HP serviceguard t2808-90006 manual # cmrunpkg packagename

Disaster Scenarios and Their Handling
Table 4-1	Disaster Scenarios and Their Handling (Continued)
Disaster Scenario		What Happens When	Recovery Process
		This Disaster Occurs
In this case, the package (P1)		The package (P1) continues	In this scenario, no attempts are
runs with RPO_TARGET set to 60		to run on N1 with md0	made to repair the first failure
seconds.		consisting of only S1 after	until the second failure occurs.
Package P1 is running on node		the first failure	Complete the following procedure
			to initiate a recovery:
N1. P1 uses a mirror md0		After the second failure, the
consisting of S1 (local to node		package does not start up	1. To recover from the first
N1, for example		on N2 because when it tries	failure, restore the FC links
/dev/hpdev/mylink-sde) and		to start with only S2 on N2,
			between the data centers. As a
S2 (local to node N2). The first		it detects that S2 is
			result, S1 is accessible from
failure occurs when all FC links		non-current for a time
			N2.
between two data centers fail,		period which is greater than
			2. After the FC links are
causing N1 to lose access to S2		the value of RPO_TARGET.
and N2 to lose access to S1.			restored, and S1 is accessible
After the package resumes			from N2, run the following
			command to restart the
activity and runs for 90 seconds,
			package on N2.
a second failure occurs causing
node N1 to fail.			# cmrunpkg <package_name>
			When the package starts up on
			N2, it automatically adds S1 back
			into the array and starts
			re-mirroring from S1 to S2. When
			re-mirroring is complete, the
			extended distance cluster detects
			and accepts S1 as part of md0
			again.
			For the second failure, restore N1.
			Once it is restored, it joins the
			cluster and can access S1 and S2.
			1. Run the following command to
			enable P1 to run on N1:
			# cmmodpkg -e P1 -n N1

92 Chapter 4