請問大俠如何配置fence 才能斷網實現節點切換, 我的cluser.conf如下: <?xml version="1.0"?> <cluster alias="test_cluster" config_version="29" name="test_cluster"> <fence_daemon post_fail_delay="0" post_join_delay="3"/> <clusternodes> <clusternode name="beg75.ex.com" nodeid="1" votes="1"> <fence> <method name="1"> <device domain="test70" name="xenfence"/> </method> </fence> </clusternode> <clusternode name="test70.ex.com" nodeid="2" votes="1"> <fence> <method name="1"> <device domain="beg75" name="xenfence"/> </method> </fence> </clusternode> </clusternodes> <cman expected_votes="1" two_node="1"/> <fencedevices> <fencedevice agent="fence_xvm" name="xenfence"/> </fencedevices> <fencedevice agent="fence_xvm" name="xenfence"/> </fencedevices> <rm> <failoverdomains> <failoverdomain name="webinfo" ordered="1" restricted="0"> <failoverdomainnode name="beg75.ex.com" priority="1"/> <failoverdomainnode name="test70.ex.com" priority="2"/> </failoverdomain> </failoverdomains> <resources> <ip address="192.168.0.100" monitor_link="0"/> <script file="/etc/init.d/httpd" name="apache"/> <fs device="/dev/sdb1" force_fsck="0" force_unmount="0" fsid="42848" fstype="ext3" mountpoint="/data" name="webdata" options="" self_fence="0"/> </resources> <service autostart="1" domain="webinfo" name="www"> <script ref="apache"> <ip ref="192.168.0.100"/> </script> <fs ref="webdata"/> </service> </rm> </cluster> 在斷網后啟動fence 也失敗,log如下: r 16 11:55:19 beg75 openais: Initialising service handler 'openais checkpoint service B.01.01' Mar 16 11:55:19 beg75 openais: Initialising service handler 'openais event service B.01.01' Mar 16 11:55:19 beg75 openais: Initialising service handler 'openais distributed locking service B.01.01' Mar 16 11:55:19 beg75 openais: Initialising service handler 'openais message service B.01.01' Mar 16 11:55:19 beg75 openais: Initialising service handler 'openais configuration service' Mar 16 11:55:19 beg75 openais: Initialising service handler 'openais cluster closed process group service v1.01' Mar 16 11:55:19 beg75 openais: Initialising service handler 'openais CMAN membership service 2.01' Mar 16 11:55:19 beg75 openais: CMAN 2.0.73 (built Sep 19 2007 16:04:02) started Mar 16 11:55:19 beg75 openais: Not using a virtual synchrony filter. Mar 16 11:55:19 beg75 openais: Creating commit token because I am the rep. Mar 16 11:55:19 beg75 openais: Saving state aru 0 high seq received 0 Mar 16 11:55:19 beg75 openais: Storing new sequence id for ring 204 Mar 16 11:55:19 beg75 openais: entering COMMIT state. Mar 16 11:55:19 beg75 openais: entering RECOVERY state. Mar 16 11:55:19 beg75 openais: position member 192.168.0.75: Mar 16 11:55:19 beg75 openais: previous ring seq 512 rep 192.168.0.75 Mar 16 11:55:19 beg75 openais: aru 0 high delivered 0 received flag 1 Mar 16 11:55:19 beg75 openais: Did not need to originate any messages in recovery. Mar 16 11:55:19 beg75 openais: Sending initial ORF token Mar 16 11:55:19 beg75 openais: CLM CONFIGURATION CHANGE Mar 16 11:55:19 beg75 openais: New Configuration: Mar 16 11:55:19 beg75 openais: Members Left: Mar 16 11:55:19 beg75 openais: Members Joined: Mar 16 11:55:19 beg75 openais: CLM CONFIGURATION CHANGE Mar 16 11:55:19 beg75 openais: New Configuration: Mar 16 11:55:19 beg75 openais: r(0) ip(192.168.0.75) Mar 16 11:55:19 beg75 openais: Members Left: Mar 16 11:55:19 beg75 openais: Members Joined: Mar 16 11:55:19 beg75 openais: r(0) ip(192.168.0.75) Mar 16 11:55:19 beg75 openais: This node is within the primary component and will provide service. Mar 16 11:55:19 beg75 openais: entering OPERATIONAL state. Mar 16 11:55:19 beg75 openais: quorum regained, resuming activity Mar 16 