Re: problems getting started

OK, last night i tried the 1.0.4 from CVS and got MUCH further.

I was able to define a node, cluster, resource, resource_group and failover_policy. I didn't try the log stuff.

One thing has got me puzzled. Sometimes the CLI wouldn't work for some tasks so i had to use the GUI, and sometimes the GUI would lock up but the CLI would work. In the end i found myself having to use both to achieve some tasks. They may have been a function of not knowing what i was doing ;)

i got to the point where i could run 'start ha_services on ... in ...' After that 'status node A' and 'status cluster C' would return up or active.

The next step is to actually DO something with all of this. Does any one have a worked example of setting up failsafe to keep apache running should it crash ? 

Is there a HOWTO somewhere ?

Scott Fagg <[email protected]>
Arup Brisbane
(07) 3023 6000


>>> Lars Marowsky-Bree <[email protected]> 13/6/02 6:48:29 pm >>>
On 2002-06-13T08:41:55,
   Scott Fagg <[email protected]> said:

> Is there a quick howto somewhere on the basics of getting a failsafe system
> up and running ? I only want to try some simple setups such as a single
> node, using failsafe to restart a crashed app or two nodes with a single
> service and have failsafe move the app from node1 to node2 in the event of
> app or os failure on node1

That should be straightforward.

> I've tried 1.0.2 RPMs , 1.0.3 source RPMSs and 1.0.4 from CVS. I've grabbed the sysadm_* rpms from sourceforge. The logs below are from 1.0.3.

What is your distribution? 1.0.3 will only work with glibc <= 2.2.2; I would
suggest using CVS version. The error you have logs very likely to be the glibc
incompatibility.



Sincerely,
    Lars Marowsky-Brée <[email protected]>

-- 
Immortality is an adequate definition of high availability for me.
	--- Gregory F. Pfister

_______________________________________________
LinuxFailSafe mailing list
[email protected] 
http://lists.community.tummy.com/mailman/listinfo/linuxfailsafe