  
    • truongv
    • ‏2013-03-15T21:29:18Z
    We don't see this problem. I have a cNFS node that is running RHEL6.3 2.6.32-279.el6.ppc64 and nfs-utils-1.2.3-26.el6.ppc64. I was able to mount the a few difference distro clients. Have you try to mount with nfs3 option on clients?
    Can you make sure that nfsv4 is disabled on the server? ( RPCNFSDARGS="-N 4" on /etc/sysconfig/nfs )
  • StorageTeamGenomeCenter
    31 Posts

    Re: packages after nfs-utils-1.2.3-15.el6 on rhel breaks CNFS GPFS version 3.5

    ‏2013-03-18T14:17:23Z  
    Can you make sure that nfsv4 is disabled on the server? ( RPCNFSDARGS="-N 4" on /etc/sysconfig/nfs )
    These are the options we set in the file. All clients mount with nfs version 3.

    MOUNTD_PORT=4046
    STATD_PORT=4047
    LOCKD_TCPPORT=4045
    LOCKD_UDPPORT=4045
    RQUOTAD_PORT=4049
    RPCNFSDARGS="-N 4"

    root@linuscs107 ~# rpcinfo -p linuscs107
    program vers proto port service
    100000 4 tcp 111 portmapper
    100000 3 tcp 111 portmapper
    100000 2 tcp 111 portmapper
    100000 4 udp 111 portmapper
    100000 3 udp 111 portmapper
    100000 2 udp 111 portmapper
    100024 1 udp 4047 status
    100024 1 tcp 4047 status
    100021 1 udp 4045 nlockmgr
    100021 3 udp 4045 nlockmgr
    100021 4 udp 4045 nlockmgr
    100021 1 tcp 4045 nlockmgr
    100021 3 tcp 4045 nlockmgr
    100021 4 tcp 4045 nlockmgr
    100011 1 udp 4049 rquotad
    100011 2 udp 4049 rquotad
    100011 1 tcp 4049 rquotad
    100011 2 tcp 4049 rquotad
    100005 1 udp 4046 mountd
    100005 1 tcp 4046 mountd
    100005 2 udp 4046 mountd
    100005 2 tcp 4046 mountd
    100005 3 udp 4046 mountd
    100005 3 tcp 4046 mountd
    100003 2 udp 2049 nfs
    100003 3 udp 2049 nfs
    100227 2 udp 2049 nfs_acl
    100227 3 udp 2049 nfs_acl
    100003 2 tcp 2049 nfs
    100003 3 tcp 2049 nfs
    100227 2 tcp 2049 nfs_acl
    100227 3 tcp 2049 nfs_acl
  • tomerperry
    tomerperry
    18 Posts

    Re: packages after nfs-utils-1.2.3-15.el6 on rhel breaks CNFS GPFS version 3.5

    ‏2013-03-18T14:35:43Z  
    These are the options we set in the file. All clients mount with nfs version 3.

    MOUNTD_PORT=4046
    STATD_PORT=4047
    LOCKD_TCPPORT=4045
    LOCKD_UDPPORT=4045
    RQUOTAD_PORT=4049
    RPCNFSDARGS="-N 4"

    root@linuscs107 ~# rpcinfo -p linuscs107
    program vers proto port service
    100000 4 tcp 111 portmapper
    100000 3 tcp 111 portmapper
    100000 2 tcp 111 portmapper
    100000 4 udp 111 portmapper
    100000 3 udp 111 portmapper
    100000 2 udp 111 portmapper
    100024 1 udp 4047 status
    100024 1 tcp 4047 status
    100021 1 udp 4045 nlockmgr
    100021 3 udp 4045 nlockmgr
    100021 4 udp 4045 nlockmgr
    100021 1 tcp 4045 nlockmgr
    100021 3 tcp 4045 nlockmgr
    100021 4 tcp 4045 nlockmgr
    100011 1 udp 4049 rquotad
    100011 2 udp 4049 rquotad
    100011 1 tcp 4049 rquotad
    100011 2 tcp 4049 rquotad
    100005 1 udp 4046 mountd
    100005 1 tcp 4046 mountd
    100005 2 udp 4046 mountd
    100005 2 tcp 4046 mountd
    100005 3 udp 4046 mountd
    100005 3 tcp 4046 mountd
    100003 2 udp 2049 nfs
    100003 3 udp 2049 nfs
    100227 2 udp 2049 nfs_acl
    100227 3 udp 2049 nfs_acl
    100003 2 tcp 2049 nfs
    100003 3 tcp 2049 nfs
    100227 2 tcp 2049 nfs_acl
    100227 3 tcp 2049 nfs_acl
    Weird - same here - without any issues...

