Re: [Linux-HA] Heartbeat in Amazon VMs doest not create virtaul ip address
- Original Message - Hi, I installed on HeartBeat,Centos 6.5 on 2 Amazon EC2 machinesthis is the If you have an option, I'd strongly recommend using the Pacemaker+CMAN stack in rhel 6.5. Red Hat began supporting pacemaker in 6.5, so it should be available to you. -- Vossel version: [root@ip-10-0-2-68 ha.d]# rpm -qa | grep heartbeat heartbeat-libs-3.0.4-2.el6.x86_64 heartbeat-3.0.4-2.el6.x86_64 heartbeat-devel-3.0.4-2.el6.x86_64 the floating IP is [root@ip-10-0-2-68 ha.d]# cat haresources ip-10-0-2-68 10.0.2.70 but it is not created on any machine, it does not matter where I do the takeover or standby commands what am I missing? is this even possible ? these are my setting in ha.cf logfacility local0 ucast eth0 10.0.2.69 auto_failback on node ip-10-0-2-68 ip-10-0-2-69 ping 10.0.2.1 use_logd yes logfacility local0 ucast eth0 10.0.2.68 auto_failback on node ip-10-0-2-68 ip-10-0-2-69 ping 10.0.2.1 use_logd yes these is the output of the route command [root@ip-10-0-2-68 ha.d]# route -n Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 10.0.2.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0 0.0.0.0 10.0.2.1 0.0.0.0 UG 0 0 0 eth0 [root@ip-10-0-2-68 ha.d]# this is how the interfaces eth0 are set up on machine 1:[root@ip-10-0-2-68 ha.d]# ifconfig eth0 Link encap:Ethernet HWaddr 12:23:49:EF:3A:53 inet addr:10.0.2.68 Bcast:10.0.2.255 Mask:255.255.255.0 inet6 addr: fe80::1023:49ff:feef:3a53/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:9001 Metric:1 RX packets:269823 errors:0 dropped:0 overruns:0 frame:0 TX packets:192305 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:167802149 (160.0 MiB) TX bytes:48341828 (46.1 MiB) Interrupt:247 these are the logs showing everything going on fine but when doing ifconfig the interface is not there: Nov 11 21:37:39 ip-10-0-2-68 heartbeat[14681]: [14681]: WARN: node ip-10-0-2-69: is dead Nov 11 21:37:39 ip-10-0-2-68 heartbeat[14681]: [14681]: info: Comm_now_up(): updating status to active Nov 11 21:37:39 ip-10-0-2-68 heartbeat[14681]: [14681]: info: Local status now set to: 'active' Nov 11 21:37:39 ip-10-0-2-68 heartbeat[14681]: [14681]: WARN: No STONITH device configured. Nov 11 21:37:39 ip-10-0-2-68 heartbeat[14681]: [14681]: WARN: Shared disks are not protected. Nov 11 21:37:39 ip-10-0-2-68 heartbeat[14681]: [14681]: info: Resources being acquired from ip-10-0-2-69. Nov 11 21:37:39 ip-10-0-2-68 mach_down(default)[14769]: info: /usr/share/heartbeat/mach_down: nice_failback: foreign resources acquired Nov 11 21:37:39 ip-10-0-2-68 heartbeat[14681]: [14681]: info: mach_down takeover complete. Nov 11 21:37:39 ip-10-0-2-68 heartbeat[14681]: [14681]: info: Initial resource acquisition complete (mach_down) Nov 11 21:37:39 ip-10-0-2-68 /usr/lib/ocf/resource.d//heartbeat/IPaddr(IPaddr_10.0.2.70)[14845]: INFO: Resource is stopped Nov 11 21:37:39 ip-10-0-2-68 heartbeat[14701]: [14701]: info: Local Resource acquisition completed. Nov 11 21:37:40 ip-10-0-2-68 /usr/lib/ocf/resource.d//heartbeat/IPaddr(IPaddr_10.0.2.70)[14958]: INFO: Resource is stopped Nov 11 21:37:40 