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: crm by andrew from
([email protected])
2. Linux-HA CVS: resources by lars from
([email protected])
----------------------------------------------------------------------
Message: 1
Date: Tue, 23 May 2006 04:00:32 -0600 (MDT)
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:
complex.c incarnation.c
Log Message:
Remove redundant code
===================================================================
RCS file: /home/cvs/linux-ha/linux-ha/crm/pengine/complex.c,v
retrieving revision 1.90
retrieving revision 1.91
diff -u -3 -r1.90 -r1.91
--- complex.c 23 May 2006 07:45:37 -0000 1.90
+++ complex.c 23 May 2006 10:00:31 -0000 1.91
@@ -1,4 +1,4 @@
-/* $Id: complex.c,v 1.90 2006/05/23 07:45:37 andrew Exp $ */
+/* $Id: complex.c,v 1.91 2006/05/23 10:00:31 andrew Exp $ */
/*
* Copyright (C) 2004 Andrew Beekhof <[EMAIL PROTECTED]>
*
@@ -212,7 +212,6 @@
if(value) {
(*rsc)->id = crm_concat(id, value, ':');
add_hash_param((*rsc)->meta, XML_RSC_ATTR_INCARNATION, value);
- add_hash_param((*rsc)->parameters,
crm_meta_name(XML_RSC_ATTR_INCARNATION), value);
} else {
(*rsc)->id = crm_strdup(id);
===================================================================
RCS file: /home/cvs/linux-ha/linux-ha/crm/pengine/incarnation.c,v
retrieving revision 1.89
retrieving revision 1.90
diff -u -3 -r1.89 -r1.90
--- incarnation.c 22 May 2006 15:31:30 -0000 1.89
+++ incarnation.c 23 May 2006 10:00:31 -0000 1.90
@@ -1,4 +1,4 @@
-/* $Id: incarnation.c,v 1.89 2006/05/22 15:31:30 andrew Exp $ */
+/* $Id: incarnation.c,v 1.90 2006/05/23 10:00:31 andrew Exp $ */
/*
* Copyright (C) 2004 Andrew Beekhof <[EMAIL PROTECTED]>
*
@@ -99,7 +99,6 @@
clone_data->child_list, child_rsc);
add_hash_param(child_rsc->meta, XML_RSC_ATTR_INCARNATION_MAX, inc_max);
- add_hash_param(child_rsc->parameters,
crm_meta_name(XML_RSC_ATTR_INCARNATION_MAX), inc_max);
print_resource(LOG_DEBUG_3, "Added", child_rsc, FALSE);
------------------------------
Message: 2
Date: Tue, 23 May 2006 10:48:56 -0600 (MDT)
From: [email protected]
Subject: [Linux-ha-cvs] Linux-HA CVS: resources by lars from
To: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
linux-ha CVS committal
Author : lars
Host :
Project : linux-ha
Module : resources
Dir : linux-ha/resources/OCF
Modified Files:
Filesystem.in
Log Message:
Perform the mount for OCFS2 at start-time, now that the notify data is
available with the actual operation too, and process pre-notifications
for start (Novell #177525).
===================================================================
RCS file: /home/cvs/linux-ha/linux-ha/resources/OCF/Filesystem.in,v
retrieving revision 1.22
retrieving revision 1.23
diff -u -3 -r1.22 -r1.23
--- Filesystem.in 22 May 2006 11:03:41 -0000 1.22
+++ Filesystem.in 23 May 2006 16:48:56 -0000 1.23
@@ -1,6 +1,6 @@
#!/bin/sh
#
-# $Id: Filesystem.in,v 1.22 2006/05/22 11:03:41 andrew Exp $
+# $Id: Filesystem.in,v 1.23 2006/05/23 16:48:56 lars Exp $
#
# Support: [EMAIL PROTECTED]
# License: GNU General Public License (GPL)
@@ -95,7 +95,7 @@
usage() {
cat <<-EOT
usage: $0 {start|stop|status|monitor|validate-all|meta-data}
- $Id: Filesystem.in,v 1.22 2006/05/22 11:03:41 andrew Exp $
+ $Id: Filesystem.in,v 1.23 2006/05/23 16:48:56 lars Exp $
EOT
}
@@ -201,16 +201,18 @@
#
Filesystem_start()
{
- if [ "$FSTYPE" = "ocfs2" ] && [ -z "$OCFS2_DO_MOUNT" ]; then
- # Sorry, start doesn't actually do anything here. Magic
- # happens in Filesystem_notify; see the comment there.