    root@hyp02 ~# rpcinfo -p localhost
    program vers proto port service
    100000 4 tcp 111 portmapper
    100000 3 tcp 111 portmapper
    100000 2 tcp 111 portmapper
    100000 4 udp 111 portmapper
    100000 3 udp 111 portmapper
    100000 2 udp 111 portmapper
    100024 1 udp 38820 status
    100024 1 tcp 53845 status
    100021 1 udp 50134 nlockmgr
    100021 3 udp 50134 nlockmgr
    100021 4 udp 50134 nlockmgr
    100021 1 tcp 58916 nlockmgr
    100021 3 tcp 58916 nlockmgr
    100021 4 tcp 58916 nlockmgr
    100011 1 udp 875 rquotad
    100011 2 udp 875 rquotad
    100011 1 tcp 875 rquotad
    100011 2 tcp 875 rquotad
    100005 1 udp 32819 mountd
    100005 1 tcp 42419 mountd
    100005 2 udp 33695 mountd
    100005 2 tcp 59263 mountd
    100005 3 udp 58739 mountd
    100005 3 tcp 52792 mountd
    100003 2 tcp 2049 nfs
    100003 3 tcp 2049 nfs
    100227 2 tcp 2049 nfs_acl
    100227 3 tcp 2049 nfs_acl
    100003 2 udp 2049 nfs
    100003 3 udp 2049 nfs
    100227 2 udp 2049 nfs_acl
    100227 3 udp 2049 nfs_acl

    root@hyp02 ~# yum list --showduplicates nfs-utils-1.2.3
    Loaded plugins: product-id, security, subscription-manager
    This system is not registered to Red Hat Subscription Management. You can use subscription-manager to register.
    Installed Packages
    nfs-utils.x86_64 1:1.2.3-36.el6 @anaconda-RedHatEnterpriseLinux-201301301459.x86_64/6.4
    Available Packages
    nfs-utils.x86_64 1:1.2.3-36.el6 local-rhels6.4-x86_64-Server
    nfs-utils.x86_64 1:1.2.3-36.el6 local-rhels6.4-x86_64-x86_64
  • StorageTeamGenomeCenter
    31 Posts

    Re: packages after nfs-utils-1.2.3-15.el6 on rhel breaks CNFS GPFS version 3.5

    ‏2013-03-18T14:47:32Z  
    Weird - same here - without any issues...

    root@hyp02 ~# rpcinfo -p localhost
    program vers proto port service
    100000 4 tcp 111 portmapper
    100000 3 tcp 111 portmapper
    100000 2 tcp 111 portmapper
    100000 4 udp 111 portmapper
    100000 3 udp 111 portmapper
    100000 2 udp 111 portmapper
    100024 1 udp 38820 status
    100024 1 tcp 53845 status
    100021 1 udp 50134 nlockmgr
    100021 3 udp 50134 nlockmgr
    100021 4 udp 50134 nlockmgr
    100021 1 tcp 58916 nlockmgr
    100021 3 tcp 58916 nlockmgr
    100021 4 tcp 58916 nlockmgr
    100011 1 udp 875 rquotad
    100011 2 udp 875 rquotad
    100011 1 tcp 875 rquotad
    100011 2 tcp 875 rquotad
    100005 1 udp 32819 mountd
    100005 1 tcp 42419 mountd
    100005 2 udp 33695 mountd
    100005 2 tcp 59263 mountd
    100005 3 udp 58739 mountd
    100005 3 tcp 52792 mountd
    100003 2 tcp 2049 nfs
    100003 3 tcp 2049 nfs
    100227 2 tcp 2049 nfs_acl
    100227 3 tcp 2049 nfs_acl
    100003 2 udp 2049 nfs
    100003 3 udp 2049 nfs
    100227 2 udp 2049 nfs_acl
    100227 3 udp 2049 nfs_acl