11:55:19 beg75 openais: got nodejoin message 192.168.0.75 Mar 16 11:55:19 beg75 openais: entering GATHER state from 11. Mar 16 11:55:19 beg75 openais: Saving state aru 9 high seq received 9 Mar 16 11:55:19 beg75 openais: Storing new sequence id for ring 208 Mar 16 11:55:19 beg75 openais: entering COMMIT state. Mar 16 11:55:19 beg75 openais: entering RECOVERY state. Mar 16 11:55:19 beg75 openais: position member 192.168.0.70: Mar 16 11:55:19 beg75 openais: previous ring seq 516 rep 192.168.0.70 Mar 16 11:55:19 beg75 openais: aru c high delivered c received flag 1 Mar 16 11:55:19 beg75 openais: position member 192.168.0.75: Mar 16 11:55:19 beg75 openais: previous ring seq 516 rep 192.168.0.75 Mar 16 11:55:19 beg75 openais: aru 9 high delivered 9 received flag 1 Mar 16 11:55:19 beg75 openais: Did not need to originate any messages in recovery. Mar 16 11:55:19 beg75 openais: CLM CONFIGURATION CHANGE Mar 16 11:55:19 beg75 openais: New Configuration: Mar 16 11:55:19 beg75 openais: r(0) ip(192.168.0.75) Mar 16 11:55:19 beg75 openais: Members Left: Mar 16 11:55:19 beg75 openais: Members Joined: Mar 16 11:55:19 beg75 openais: CLM CONFIGURATION CHANGE Mar 16 11:55:19 beg75 openais: New Configuration: Mar 16 11:55:19 beg75 openais: r(0) ip(192.168.0.70) Mar 16 11:55:19 beg75 openais: r(0) ip(192.168.0.75) Mar 16 11:55:19 beg75 openais: Members Left: Mar 16 11:55:19 beg75 openais: Members Joined: Mar 16 11:55:19 beg75 openais: r(0) ip(192.168.0.70) Mar 16 11:55:19 beg75 openais: This node is within the primary component and will provide service. Mar 16 11:55:19 beg75 openais: entering OPERATIONAL state. Mar 16 11:55:19 beg75 openais: got nodejoin message 192.168.0.70 Mar 16 11:55:19 beg75 openais: got nodejoin message 192.168.0.75 Mar 16 11:55:19 beg75 openais: got joinlist message from node 2 Mar 16 11:55:19 beg75 groupd: found uncontrolled kernel object rgmanager in /sys/kernel/dlm Mar 16 11:55:19 beg75 groupd: local node must be reset to clear 1 uncontrolled instances of gfs and/or dlm Mar 16 11:55:19 beg75 openais: cman killed by node 1 because we were killed by cman_tool or other application Mar 16 11:55:21 beg75 fenced: cman_init error 0 111 Mar 16 11:55:22 beg75 dlm_controld: cman_init error 0 111 Mar 16 11:55:22 beg75 gfs_controld: cman_init error 111 ar 16 11:55:49 beg75 ccsd: Unable to connect to cluster infrastructure after 780 seconds. Mar 16 11:55:50 beg75 fence_node: agent "fence_xvm" reports: Adding IP 127.0.0.1 to list (family 2) Adding IP 192.168.0.75 to list (family 2) Adding IP 192.168.0.100 to list (family 2) ipv4_listen: Setting up ipv4 listen socket ipv4_listen: Success; fd = 2 Setting up ipv4 multicast send (225.0.0.12:1229) Joining IP Mar 16 11:55:50 beg75 fence_node: agent "fence_xvm" reports: Multicast group (pass 1) Joining IP Multicast group (pass 2) Setting TTL to 2 for fd5 ipv4_send_sk: success, fd = 5 Setting up ipv4 multicast send (225.0.0.12:1229) Joining IP Multicast group (pass 1) Joining IP Multicast group (pass 2) Setting TTL to 2 Mar 16 11:55:50 beg75 fence_node: agent "fence_xvm" reports: for fd5 ipv4_send_sk: success, fd = 5 Setting up ipv4 multicast send (225.0.0.12:1229) Joining IP Multicast group (pass 1) Joining IP Multicast group (pass 2) Setting TTL to 2 for fd5 ipv4_send_sk: success, fd = 5 Setting up ipv4 multicast send (225.0.0.1 Mar 16 11:55:50 beg75 fence_node: agent "fence_xvm" reports: 2:1229) Joining IP Multicast group (pass 1) Joining IP Multicast group (pass 2) Setting TTL to 2 for fd5 ipv4_send_sk: success, fd = 5 Setting up ipv4 multicast send (225.0.0.12:1229) Joining IP Multicast group (pass 1) Joining IP Multicast group (pass 2)