Re: [Openais] delayed shutdown

2009-04-16 Thread Chrissie Caulfield
Fabio M. Di Nitto wrote: On Wed, 2009-04-15 at 14:35 -0500, David Teigland wrote: If I run 'cman_tool leave' on four nodes in parallel, node1 will leave right away, but the other three nodes don't leave until the token timeout expires for node1 causing a confchg for it, after which the other

Re: [Openais] Patch for WhiteTank to prevent shared memory leak

2009-04-16 Thread Jérôme Flesch
This patch is definitively about fixing a bug due to a BSD-ism. My test case: On a FreeBSD system: - a client connects to Aisexec and creates the SHM - the client does its job - the client disconnects, the SHM is not destroyed by Aisexec After some debugging, I've found out that, at the

Re: [Openais] howto distribute data accross all nodes?

2009-04-16 Thread Dietmar Maurer
Check out: http://www.openais.org/doku.php?id=dev:paritition_recovery Instead of using the token callback method, you could write your own methodology for executing the state machine. Ah, OK - I think that is what I already do. What is miss is something like a 'merge' function? Seems the

[Openais] Partition Recovery and CPG

2009-04-16 Thread Dietmar Maurer
Lest assume the cluster is partitioned: Part1: node1 node2 node3 Part2: node4 node5 After recovery, what join/leave messaged do I receive with a CPG: A.) JOIN: node4 node5 or B.) JOIN: node1 node2 node3 or anything else? - Dietmar ___ Openais

Re: [Openais] howto distribute data accross all nodes?

2009-04-16 Thread Steven Dake
On Thu, 2009-04-16 at 08:20 +0200, Dietmar Maurer wrote: You might try taking a look at exec/sync.c it is a synchronization engine. Basically it takes configuration changes into account to call sync_init, sync_process, sync_abort, or sync_activate. These 4 states then activate the new

Re: [Openais] Partition Recovery and CPG

2009-04-16 Thread David Teigland
On Thu, Apr 16, 2009 at 12:38:19PM +0200, Dietmar Maurer wrote: Lest assume the cluster is partitioned: Part1: node1 node2 node3 Part2: node4 node5 After recovery, what join/leave messaged do I receive with a CPG: A.) JOIN: node4 node5 or B.) JOIN: node1 node2 node3 or anything

Re: [Openais] [RFC] corosync/trunk: logsys v3 implementation take 3

2009-04-16 Thread Steven Dake
Great work! comments that should be addressed: logsys.h: please no single line comments, especially in header files. can you multiline this or put it in the comment above the definitions? /* FORK and THREADED are ignored for SUBSYSTEMS */ How would you feel about changing LOG_MODE_* to

Re: [Openais] Partition Recovery and CPG

2009-04-16 Thread Robert Wipfel
On 4/16/2009 at 11:29 AM, in message 20090416172927.ge21...@redhat.com, David Teigland teigl...@redhat.com wrote: On Thu, Apr 16, 2009 at 12:38:19PM +0200, Dietmar Maurer wrote: Lest assume the cluster is partitioned: Part1: node1 node2 node3 Part2: node4 node5 After recovery, what

[Openais] [PATCH openais trunk] AMF: allow a component to start before openais

2009-04-16 Thread angus salkeld
This patch allows a component to start before AMF and poll the saAmfComponetRegister() function. -Angus Index: services/amf.c === --- services/amf.c (revision 1791) +++ services/amf.c (working copy) @@ -1843,7 +1843,16 @@

Re: [Openais] [RFC] corosync/trunk: logsys v3 implementation take 3

2009-04-16 Thread Fabio M. Di Nitto
Hi Steven, On Thu, 2009-04-16 at 16:28 -0700, Steven Dake wrote: Great work! comments that should be addressed: logsys.h: please no single line comments, especially in header files. can you multiline this or put it in the comment above the definitions? Absolutely not a problem. I'll