    root@hyp02 ~# yum list --showduplicates nfs-utils-1.2.3
    Loaded plugins: product-id, security, subscription-manager
    This system is not registered to Red Hat Subscription Management. You can use subscription-manager to register.
    Installed Packages
    nfs-utils.x86_64 1:1.2.3-36.el6 @anaconda-RedHatEnterpriseLinux-201301301459.x86_64/6.4
    Available Packages
    nfs-utils.x86_64 1:1.2.3-36.el6 local-rhels6.4-x86_64-Server
    nfs-utils.x86_64 1:1.2.3-36.el6 local-rhels6.4-x86_64-x86_64
    Well I did some further testing and since I upgraded to the latest kernel it seems to be working fine. I may downgrade to the last kernel just to make sure that the kernel and nfs-utils combo was causing this issue.
    This combination seems to work fine.
    root@linuscs109 ~# uname -a
    Linux linuscs109 2.6.32-358.0.1.el6.x86_64 #1 SMP Wed Feb 20 11:05:23 EST 2013 x86_64 x86_64 x86_64 GNU/Linux
    root@linuscs109 ~# rpm -qa | grep gpfs
    gpfs.gplbin-2.6.32-358.0.1.el6.x86_64-3.5.0-7.x86_64
    gpfs.docs-3.5.0-7.noarch
    gpfs.gpl-3.5.0-7.noarch
    gpfs.base-3.5.0-7.x86_64
    gpfs.msg.en_US-3.5.0-7.noarch
    root@linuscs109 ~# rpm -qa | grep nfs-utils
    nfs-utils-lib-1.1.5-6.el6.x86_64
    nfs-utils-1.2.3-36.el6.x86_64
  • StorageTeamGenomeCenter
    31 Posts

    Re: packages after nfs-utils-1.2.3-15.el6 on rhel breaks CNFS GPFS version 3.5

    ‏2013-03-18T15:11:15Z  
    Well I did some further testing and since I upgraded to the latest kernel it seems to be working fine. I may downgrade to the last kernel just to make sure that the kernel and nfs-utils combo was causing this issue.
    This combination seems to work fine.
    root@linuscs109 ~# uname -a
    Linux linuscs109 2.6.32-358.0.1.el6.x86_64 #1 SMP Wed Feb 20 11:05:23 EST 2013 x86_64 x86_64 x86_64 GNU/Linux
    root@linuscs109 ~# rpm -qa | grep gpfs
    gpfs.gplbin-2.6.32-358.0.1.el6.x86_64-3.5.0-7.x86_64
    gpfs.docs-3.5.0-7.noarch
    gpfs.gpl-3.5.0-7.noarch
    gpfs.base-3.5.0-7.x86_64
    gpfs.msg.en_US-3.5.0-7.noarch
    root@linuscs109 ~# rpm -qa | grep nfs-utils
    nfs-utils-lib-1.1.5-6.el6.x86_64
    nfs-utils-1.2.3-36.el6.x86_64
    confirmed something is broke with this combination. The newest kernel seems to fix the problem.

