Peter Thümmel wrote:
Hi,
we have 2 Servers, intend to run an failsave (Pro)Ftp_server.
DRBD is running fine over an ethernet link.
When i disconnect a server the other server should go to primary-Status.
The drbd resource (V2) is installed an run fine.
The pingd is cloned and run also fine.
Please help me with the config of the rcs_location, that the drbd following the
pingd (get to secondary on the disconnectet server and go to the primary-state
on the other server.)
In my setup, a node with connectivity gets a score bonus of 5000:
Read my rsc_location:
<rsc_location id="my_resource_connected" rsc="ms-drbd3">
<rule id="my_resource_connected_rule" score="5000"
boolean_op="and" role="master">
<expression id="my_resource_connected_expr_undefined"
attribute="pingd" operation="defined"/>
<expression id="my_resource_connected_expr_zero"
attribute="pingd" operation="ne" value="0"/>
</rule>
</rsc_location>
This means, if the master node looses ping connectivity, its score for
the master role is decreased by 5000, which is always lower than the
score of the other node (which hopefully still has the 5000 bonus due to
connectivity). That way, a failover is initiated.
Before you use my score of 5000, you might want to look at your scores
and set a proper value for your environment. Use the showscores.sh
script from pacemaker to display your scores.
Regards
Dominik
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
