Allan,
there seems to be (at least one) bug in corosync configuration checker.
Your configuration is missing nodeid parameter. Also make sure to NOT
specify mcastaddr, because it will ends up with message you've described
(configuration which you've pasted doesn't contain it, so that's correct).
Honza
Allan Latham napsal(a):
> Hi all
>
> just to let you know that our first series of tests works fine if I
> substitute the ipv4 address in the same script.
>
> Does anyone actually use corosync live on ipv6?
>
> Is staying with ipv4 the right solution?
>
> All the best
>
> Allan
>
> On 25/06/13 16:43, Allan Latham wrote:
>> Hi
>>
>> We have used corosync (and pacemaker) successfully some years in an IPV4
>> environment.
>>
>> Now we should be using IPV6 for the next project but this is proving
>> difficult. I presume I have not understood the configuration but I have
>> searched high and low to find examples.
>>
>> Results with Debian Wheezy version 1.4.2 and self compiled 1.4.6
>> versions are identical:
>>
>> 2013-06-25_14:23:34.87618 Jun 25 14:23:34 corosync [MAIN ] Successfully
>> read main configuration file '/etc/corosync/corosync.conf'.
>> 2013-06-25_14:23:34.87706 Jun 25 14:23:34 corosync [TOTEM ] Initializing
>> transport (UDP/IP Unicast).
>> 2013-06-25_14:23:34.87707 Jun 25 14:23:34 corosync [TOTEM ] Initializing
>> transmit/receive security: libtomcrypt SOBER128/SHA1HMAC (mode 0).
>> 2013-06-25_14:23:34.98701 corosync: totemip.c:104: totemip_equal:
>> Assertion `addrlen' failed.
>>
>> Here is the config file (IP addresses partly hidden):
>>
>> Of course I have also tried with abbreviated ip6 addresses (no leading 0
>> and :: for consecutive :0:0: groups)
>>
>> I even tried without any : at all. Same result every time.
>>
>> totem {
>> version: 2
>> nodeid: 1 # A
>> secauth: on
>> vsftype: none
>> token: 4000
>> consensus: 7000
>> token_retransmits_before_loss_const: 20
>> window_size: 20
>> max_messages: 20
>> transport: udpu
>> interface {
>> member {
>> memberaddr: 2aff:ffff:0131:0203:0000:0000:0001:0002 # A
>> }
>> member {
>> memberaddr: 2aff:ffff:0100:40eb:0000:0000:0001:0002 # B
>> }
>> member {
>> memberaddr: 2aff:ffff:0d16:2d90:0000:0000:0000:0002 # C
>> }
>> ringnumber: 0
>> bindnetaddr: 2aff:ffff:0131:0203:0000:0000:0001:0002 # A
>> mcastport: 694
>> }
>> }
>> amf {
>> mode: disabled
>> }
>>
>> service {
>> # Load the Pacemaker Cluster Resource Manager
>> ver: 0
>> name: pacemaker
>> }
>>
>> aisexec {
>> user: root
>> group: root
>> }
>>
>> logging {
>> fileline: off
>> to_syserr: yes
>> debug: off
>> timestamp: on
>> logger_subsys {
>> subsys: AMF
>> debug: off
>> }
>> }
>>
>> If anyone can tell me where I'm going wrong I'd be very thankful.
>>
>> Best regards to all
>>
>> Allan
>> _______________________________________________
>> discuss mailing list
>> discuss@xxxxxxxxxxxx
>> http://lists.corosync.org/mailman/listinfo/discuss
>>
>
> _______________________________________________
> discuss mailing list
> discuss@xxxxxxxxxxxx
> http://lists.corosync.org/mailman/listinfo/discuss
>
_______________________________________________
discuss mailing list
discuss@xxxxxxxxxxxx
http://lists.corosync.org/mailman/listinfo/discuss