ip-10-0-2-68 IPaddr(IPaddr_10.0.2.70)[15057]: INFO: /usr/libexec/heartbeat/send_arp -i 200 -r 5 -p /var/run/resource-agents/send_arp-10.0.2.70 eth0 10.0.2.70 auto not_used not_used Nov 11 21:37:40 ip-10-0-2-68 /usr/lib/ocf/resource.d//heartbeat/IPaddr(IPaddr_10.0.2.70)[15064]: INFO: Success Nov 11 21:37:49 ip-10-0-2-68 heartbeat[14681]: [14681]: info: Local Resource acquisition completed. (none) Nov 11 21:37:49 ip-10-0-2-68 heartbeat[14681]: [14681]: info: local resource transition completed. Nov 11 21:38:16 ip-10-0-2-69 heartbeat: [18350]: WARN: node ip-10-0-2-68: is dead Nov 11 21:38:16 ip-10-0-2-69 heartbeat: [18350]: info: Comm_now_up(): updating status to active Nov 11 21:38:16 ip-10-0-2-69 heartbeat: [18350]: info: Local status now set to: 'active' Nov 11 21:38:16 ip-10-0-2-69 heartbeat: [18350]: WARN: No STONITH device configured. Nov 11 21:38:16 ip-10-0-2-69 heartbeat: [18350]: WARN: Shared disks are not protected. Nov 11 21:38:16 ip-10-0-2-69 heartbeat: [18350]: info: Resources being acquired from ip-10-0-2-68. Nov 11 21:38:16 ip-10-0-2-69 heartbeat: [18360]: info: No local resources [/usr/share/heartbeat/ResourceManager listkeys ip-10-0-2-69] to acquire. Nov 11 21:38:17 ip-10-0-2-69 /usr/lib/ocf/resource.d//heartbeat/IPaddr(IPaddr_10.0.2.70)[18441]: INFO: Resource is stopped Nov 11 21:38:17 ip-10-0-2-69 IPaddr(IPaddr_10.0.2.70)[18537]: INFO: /usr/libexec/heartbeat/send_arp -i 200 -r 5 -p /var/run/resource-agents/send_arp-10.0.2.70 eth0 10.0.2.70 auto not_used not_used Nov 11 21:38:17 ip-10-0-2-69
Re: [Linux-HA] MySQL resource agent and MySQL Master/Slave replication
- Original Message - Dear guys, I have set up a MySQL Master/Slave replication and tried to make it HA using corosync/pacemaker. I used the Mysql Resource Agent currently provided by Debian Wheezy and set up the corosync primitives as mentioned in the examples[1]. When migrating the mysql resource from one node to the other, the mysql resource agent didn't handle the slaves at all. As I inspected the code of the mysql resource agent, I figured out, that all the slave handling is done in the mysql_notify function of the resource agent. To enable the multi state resource ms_mysql to send notifications to the mysql resource agent, I added: ms ms_mysql p_mysql \ meta clone-max=3 \ *meta notify=true* Now the MySQL replication slaves get started/stopped as expected. You may add this parameter to your documentation[2], so other users don't have to figure it out themselves :-) good observation, thanks for the feedback! -- Vossel cheers Tom -- unixum Tom Hutter Hufnertwiete 3 D-22305 Hamburg Telefon: +49 40 67 39 21 52 mobil : +49 174 400 24 16 E-Mail: tom.hut...@unixum.de www.unixum.de Steuer-Nr.: 43/102/01125 Geschäftsführung: Tom Hutter [1] http://www.linux-ha.org/wiki/Mysql_%28resource_agent%29 [2] http://www.linux-ha.org/wiki/Mysql_%28resource_agent%29#MySQL_master.2Fslave_replication ___ Linux-HA mailing list Linux-HA@lists.linux-ha.org http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems ___ Linux-HA mailing list Linux-HA@lists.linux-ha.org http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
