Send Linux-ha-cvs mailing list submissions to
[email protected]
To subscribe or unsubscribe via the World Wide Web, visit
http://lists.community.tummy.com/mailman/listinfo/linux-ha-cvs
or, via email, send a message with subject or body 'help' to
[EMAIL PROTECTED]
You can reach the person managing the list at
[EMAIL PROTECTED]
When replying, please edit your Subject line so it is more specific
than "Re: Contents of Linux-ha-cvs digest..."
Today's Topics:
1. Linux-HA CVS: include by andrew from
([email protected])
2. Linux-HA CVS: lib by andrew from
([email protected])
3. Linux-HA CVS: crm by andrew from
([email protected])
4. Linux-HA CVS: crm by andrew from
([email protected])
----------------------------------------------------------------------
Message: 1
Date: Tue, 10 Jan 2006 06:29:20 -0700 (MST)
From: [email protected]
Subject: [Linux-ha-cvs] Linux-HA CVS: include by andrew from
To: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
linux-ha CVS committal
Author : andrew
Host :
Project : linux-ha
Module : include
Dir : linux-ha/include/clplumbing
Modified Files:
ipc.h
Log Message:
Expose socket_disconnect() to the world.
Sometimes its useful to close an IPC channel without destroying its buffers.
Signed off by Alan: "You're right. I guess we need to export the disconnect
function through the IPC_OPS interface."
===================================================================
RCS file: /home/cvs/linux-ha/linux-ha/include/clplumbing/ipc.h,v
retrieving revision 1.54
retrieving revision 1.55
diff -u -3 -r1.54 -r1.55
--- ipc.h 15 Oct 2005 02:46:24 -0000 1.54
+++ ipc.h 10 Jan 2006 13:29:20 -0000 1.55
@@ -1,4 +1,4 @@
-/* $Id: ipc.h,v 1.54 2005/10/15 02:46:24 gshi Exp $ */
+/* $Id: ipc.h,v 1.55 2006/01/10 13:29:20 andrew Exp $ */
/*
* ipc.h IPC abstraction data structures.
*
@@ -586,7 +586,9 @@
*/
int (*get_conntype)(struct IPC_CHANNEL* ch);
-
+
+ int (*disconnect)(struct IPC_CHANNEL* ch);
+
};
------------------------------
Message: 2
Date: Tue, 10 Jan 2006 06:29:20 -0700 (MST)
From: [email protected]
Subject: [Linux-ha-cvs] Linux-HA CVS: lib by andrew from
To: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
linux-ha CVS committal
Author : andrew
Host :
Project : linux-ha
Module : lib
Dir : linux-ha/lib/clplumbing
Modified Files:
ipcsocket.c
Log Message:
Expose socket_disconnect() to the world.
Sometimes its useful to close an IPC channel without destroying its buffers.
Signed off by Alan: "You're right. I guess we need to export the disconnect
function through the IPC_OPS interface."
===================================================================
RCS file: /home/cvs/linux-ha/linux-ha/lib/clplumbing/ipcsocket.c,v
retrieving revision 1.165
retrieving revision 1.166
diff -u -3 -r1.165 -r1.166
--- ipcsocket.c 8 Jan 2006 21:47:36 -0000 1.165
+++ ipcsocket.c 10 Jan 2006 13:29:20 -0000 1.166
@@ -1,4 +1,4 @@
-/* $Id: ipcsocket.c,v 1.165 2006/01/08 21:47:36 lars Exp $ */
+/* $Id: ipcsocket.c,v 1.166 2006/01/10 13:29:20 andrew Exp $ */
/*
* ipcsocket unix domain socket implementation of IPC abstraction.
*
@@ -2485,4 +2485,5 @@
is_sendq_full: socket_is_sendq_full,
is_recvq_full: socket_is_recvq_full,
get_conntype: socket_get_conntype,
+ disconnect: socket_disconnect,
};
------------------------------
Message: 3
Date: Tue, 10 Jan 2006 06:46:41 -0700 (MST)
From: [email protected]
Subject: [Linux-ha-cvs] Linux-HA CVS: crm by andrew from
To: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
linux-ha CVS committal
Author : andrew
Host :
Project : linux-ha
Module : crm
Dir : linux-ha/crm/pengine
Modified Files:
unpack.c
Log Message:
Timerless elections
- required nodes to send "no-votes" when they loose.
- nodes can no longer ignore their own votes or do vote compression
- elections are owned and unique. we must recieve 1 no-vote from each
node in the cluster before continuing
Have all shutdowns go via the PE
- fixes the SimulStop case where resources were stopped out of order
- required "shutdown" status to no longer be considered during elections
- the TE needs to be invoked even if the DC is shutting down
- DC's shut themselves down and the cluster can elect a new one if it needs to
XML_CIB_ATTR_SHUTDOWN is now checked for non-zero instead of NULL to decide if
a node wants to shut down.
