Hi,

On Fri, Nov 30, 2007 at 11:05:29AM +0100, China wrote:
> Hi, I've a problem with a new Linux-HA installation.
> I've a test environment with a hub and 2 Mini Mac (1 ppc and 1 intel) with a
> new installation Fedora core 7.
> HA is installed in both Mac using source package (tgz) through "ConfigureMe
> rpm; rpm -i"
> 
> Each Mac has two eth interfaces:
> eth0 "export" services and is connected with the hub
> eth1 is for heartbeat messages and is connected with the eth1 of the other
> node
> 
> The configuration files are:
> 
> HARESOURCES
> test-ppc 192.168.122.8/24 httpd
> 
> CIB.XML
> deriving from haresources
> 
> HA.CF
> bcast eth1
> node test-ppc test
> respawn root /rnd/apps/components/heartbeat/lib/heartbeat/pingd -m 100 -d 5s
> ping_group test 192.168.122.100
> crm yes
> 
> AUTHKEYS
> auth 3
> 1 crc
> 2 md5 mypwd
> 3 sha1 mypwd
> 
> When I start the cluster I've a lot of this messages in test log file:
> WARN: string2msg_ll: node [test-ppc] failed authentication
> 
> and a lot of this messages in test-ppc log file:
> WARN: string2msg_ll: node [test] failed authentication
> 
> I've tried with ucast, but the result is the same.
> Instead with crc and md5 all is OK. So I think it is a problem related sha.

Looks like a bug. Can you open a bugzilla for it.

> How can I resolve this?

Use md5. Or do you insist on sha1?

Thanks,

Dejan

> Thanks in advance.
> 
> -- 
> 
> Davide Belloni
> _______________________________________________
> Linux-HA mailing list
> [email protected]
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Reply via email to