[Cluster-devel] SAP resource agents

[Cluster-devel] SAP resource agents

[Mark Hlawatschek]

Hi,

we have made another step to improve the way how to run SAP on a Red Hat high 
availability cluster. 

As a result, you'll find the latest versions of the following resource agents 
and event-scripts attached:
Many thanks to Alexander Krauth for providing the SAP resource agents.

- SAPDatabase:
The purpose of the resource agent is to start, stop and monitor the database 
instance of a SAP system.

- SAPInstance:
Manages a single SAP Instance as a High-Availability resource. 

- follow-service.sl
Implements the "follow service" logic based on rgmanagers RIND event 
mechanism. This is needed to implement a replicated enqueue failover 
scenario.

Mark

-- 
Gruss / Regards,

Dipl.-Ing. Mark Hlawatschek
http://www.atix.de/
http://www.open-sharedroot.org/
-------------- next part --------------
% follow-service.sl
%
% Description:	Implements the "follow service" mechanism based on the Red Hat RIND event 
%               scripting mechanism.
%
% Author:       Marc Grimme, Mark Hlawatschek, October 2008
% Support:      support at atix.de
% License:      GNU General Public License (GPL)
% Copyright:    (c) 2008 ATIX AG


debug("*** follow-service.sl");


%
% Returns a list of nodes for the given service that are online and in the failoverdomain.
%
define nodelist_online(service_name) {
   variable nodes, nofailback, restricted, ordered, node_list;
   nodes=nodes_online();
   
   (nofailback, restricted, ordered, node_list) = service_domain_info(service_name);
   
   return intersection(nodes, node_list);
}

%
% Idea: 
%   General purpose function of a construct when Service(svc1) and Service(svc2) 
%   should not be running on the same node even after failover.
%   There are to options to influence the behaviour. If both services have to be 
%   running on the same node (only one node is left in the failovergroup) what 
%   service is the master and should both services be running or only the master
%   service survives. If master is not svc1 or svc2 both service might run on the 
%   same node. If master is either svc1 or svc2 the specified one will be the 
%   surviving service.
%   If followslave is not 0 the svc1 always follows svc2. That means it will be 
%   started on on the same node as svc1. And if available svc2 will be relocated
%   to any other node.
%
define follow_service(svc1, svc2, master) %, followslave)
{
	variable state, owner_svc1, owner_svc2;
	variable nodes1, nodes2, allowed;

	debug("*** FOLLOW_SERVICE: follow_service(",svc1,", ",svc2,", ", master, ")");
	debug("*** FOLLOW_SERVICE: event_type: ", event_type, ", service_name: ", service_name, ", service_state: ", service_state);

	%
	% setup the master
	%
	if ((master != svc1) and (master != svc2)) {
		debug("*** FOLLOW_SERVICE: master=NULL");
		master=NULL;
	}

	% get infos we need to decide further
	(owner_svc1, state) = service_status(svc1);
	(owner_svc2, state) = service_status(svc2);
	nodes1 = nodelist_online(svc1);
	nodes2 = nodelist_online(svc2);
	debug("*** FOLLOW_SERVICE: service_status(",svc1,"): ", service_status(svc1));
	debug("*** FOLLOW_SERVICE: owner_svc1: ", owner_svc1, ", owner_svc2: ", owner_svc2, ", nodes1: ", nodes1, ", nodes2: ", nodes2);

	if (((event_type == EVENT_NODE)    and (owner_svc1 == node_id) and (node_state == NODE_OFFLINE) and (owner_svc2 >=0)) or 
		((event_type == EVENT_SERVICE) and (service_name == svc1)  and (service_state == "recovering" ) and (owner_svc2 >= 0))) {
		%
		% uh oh, the owner of the master server died.  Restart it
		% on the node running the slave server or if we should not 
		% follow the slave start it somewhere else.
		% We should end up here if svc1 has to be restarted

		%
		% If this was a service event, don't execute the default event
		% script trigger after this script completes.
		%
		if (event_type == EVENT_SERVICE) {
			stop_processing();
		}
		% were to start svc2
		allowed=subtract(nodes2, owner_svc2);
		if (length(allowed) > 1) {
			allowed=subtract(allowed, service_last_owner);
		}
		debug("*** FOLLOW SERVICE: service event triggered following svc2 to ",owner_svc2, " svc2 on : ",allowed);