- means we can do away with the XML_CIB_ATTR_CLEAR_SHUTDOWN and the special
code for it in the CIB
===================================================================
RCS file: /home/cvs/linux-ha/linux-ha/crm/pengine/unpack.c,v
retrieving revision 1.151
retrieving revision 1.152
diff -u -3 -r1.151 -r1.152
--- unpack.c 9 Jan 2006 18:45:00 -0000 1.151
+++ unpack.c 10 Jan 2006 13:46:41 -0000 1.152
@@ -1,4 +1,4 @@
-/* $Id: unpack.c,v 1.151 2006/01/09 18:45:00 andrew Exp $ */
+/* $Id: unpack.c,v 1.152 2006/01/10 13:46:41 andrew Exp $ */
/*
* Copyright (C) 2004 Andrew Beekhof <[EMAIL PROTECTED]>
*
@@ -489,18 +489,19 @@
crm_element_value(node_state, XML_CIB_ATTR_INCCM);
const char *ha_state =
crm_element_value(node_state, XML_CIB_ATTR_HASTATE);
- const char *shutdown =
- crm_element_value(node_state, XML_CIB_ATTR_SHUTDOWN);
-
+ int shutdown = 0;
+
if(this_node == NULL) {
pe_config_err("No node to check");
return online;
}
+ ha_msg_value_int(node_state, XML_CIB_ATTR_SHUTDOWN, &shutdown);
+
if(safe_str_eq(exp_state, CRMD_JOINSTATE_MEMBER)) {
this_node->details->expected_up = TRUE;
}
- if(shutdown != NULL) {
+ if(shutdown != 0) {
this_node->details->shutdown = TRUE;
#if 0
this_node->details->expected_up = FALSE;
------------------------------
Message: 4
Date: Tue, 10 Jan 2006 06:46:41 -0700 (MST)
From: [email protected]
Subject: [Linux-ha-cvs] Linux-HA CVS: crm by andrew from
To: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
linux-ha CVS committal
Author : andrew
Host :
Project : linux-ha
Module : crm
Dir : linux-ha/crm/tengine
Modified Files:
unpack.c
Log Message:
Timerless elections
- required nodes to send "no-votes" when they loose.
- nodes can no longer ignore their own votes or do vote compression
- elections are owned and unique. we must recieve 1 no-vote from each
node in the cluster before continuing
Have all shutdowns go via the PE
- fixes the SimulStop case where resources were stopped out of order
- required "shutdown" status to no longer be considered during elections
- the TE needs to be invoked even if the DC is shutting down
- DC's shut themselves down and the cluster can elect a new one if it needs to
XML_CIB_ATTR_SHUTDOWN is now checked for non-zero instead of NULL to decide if
a node wants to shut down.
- means we can do away with the XML_CIB_ATTR_CLEAR_SHUTDOWN and the special
code for it in the CIB
===================================================================
RCS file: /home/cvs/linux-ha/linux-ha/crm/tengine/unpack.c,v
retrieving revision 1.51
retrieving revision 1.52
diff -u -3 -r1.51 -r1.52
--- unpack.c 19 Dec 2005 16:54:44 -0000 1.51
+++ unpack.c 10 Jan 2006 13:46:41 -0000 1.52
@@ -1,4 +1,4 @@
-/* $Id: unpack.c,v 1.51 2005/12/19 16:54:44 andrew Exp $ */
+/* $Id: unpack.c,v 1.52 2006/01/10 13:46:41 andrew Exp $ */
/*
* Copyright (C) 2004 Andrew Beekhof <[EMAIL PROTECTED]>
*
@@ -240,6 +240,7 @@
gboolean
extract_event(crm_data_t *msg)
{
+ int shutdown = 0;
const char *event_node = NULL;
struct abort_blob_s
{
@@ -328,7 +329,9 @@
}
}
- if(crm_element_value(node_state, XML_CIB_ATTR_SHUTDOWN) !=
NULL) {
+ shutdown = 0;
+ ha_msg_value_int(node_state, XML_CIB_ATTR_SHUTDOWN, &shutdown);
+ if(shutdown != 0) {
crm_info("Aborting on "XML_CIB_ATTR_SHUTDOWN"
attribute");
if(blob.text == NULL) {
blob.text = "Aborting on
"XML_CIB_ATTR_SHUTDOWN" attribute";
------------------------------
_______________________________________________
Linux-ha-cvs mailing list
[email protected]
http://lists.community.tummy.com/mailman/listinfo/linux-ha-cvs
End of Linux-ha-cvs Digest, Vol 26, Issue 11
********************************************
