[Pacemaker] fence_xvm / fence_virtd problem

Vladislav Bogdanov bubble at hoster-ok.com
Sat Jun 15 22:04:22 UTC 2013


15.06.2013 20:26, Digimer wrote:
> Ah, I think it's a problem with the firewall rules on the host. Not sure
> how to fix it though...

You probably need to open port 1229/tcp in filter INPUT on virtual
cluster members. That is where fence_xvm listens for a connection from
fence_virtd after it sends multicast request iirc. Design deficiency,
fence_xvm works only one copy on a system at a time, it listens for tcp
connection on a predefined port.

> 
> lemass:/home/digimer# iptables-save
> # Generated by iptables-save v1.4.16.2 on Sat Jun 15 13:26:33 2013
> *nat
> :PREROUTING ACCEPT [246583:89552160]
> :INPUT ACCEPT [2335:362026]
> :OUTPUT ACCEPT [11740:741351]
> :POSTROUTING ACCEPT [11706:738225]
> -A POSTROUTING -s 192.168.122.0/24 ! -d 192.168.122.0/24 -p tcp -j
> MASQUERADE --to-ports 1024-65535
> -A POSTROUTING -s 192.168.122.0/24 ! -d 192.168.122.0/24 -p udp -j
> MASQUERADE --to-ports 1024-65535
> -A POSTROUTING -s 192.168.122.0/24 ! -d 192.168.122.0/24 -j MASQUERADE
> COMMIT
> # Completed on Sat Jun 15 13:26:33 2013
> # Generated by iptables-save v1.4.16.2 on Sat Jun 15 13:26:33 2013
> *mangle
> :PREROUTING ACCEPT [3250861:2486027770]
> :INPUT ACCEPT [2557761:1301267981]
> :FORWARD ACCEPT [444644:1094901100]
> :OUTPUT ACCEPT [1919457:2636518995]
> :POSTROUTING ACCEPT [2364615:3731498365]
> -A POSTROUTING -o virbr0 -p udp -m udp --dport 68 -j CHECKSUM
> --checksum-fill
> COMMIT
> # Completed on Sat Jun 15 13:26:33 2013
> # Generated by iptables-save v1.4.16.2 on Sat Jun 15 13:26:33 2013
> *filter
> :INPUT ACCEPT [2557761:1301267981]
> :FORWARD ACCEPT [0:0]
> :OUTPUT ACCEPT [1919457:2636518995]
> -A INPUT -i virbr0 -p udp -m udp --dport 53 -j ACCEPT
> -A INPUT -i virbr0 -p tcp -m tcp --dport 53 -j ACCEPT
> -A INPUT -i virbr0 -p udp -m udp --dport 67 -j ACCEPT
> -A INPUT -i virbr0 -p tcp -m tcp --dport 67 -j ACCEPT
> -A FORWARD -d 192.168.122.0/24 -o virbr0 -m conntrack --ctstate
> RELATED,ESTABLISHED -j ACCEPT
> -A FORWARD -s 192.168.122.0/24 -i virbr0 -j ACCEPT
> -A FORWARD -i virbr0 -o virbr0 -j ACCEPT
> -A FORWARD -o virbr0 -j REJECT --reject-with icmp-port-unreachable
> -A FORWARD -i virbr0 -j REJECT --reject-with icmp-port-unreachable
> COMMIT
> # Completed on Sat Jun 15 13:26:33 2013
> 
> digimer
> 
> On 06/15/2013 01:09 PM, Digimer wrote:
>> Hi all,
>>
>>    I'm trying to play with pacemaker on fedora 19 (pre-release) and I am
>> having trouble getting the guests to talk to the host.
>>
>>  From the host, I can run;
>>
>> lemass:/home/digimer# fence_xvm -o list
>> pcmk1                83f6abdc-bb48-d794-4aca-13f091f32c8b on
>> pcmk2                2d778455-de7d-a9fa-994c-69d7b079fda8 on
>>
>> I can fence the guests from the host as well. However, I can not get the
>> list (or fence) from the quests;
>>
>> [root at pcmk1 ~]# fence_xvm -o list
>> Timed out waiting for response
>> Operation failed
>>
>> I suspect a multicast issue, but so far as I can tell, multicast is
>> enabled on the bridge;
>>
>> lemass:/home/digimer# ifconfig
>> virbr0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
>>          inet 192.168.122.1  netmask 255.255.255.0  broadcast
>> 192.168.122.255
>>          ether 52:54:00:da:90:a1  txqueuelen 0  (Ethernet)
>>          RX packets 103858  bytes 8514464 (8.1 MiB)
>>          RX errors 0  dropped 0  overruns 0  frame 0
>>          TX packets 151988  bytes 177742562 (169.5 MiB)
>>          TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
>>
>> vnet0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
>>          inet6 fe80::fc54:ff:feed:3701  prefixlen 64  scopeid 0x20<link>
>>          ether fe:54:00:ed:37:01  txqueuelen 500  (Ethernet)
>>          RX packets 212828  bytes 880551892 (839.7 MiB)
>>          RX errors 0  dropped 0  overruns 0  frame 0
>>          TX packets 225430  bytes 182955760 (174.4 MiB)
>>          TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
>>
>> vnet1: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
>>          inet6 fe80::fc54:ff:fe45:e9ae  prefixlen 64  scopeid 0x20<link>
>>          ether fe:54:00:45:e9:ae  txqueuelen 500  (Ethernet)
>>          RX packets 4840  bytes 587902 (574.1 KiB)
>>          RX errors 0  dropped 0  overruns 0  frame 0
>>          TX packets 7495  bytes 899578 (878.4 KiB)
>>          TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
>>
>> I tried specifying the mcast address and port without success.
>>
>> The host's config is:
>>
>> lemass:/home/digimer# cat /etc/fence_virt.conf
>> backends {
>>      libvirt {
>>          uri = "qemu:///system";
>>      }
>>
>> }
>>
>> listeners {
>>      multicast {
>>          port = "1229";
>>          family = "ipv4";
>>          interface = "virbr0";
>>          address = "239.192.214.190";
>>          key_file = "/etc/cluster/fence_xvm.key";
>>      }
>>
>> }
>>
>> fence_virtd {
>>      module_path = "/usr/lib64/fence-virt";
>>      backend = "libvirt";
>>      listener = "multicast";
>> }
>>
>> The cluster forms and corosync is using multicast, so I am not sure if
>> mcast really is the problem.
>>
>> Any tips/help?
>>
>> Thanks!
>>
> 
> 





More information about the Pacemaker mailing list