    root@linuscs109 ~# uname -a
    Linux linuscs109 2.6.32-279.19.1.el6.x86_64 #1 SMP Sat Nov 24 14:35:28 EST 2012 x86_64 x86_64 x86_64 GNU/Linux
    root@linuscs109 ~# rpm -qa | grep nfs
    nfs-utils-lib-1.1.5-6.el6.x86_64
    nfs4-acl-tools-0.3.3-6.el6.x86_64
    nfs-utils-1.2.3-36.el6.x86_64
    root@linuscs109 ~# rpm -qa | grep gpfs
    gpfs.gplbin-2.6.32-358.0.1.el6.x86_64-3.5.0-7.x86_64
    gpfs.docs-3.5.0-7.noarch
    gpfs.gpl-3.5.0-7.noarch
    gpfs.gplbin-2.6.32-279.19.1.el6.x86_64-3.5.0-7.x86_64
    gpfs.base-3.5.0-7.x86_64
    gpfs.msg.en_US-3.5.0-7.noarch
    server side
    Mar 18 10:10:16 linuscs109 CNFS: Starting NFS services
    Mar 18 10:10:16 linuscs109 rpc.mountd13516: Kernel does not have pseudo root support.
    Mar 18 10:10:16 linuscs109 rpc.mountd13516: NFS v4 mounts will be disabled unless fsid=0
    Mar 18 10:10:16 linuscs109 rpc.mountd13516: is specfied in /etc/exports file.
    Mar 18 10:10:16 linuscs109 rpc.mountd13518: Version 1.2.3 starting
    Mar 18 10:10:16 linuscs109 kernel: NFSD: Using /var/lib/nfs/v4recovery as the NFSv4 state recovery directory
    Mar 18 10:10:16 linuscs109 kernel: NFSD: starting 90-second grace period
    Mar 18 10:10:21 linuscs109 CNFS: NFS clients of node 10.0.28.59 notified to reclaim NLM locks
    Mar 18 10:10:21 linuscs109 sm-notify14122: Version 1.2.3 starting
    Mar 18 10:10:21 linuscs109 sm-notify14122: Already notifying clients; Exiting!
    Mar 18 10:10:21 linuscs109 CNFS: Monitor has started pid=14158
    Mar 18 10:10:45 linuscs109 kernel: svc: 10.0.28.57, port=796: unknown version (4 for prog 100003, nfsd)
    Mar 18 10:10:45 linuscs109 rpc.mountd13518: authenticated mount request from 10.0.28.57:959 for /vol/aggr0 (/vol/aggr0)
    Mar 18 10:10:45 linuscs109 rpc.mountd13518: internal: no supported addresses in nfs_client
    Mar 18 10:10:45 linuscs109 rpc.mountd13518: getfh failed: Operation not permitted
    Mar 18 10:11:02 linuscs109 kernel: svc: 10.0.28.57, port=776: unknown version (4 for prog 100003, nfsd)
    Mar 18 10:11:02 linuscs109 rpc.mountd13518: authenticated mount request from 10.0.28.57:846 for /vol/aggr0 (/vol/aggr0)
    Mar 18 10:11:02 linuscs109 rpc.mountd13518: internal: no supported addresses in nfs_client
    Mar 18 10:11:02 linuscs109 rpc.mountd13518: getfh failed: Operation not permitted

    client
    root@linuscs107 mnt# mount linuscs109:/vol/aggr0 /mnt/test
    mount.nfs: access denied by server while mounting linuscs109:/vol/aggr0
  • StorageTeamGenomeCenter
    31 Posts

    Re: packages after nfs-utils-1.2.3-15.el6 on rhel breaks CNFS GPFS version 3.5

    ‏2013-03-18T16:40:19Z  
    confirmed something is broke with this combination. The newest kernel seems to fix the problem.

    root@linuscs109 ~# uname -a
    Linux linuscs109 2.6.32-279.19.1.el6.x86_64 #1 SMP Sat Nov 24 14:35:28 EST 2012 x86_64 x86_64 x86_64 GNU/Linux
    root@linuscs109 ~# rpm -qa | grep nfs
    nfs-utils-lib-1.1.5-6.el6.x86_64
    nfs4-acl-tools-0.3.3-6.el6.x86_64
    nfs-utils-1.2.3-36.el6.x86_64
    root@linuscs109 ~# rpm -qa | grep gpfs
    gpfs.gplbin-2.6.32-358.0.1.el6.x86_64-3.5.0-7.x86_64
    gpfs.docs-3.5.0-7.noarch
    gpfs.gpl-3.5.0-7.noarch
    gpfs.gplbin-2.6.32-279.19.1.el6.x86_64-3.5.0-7.x86_64
    gpfs.base-3.5.0-7.x86_64
    gpfs.msg.en_US-3.5.0-7.noarch
    server side
    Mar 18 10:10:16 linuscs109 CNFS: Starting NFS services
    Mar 18 10:10:16 linuscs109 rpc.mountd13516: Kernel does not have pseudo root support.
    Mar 18 10:10:16 linuscs109 rpc.mountd13516: NFS v4 mounts will be disabled unless fsid=0
    Mar 18 10:10:16 linuscs109 rpc.mountd13516: is specfied in /etc/exports file.
    Mar 18 10:10:16 linuscs109 rpc.mountd13518: Version 1.2.3 starting
    Mar 18 10:10:16 linuscs109 kernel: NFSD: Using /var/lib/nfs/v4recovery as the NFSv4 state recovery directory
    Mar 18 10:10:16 linuscs109 kernel: NFSD: starting 90-second grace period
    Mar 18 10:10:21 linuscs109 CNFS: NFS clients of node 10.0.28.59 notified to reclaim NLM locks
    Mar 18 10:10:21 linuscs109 sm-notify14122: Version 1.2.3 starting
    Mar 18 10:10:21 linuscs109 sm-notify14122: Already notifying clients; Exiting!
    Mar 18 10:10:21 linuscs109 CNFS: Monitor has started pid=14158
    Mar 18 10:10:45 linuscs109 kernel: svc: 10.0.28.57, port=796: unknown version (4 for prog 100003, nfsd)
    Mar 18 10:10:45 linuscs109 rpc.mountd13518: authenticated mount request from 10.0.28.57:959 for /vol/aggr0 (/vol/aggr0)
    Mar 18 10:10:45 linuscs109 rpc.mountd13518: internal: no supported addresses in nfs_client
    Mar 18 10:10:45 linuscs109 rpc.mountd13518: getfh failed: Operation not permitted
    Mar 18 10:11:02 linuscs109 kernel: svc: 10.0.28.57, port=776: unknown version (4 for prog 100003, nfsd)
    Mar 18 10:11:02 linuscs109 rpc.mountd13518: authenticated mount request from 10.0.28.57:846 for /vol/aggr0 (/vol/aggr0)
    Mar 18 10:11:02 linuscs109 rpc.mountd13518: internal: no supported addresses in nfs_client
    Mar 18 10:11:02 linuscs109 rpc.mountd13518: getfh failed: Operation not permitted

