Hi All. I have a setup in which I have two servers serving nfs share. The nfs service is made highly available with pacemaker. When the primary server goes down the secondary starts nfs service. Service IP is floating between servers but they have NO "shared" storage/filesystem so NFS state/connection information in case of failover is lost. I have two clients. When the failover from primary to secondary occurs the mount is stale and I need to manually remount the share. Is there a way in linux/CentOS to automatically remount nfs share in such case? Or should I just write a script which (for example) check /proc/mounts and execute it from crontab? I am curious if it can be done with "standard" linux services (automounter?) ;) Best regards, Rafal Radecki.
Am 20.02.2013 um 11:29 schrieb Rafa? Radecki <radecki.rafal at gmail.com>:> Hi All. > > I have a setup in which I have two servers serving nfs share. The nfs > service is made highly available with pacemaker. When the primary > server goes down the secondary starts nfs service. Service IP is > floating between servers but they have NO "shared" storage/filesystem > so NFS state/connection information in case of failover is lost. I > have two clients. When the failover from primary to secondary occurs > the mount is stale and I need to manually remount the share. > Is there a way in linux/CentOS to automatically remount nfs share in > such case? Or should I just write a script which (for example) check > /proc/mounts and execute it from crontab? I am curious if it can be > done with "standard" linux services (automounter?) ;)what is the order of the resources nfs and ip? -- LF
Why not use DRBD in lieu of shared storage? On Wed, Feb 20, 2013 at 2:52 AM, Leon Fauster <leonfauster at googlemail.com>wrote:> Am 20.02.2013 um 11:29 schrieb Rafa? Radecki <radecki.rafal at gmail.com>: > > Hi All. > > > > I have a setup in which I have two servers serving nfs share. The nfs > > service is made highly available with pacemaker. When the primary > > server goes down the secondary starts nfs service. Service IP is > > floating between servers but they have NO "shared" storage/filesystem > > so NFS state/connection information in case of failover is lost. I > > have two clients. When the failover from primary to secondary occurs > > the mount is stale and I need to manually remount the share. > > Is there a way in linux/CentOS to automatically remount nfs share in > > such case? Or should I just write a script which (for example) check > > /proc/mounts and execute it from crontab? I am curious if it can be > > done with "standard" linux services (automounter?) ;) > > > > what is the order of the resources nfs and ip? > > -- > LF > > _______________________________________________ > CentOS mailing list > CentOS at centos.org > http://lists.centos.org/mailman/listinfo/centos >