glusterfs

History

Kaleb S KEITHLEY 40a24f5ab9 common-ha: reliable grace using pacemaker notify actions

Using *-dead_ip-1 resources to track on which nodes the ganesha.nfsd
had died was found to be unreliable.

Running `pcs status` in the ganesha_grace monitor action was seen to
time out during failover; the HA devs opined that it was, generally,
not a good idea to run `pcs status` in a monitor action in any event.
They suggested using the notify feature, where the resources on all
the nodes are notified when a clone resource agent dies.

This change adds a notify action to the ganesha_grace RA. The ganesha_mon
RA monitors its ganesha.nfsd daemon. While the daemon is running, it
creates two attributes: ganesha-active and grace-active. When the daemon
stops for any reason, the attributes are deleted. Deleting the
ganesha-active attribute triggers the failover of the virtual IP (the
IPaddr RA) to another node where ganesha.nfsd is still running. The
ganesha_grace RA monitors the grace-active attribute. When the
grace-active attibute is deleted, the ganesha_grace RA stops, and will
not restart. This triggers pacemaker to trigger the notify action in
the ganesha_grace RAs on the other nodes in the cluster; which send a
DBUS message to their ganesha.nfsd.

(N.B. grace-active is a bit of a misnomer. while the grace-active
attribute exists, everything is normal and healthy. Deleting the
attribute triggers putting the surviving ganesha.nfsds into GRACE.)

To ensure that the remaining/surviving ganesha.nfsds are put into
NFS-GRACE before the IPaddr (virtual IP) fails over there is a short
delay (sleep) between deleting the grace-active attribute and the
ganesha-active attribute. To summarize:
  1. on node 2 ganesha_mon:monitor notices that ganesha.nfsd has died
  2. on node 2 ganesha_mon:monitor deletes its grace-active attribute
  3. on node 2 ganesha_grace:monitor notices that grace-active is gone
     and returns OCF_ERR_GENERIC, a.k.a. new error. When pacemaker
     tries to (re)start ganesha_grace, its start action will return
     OCF_NOT_RUNNING, a.k.a. known error, don't attempt further
     restarts.
  4. on nodes 1, 3, etc., ganesha_grace:notify receives a post-stop
     notification indicating that node 2 is gone, and sends a DBUS
     message to its ganesha.nfsd putting it into NFS-GRACE.
  5. on node 2 ganesha_mon:monitor waits a short period, then deletes
     its ganesha-active attribute. This triggers the IPaddr (virt IP)
     failover according to constraint location rules.

ganesha_nfsd modified to run for the duration, start action is invoked
to setup the /var/lib/nfs symlink, stop action is invoked to restore it.
ganesha-ha.sh modified accordingly to create it as a clone resource.

BUG: 1290865
Change-Id: I1ba24f38fa4338b3aeb17c65645e9f439387ff57
Signed-off-by: Kaleb S KEITHLEY <kkeithle@redhat.com>
Reviewed-on: http://review.gluster.org/12964
Smoke: Gluster Build System <jenkins@build.gluster.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.com>
Reviewed-on: http://review.gluster.org/13725

2016-03-14 21:34:27 -07:00

debugging

doc : Steps to analize coredumps generated by regression machines.

2015-06-29 08:14:29 -07:00

developer-guide

docs: move contributor docs to the glusterdocs repository

2015-11-15 05:19:37 -08:00

features

common-ha: reliable grace using pacemaker notify actions

2016-03-14 21:34:27 -07:00

gluster.8

gluster man page: adding geo-replication chapter