    client
    root@linuscs107 mnt# mount linuscs109:/vol/aggr0 /mnt/test
    mount.nfs: access denied by server while mounting linuscs109:/vol/aggr0
    Unfortunately this problems still remains with the newest kernel. Once GPFS starts the nfs problem can be fixed with a restart of nfs.

    /etc/init.d/nfs restart

    This problem seems to be in the way GPFS starts nfs.

    Any GPFS developers out there?
  • StorageTeamGenomeCenter
    31 Posts

    Re: packages after nfs-utils-1.2.3-15.el6 on rhel breaks CNFS GPFS version 3.5

    ‏2013-03-18T21:45:24Z  
    Unfortunately this problems still remains with the newest kernel. Once GPFS starts the nfs problem can be fixed with a restart of nfs.

    /etc/init.d/nfs restart

    This problem seems to be in the way GPFS starts nfs.

    Any GPFS developers out there?
    This problem is similar to https://bugzilla.redhat.com/show_bug.cgi?id=164186

    > root@linuscs109 ~# cat /proc/fs/nfs/exports
    > # Version 1.2
    > # Path Client(Flags) # IPs
    > root@linuscs109 ~# exportfs
    > /vol/aggr0 linus225
    > /vol/aggr0 <world>
    > /tmp/test <world>
    > root@linuscs109 ~# exportfs -rv
    > exporting linus225:/vol/aggr0
    > exporting *:/tmp/test
    > exporting *:/vol/aggr0
    > root@linuscs109 ~# cat /proc/fs/nfs/exports
    > # Version 1.2
    > # Path Client(Flags) # IPs
    > root@linuscs109 ~# /etc/init.d/nfs restart
    > Shutting down NFS daemon: OK
    > Shutting down NFS mountd: OK
    > Shutting down NFS quotas: OK
    > Shutting down NFS services: OK
    > Starting NFS services: OK
    > Starting NFS quotas: OK
    > Starting NFS mountd: OK
    > Stopping RPC idmapd: OK
    > Starting RPC idmapd: OK
    > Starting NFS daemon: OK
    > root@linuscs109 ~# cat /proc/fs/nfs/exports
    > # Version 1.2
    > # Path Client(Flags) # IPs
    > /vol/aggr0 *(rw,root_squash,sync,wdelay,no_subtree_check,fsid=13000,uuid=0000ef96:00000001:00000000:00000000)
  • StorageTeamGenomeCenter
    31 Posts

    Re: packages after nfs-utils-1.2.3-15.el6 on rhel breaks CNFS GPFS version 3.5

    ‏2013-03-18T22:10:50Z  
    This problem is similar to https://bugzilla.redhat.com/show_bug.cgi?id=164186

