[Linux-HA] resource group not moved to available node

[Andrew Beekhof]

On 3/7/07, Yan Fitterer <yfitterer at novell.com> wrote:
> > It not only "failed" on idm01, it failed to start. A start failure means
> > the node is not considered eligible to run the resource until you use
> > crm_resource - C.
>
> OK this is fine - by why does it not try to run on idm02 after failing on idm04 ?

according to ptest, idm02 has a score of -100 and is therefor unavailable

ptest[27281]: 2007/03/08_14:40:59 debug: native_print: Allocating:
R_meta_master_disk   (heartbeat:evms_failover-tc):   Stopped
ptest[27281]: 2007/03/08_14:40:59 debug: native_assign_node: Color
R_meta_master_disk, Node[0] idm02: -100
ptest[27281]: 2007/03/08_14:40:59 debug: native_assign_node: Color
R_meta_master_disk, Node[1] idm04: -500
ptest[27281]: 2007/03/08_14:40:59 debug: native_assign_node: Color
R_meta_master_disk, Node[2] idm01: -1000000
ptest[27281]: 2007/03/08_14:40:59 debug: native_assign_node: Color
R_meta_master_disk, Node[3] idm03: -1000000
ptest[27281]: 2007/03/08_14:40:59 debug: native_assign_node: Even
highest ranked node for R_meta_master_disk, had weight -100
ptest[27281]: 2007/03/08_14:40:59 WARN: native_color: Resource
R_meta_master_disk cannot run anywhere


btw. i found:
ptest[24777]: 2007/03/08_14:33:20 ERROR: unpack_rsc_order: Constraint
OR_userapp_wf: no resource found for RHS of group_userapp