Thank you all for your help ! I've got one more HORCM related question after reading chapter 2.8.5 "Command Device" of CCI release 27. According to this chapter if we lost command device the worst could happen with system is we won't be able to send commands to SAN. However when we removed both fibers of paired HBAs used to communicate with command device in couple of minutes and multiple core dumps node crashed. When we did same with pair of HBAs which aren't used to communicate with command device everything quite smoothly failed over to standby node.
Is this another case of "works as designed, but probably not as desired" or could it be a HORCM bug ? Thanks, Sergei -- This message posted from opensolaris.org
