rpower man page states: "don’t try to run wake against the ’on’ state node, it would cause the node gets to ’off’ state."
but it seems better to have our code check whether the status is on and if so return and error for the wake command rather than sending the wake command to a node that is already on.
Please evaluate if we should do this. I am moving to 2.8.5 for now.
Ticket moved from /p/xcat/bugs/3892/
Can't be converted:
Jarrod fixed this problem in 2.8.5 and 2.9.
2.8.5:commit 321e6696f89b7ac8852c44d316c38bcbe8350d47
2.9: commit 6e9d88a8428fb5edfb6bc605c10da3542730604c