Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Hi! I am try to use two nodes active active drbd (8.4.2) configuration with pacemaker(1.1.10,1.1.9 ). The drbd is main resource in my cluster, so there is the first resource in order,and other resources is co-located on drbd. The two cluster nodes starts without any problem, and resource is correctly migrates to a on-line node, when I put another one to standby mode. But if I put standby node on-line the drbd ocf resource try to demote drbd, so it stop co-located resources ( such as virtual domain ) before. I think, this is not what I want from cluster. I look in ocf source file and find this code in drbd_start() function drbd_start() { .... $OCF_RUNNING_MASTER) ocf_log warn "$DRBD_RESOURCE already Primary, demoting." do_drbdadm secondary $DRBD_RESOURCE .... } So I have some questions for developers: 1. Is it nessesary to put drbd in secondary mode if it run as master? 2. Can this part of code stop my resources when i return standby node? online? 3. What can happend if I remove demoting code? I think it must be like this drbd_start() { .... $OCF_RUNNING_MASTER) ocf_log warn "$DRBD_RESOURCE already Primary." rc=$OCF_SUCCESS break ;; ... } I shall try this, may be next night (to prevent rebooting my virtual servers in work time). Thanks for drbd. Bye!