		% either svc1 is the master or there are node were to start svc2
		if ((master == svc1) or (length(allowed) > 0)) {
			()=service_start(svc1, owner_svc2);
		}
		% either svc2 is the master or there are node were to start svc2
		if ((master == svc2) or (length(allowed) > 0)) {
			()=service_stop(svc2);
        	()=service_start(svc2, allowed);
		} 
	}
	else if (((event_type == EVENT_NODE) and (owner_svc2 == node_id) and (node_state == NODE_OFFLINE) and (owner_svc2 >=0)) or 
		((event_type == EVENT_SERVICE) and (service_name == svc2) and (service_state == "recovering" ) and (owner_svc1 >= 0))) {
		%
		% uh oh, the owner of the svc2 died.  Restart it
		% on any other node but not the one running the svc1.
		% If svc1 is the only one left only start it there 
		% if master==svc2
		%
		% Just relocate svc2 or if svc2 is master stop svc1 and start svc2 on owner_svc1

		%
		% If this was a service event, don't execute the default event
		% script trigger after this script completes.
		%
	  
		if (event_type == EVENT_SERVICE) {
			stop_processing();
		}

		allowed=subtract(nodes2, owner_svc1);
		if (length(allowed) > 1) {
			allowed=subtract(allowed, service_last_owner);
		}

		debug("*** FOLLOW SERVICE: service event triggered relocating svc2 to ",allowed, " svc1 on : ",owner_svc1);

		if (length(allowed) > 0) {
			()=service_stop(svc2);
			()=service_start(svc2, allowed);
		} else if (master == svc2) {
			()=service_stop(svc1);
			()=service_start(svc2, owner_svc1);
		}
	}
	else if (((event_type == EVENT_SERVICE) and (service_state == "started") and (owner_svc2 == owner_svc1) and (owner_svc1 > 0) and (owner_svc2 > 0)) or
    		((event_type == EVENT_CONFIG) and (owner_svc2 == owner_svc1))) {
		allowed=subtract(nodes2, owner_svc1);
		debug("*** FOLLOW SERVICE: service event both running on same node triggered.", allowed);
		if (length(allowed) > 0) {
			%()=service_stop(svc1);
			%()=service_start(svc1, owner_svc2);
			()=service_stop(svc2);
			()=service_start(svc2, allowed);
		} else if ((master == svc2) and (owner_svc2 > 0)){
			debug("*** FOLLOW SERVICE: will stop service .", svc1); 
			()=service_stop(svc1);
		} else if ((master == svc1) and (owner_svc1 > 0)) {
			debug("*** FOLLOW SERVICE: will stop service .", svc2);
			()=service_stop(svc2);
		} else {
			debug("*** FOLLOW SERVICE: both services running on the same node or only one is running.", allowed, ", ", master);
		}
	}
	return;
}
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SAPInstance
Type: application/x-shellscript
Size: 18245 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/cluster-devel/attachments/20081112/4bb7e565/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SAPDatabase
Type: application/x-shellscript
Size: 27244 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/cluster-devel/attachments/20081112/4bb7e565/attachment-0001.bin>

[Cluster-devel] SAP resource agents

[Fabio M. Di Nitto]

Hi Mark,

On Wed, 12 Nov 2008, Mark Hlawatschek wrote:

> Hi,
>
> we have made another step to improve the way how to run SAP on a Red Hat high
> availability cluster.

thanks for the submission.

>
> As a result, you'll find the latest versions of the following resource agents
> and event-scripts attached:

A few things tho...

> Many thanks to Alexander Krauth for providing the SAP resource agents.
>
> - SAPDatabase:
> The purpose of the resource agent is to start, stop and monitor the database
> instance of a SAP system.
>
> - SAPInstance:
> Manages a single SAP Instance as a High-Availability resource.

on what platform have those scripts been tested?

> - follow-service.sl
> Implements the "follow service" logic based on rgmanagers RIND event
> mechanism. This is needed to implement a replicated enqueue failover
> scenario.

depending on what branch this should land, we need to look into the 
licence. master/stable2 are now gpl2+. New submission should respect that 
licence. Updating old bits (like the SAP* files is still ok with the old 
licence for now).

I'd also like to have a proper diff on top of the branch with explanation 
of the changes. Some of them are related to defaults and depending where 
they should be applied, it might be important to review them all together.

Fabio

--
I'm going to make him an offer he can't refuse.