On Thu, Aug 28, 2008 at 15:40, Gianluca Cecchi <[EMAIL PROTECTED]> wrote: > On Thu, Aug 28, 2008 at 2:23 PM, Lars Marowsky-Bree <[EMAIL PROTECTED]> wrote: > >> >> Set the "resource-stickiness" meta_attribute to "infinity"; you can do >> this for the whole cluster with the crm_config setting of >> "default-resource-stickiness" - the GUI provides access to these, or >> use: >> >> crm_attribute -v INFINITY -t crm_config -n default-resource-stickiness >> >> > Ok. I tried and it worked after changing config from command line as > indicated. Thanks > What is the correct file/files to check for these parameters and all their > possible values?
type: /usr/lib/heartbeat/pengine metadata for a list of available options, or read the docs at: http://clusterlabs.org/mw/Image:Configuration_Explained.pdf > I found the /usr/share/heartbeat/crm.dtd but I don't know if it covers all > what needed... > In particular, what could be the sense of negative values for > default-resource-stickiness or even -INFINITY (resources will always move > away from their current location)? > > A question about config propagation: is it supposed that one run these kind > of commands (as the crm_attribute above) only when all the nodes are up and > running? As long as one node is running, it should all work. > Or by design a cluster node that was down wil receive and correctly > merge the config stored in its file system with the current one? In these > days tests it seemed to me to get strange behaviours in this case..... but I > could be wrong... > Could I for example safely change a resource configuration from a node > command line while another node is down and then silently/smoothly start the > latter up again to get it transparently applied ? > > Gianluca > _______________________________________________ > Linux-HA mailing list > [email protected] > http://lists.linux-ha.org/mailman/listinfo/linux-ha > See also: http://linux-ha.org/ReportingProblems > _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
