looks like memory corruption in the heartbeat process
you should file a bug report for that and mark it blocker
On Nov 26, 2007, at 10:43 PM, Audet, Jean-Michel wrote:
When I crash, I get the following messages in my /var/log/messages
heartbeat: ERROR: ipc_bufpool_update: magic number in head does not
match.Something very bad happened, abort now, fearside pid 27029
heartbeat: ERROR: magic=616d6a2f, expected=abcd
heartbeat: info: pool: refcount=1, startpos=0x9ad4ba0,
currpos=0x9ad515a, consumerpos=0x9ad50ba, endpos=0x9ad6b88, size 4096
heartbeat: info: nmsg=0
attrd: CRIT: attrd_ha_dispatch: Lost connection to heartbeat service.
JM
-----Message d'origine-----
De : Audet, Jean-Michel
Envoyé : Monday, November 26, 2007 4:30 PM
À : 'General Linux-HA mailing list'
Objet : RE: [Linux-HA] Heartbeat vs CIB
Hi again,
I am trying to merge the code from crm_mon since Friday and I am
always hitting the wall on function set_working_set_defaults.
My code simply crash there and stop.
CRM_MON application compile and work fine. Any idea what can cause
the problem?
Thanks again,
Jean-Michel Audet
-----Message d'origine-----
De : [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
] De la part de Andrew Beekhof
Envoyé : Monday, November 26, 2007 1:49 PM
À : General Linux-HA mailing list
Objet : Re: [Linux-HA] Heartbeat vs CIB
On Nov 26, 2007, at 5:47 PM, Audet, Jean-Michel wrote:
Thanks for the feedback.
I had already used the walking node functions:
Here is the kind of information I am able to get:
MyNode1 type=normal
MyNode1 status=active
MyNode2 type=normal
MyNode2 status=active
192.168.0.1 type=ping
MyNode1 status=ping
However, I would like to know which one is the Master (Hot) node
(Re. The node that currently the DC).
DC status is irrelevant. it doesn't indicate the node has any
resources.
if you want to know what's running where, try looking at the crm_mon
program
Thanks!
JM
-----Message d'origine-----
De : [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
] De la part de Andrew Beekhof
Envoyé : Monday, November 26, 2007 10:17 AM
À : General Linux-HA mailing list
Objet : Re: [Linux-HA] Heartbeat vs CIB
On Nov 23, 2007, at 3:25 PM, Audet, Jean-Michel wrote:
I am currently using Heartbeat v2 with a cib.xml file.
I need an Active/Standby (Master/Slave) configuration
I am a newbie and I need to connect my software to the heartbeat
daemons
using the client library and then do check pointing between my 2
applications to get the standby node in sync.
I found some docs and examples on the hb-api interface. I connect
successfully using ll_cluster_new, signon, etc... However, I don't
know
how to know the current state of a node.
look for do_node_walk() in tools/pingd.c
I found from the crm_mon application source that there is also
another
API to manage and work with nodes (cib.h). It looks like the kind
interface is similar but not exactly the same.
the cib is probably not what you need
What are you suggesting for a "new" development? Which API? Like I
said, I need an Active/Standby implementation and I need to send
data
from my application to do check pointing. I need to be able to
retrieve
the current state of my node. I need to be able to do cooperative
switch-over.
Thanks!
Jean-Michel Audet
_______________________________________________
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
_______________________________________________
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
_______________________________________________
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