    > root@linuscs109 ~# cat /proc/fs/nfs/exports
    > # Version 1.2
    > # Path Client(Flags) # IPs
    > root@linuscs109 ~# exportfs
    > /vol/aggr0 linus225
    > /vol/aggr0 <world>
    > /tmp/test <world>
    > root@linuscs109 ~# exportfs -rv
    > exporting linus225:/vol/aggr0
    > exporting *:/tmp/test
    > exporting *:/vol/aggr0
    > root@linuscs109 ~# cat /proc/fs/nfs/exports
    > # Version 1.2
    > # Path Client(Flags) # IPs
    > root@linuscs109 ~# /etc/init.d/nfs restart
    > Shutting down NFS daemon: OK
    > Shutting down NFS mountd: OK
    > Shutting down NFS quotas: OK
    > Shutting down NFS services: OK
    > Starting NFS services: OK
    > Starting NFS quotas: OK
    > Starting NFS mountd: OK
    > Stopping RPC idmapd: OK
    > Starting RPC idmapd: OK
    > Starting NFS daemon: OK
    > root@linuscs109 ~# cat /proc/fs/nfs/exports
    > # Version 1.2
    > # Path Client(Flags) # IPs
    > /vol/aggr0 *(rw,root_squash,sync,wdelay,no_subtree_check,fsid=13000,uuid=0000ef96:00000001:00000000:00000000)
    Ok my temp fix was to modify the fsfuncs to replace the exportfs -r with something different. if nfs is set to start during boot this may not be a problem.

    vi /usr/lpp/mmfs/bin/mmnfsfuncs

    nfsService() {
    -f $sysnfs && . $sysnfs

    case $1 in
    start)
    msg "Starting NFS services"
    sysctl -e -q -w fs.nfs.use_underlying_lock_ops=1 2> /dev/null

    # Reload exportfs anyway since starting nfs server may not do this
    # if it was already running
    #exportfs -r
    /etc/init.d/nfs restart
  • StorageTeamGenomeCenter
    31 Posts

    Re: packages after nfs-utils-1.2.3-15.el6 on rhel breaks CNFS GPFS version 3.5

    ‏2013-03-19T15:23:27Z  
    Ok my temp fix was to modify the fsfuncs to replace the exportfs -r with something different. if nfs is set to start during boot this may not be a problem.

    vi /usr/lpp/mmfs/bin/mmnfsfuncs

    nfsService() {
    -f $sysnfs && . $sysnfs

    case $1 in
    start)
    msg "Starting NFS services"
    sysctl -e -q -w fs.nfs.use_underlying_lock_ops=1 2> /dev/null

    # Reload exportfs anyway since starting nfs server may not do this
    # if it was already running
    #exportfs -r
    /etc/init.d/nfs restart
    All,

    The init script will also fail as it uses exportfs.

    If you add a new export to /etc/exports and do a "/etc/init.d/nfs relaod" it will not export the new volume.

    reload | force-reload)
    /usr/sbin/exportfs -r
    -f /var/lock/subsys/nfs && touch /var/lock/subsys/nfs
    ;;

    Can anyone reproduce this or are we the only ones with this problem.

    should be able to reproduce it with version nfs-utils-1.2.3-15.el6_2.1 to
    current.
  • StorageTeamGenomeCenter
    31 Posts

    Re: packages after nfs-utils-1.2.3-15.el6 on rhel breaks CNFS GPFS version 3.5

    ‏2013-05-13T20:59:11Z  
    All,

    The init script will also fail as it uses exportfs.

    If you add a new export to /etc/exports and do a "/etc/init.d/nfs relaod" it will not export the new volume.

    reload | force-reload)
    /usr/sbin/exportfs -r
    -f /var/lock/subsys/nfs && touch /var/lock/subsys/nfs
    ;;

    Can anyone reproduce this or are we the only ones with this problem.

    should be able to reproduce it with version nfs-utils-1.2.3-15.el6_2.1 to
    current.

    defect# 876886

    This is fixed in version 3.5.0.10

           This is due to a bug fix in RHEL 6.3 that change the order of
           mountd and nfsd startup

    looks like they just added a check for version >=6 then it does not unmount /proc/fs/nfsd