- ocf_log debug "$DEVICE: ocfs2 - skipping start."
- return $OCF_SUCCESS
+ if [ "$FSTYPE" = "ocfs2" ]; then
+ # "start" now has the notification data available; that
+ # we're being started means we didn't get the
+ # pre-notification, because we weren't running, so
+ # process the information now first.
+ OCF_RESKEY_CRM_meta_notify_type="pre"
+ OCF_RESKEY_CRM_meta_notify_operation="start"
+ Filesystem_notify
fi
# See if the device is already mounted.
- Filesystem_status >/dev/null 2>&1
- if [ $? -eq $OCF_SUCCESS ] ; then
+ if Filesystem_status >/dev/null 2>&1 ; then
ocf_log info "Filesystem $MOUNTPOINT is already mounted."
return $OCF_SUCCESS
fi
@@ -293,32 +295,20 @@
# giving user-space membership events to a cluster-aware
# filesystem. Right now, only OCFS2 is supported.
#
- # We get notifications from hb2 that some operation (start or
- # stop) has completed; we then (1) compare the list of nodes
- # which are active in the fs membership with the list of nodes
- # which hb2 wants to be participating and remove those which
- # aren't supposed to be around. And vice-versa, (2) we add nodes
- # which aren't yet members, but which hb2 _does_ want to be
- # active.
+ # When we get a pre-start notification, we set up all the nodes
+ # which will be active in our membership for the filesystem.
+ # (For the resource to be started, this happens at the time of
+ # the actual 'start' operation.)
#
- # Eventually, if (3) we figure that we ourselves are on the list
- # of nodes which weren't active yet, we initiate a mount
- # operation.
+ # At a post-start, actually there's nothing to do for us really,
+ # but no harm done in re-syncing either.
#
- # That's it.
+ # pre-stop is meaningless; we can't remove any node yet, it
+ # first needs to unmount.
#
- # If you wonder why we don't process pre-notifications, or don't
- # do anything in "start": pre-start doesn't help us, because we
- # don't get it on the node just starting. pre-stop doesn't help
- # us either, because we can't remove any nodes while still
- # having the fs mounted. And because we can't mount w/o the
- # membership populated, we have to wait for the post-start
- # event.
+ # post-stop: the node is removed from the membership of the
+ # other nodes.
#
- # This approach _does_ have the advantage of being rather
- # robust, I hope. We always re-sync the current membership with
- # the expected membership.
- #
# Note that this expects that the base cluster is already
# active; ie o2cb has been started and populated
# $OCFS2_CLUSTER_ROOT/node/ already. This can be achieved by
@@ -337,51 +327,33 @@
local n_type="$OCF_RESKEY_CRM_meta_notify_type"
local n_op="$OCF_RESKEY_CRM_meta_notify_operation"
local n_active="$OCF_RESKEY_CRM_meta_notify_active_uname"
+ local n_stop="$OCF_RESKEY_CRM_meta_notify_stop_uname"
+ local n_start="$OCF_RESKEY_CRM_meta_notify_start_uname"
- case "$OCF_RESKEY_CRM_meta_notify_type" in
- *pre*)
- case "$OCF_RESKEY_CRM_meta_notify_operation" in
- *start*)
- # remove unames from notify_stop_uname
- for UNAME in "$OCF_RESKEY_CRM_meta_notify_stop_uname"
- do
- n_active=${n_active//$UNAME}
- done
- ;;
- *stop*)
- # leave notify_active_uname as is
- ;;
- esac
- ;;
- *post*)
- case "$OCF_RESKEY_CRM_meta_notify_operation" in
- *start*)
- # remove unames from notify_stop_uname
- for UNAME in "$OCF_RESKEY_CRM_meta_notify_stop_uname"
- do
- n_active=${n_active//$UNAME}
- done
- # add unames from notify_start_uname
- n_active="${n_active}
$OCF_RESKEY_CRM_meta_notify_start_uname"
- ;;
- *stop*)
- # remove unames from notify_stop_uname
- for UNAME in "$OCF_RESKEY_CRM_meta_notify_stop_uname"
- do
- n_active=${n_active//$UNAME}
- done
- ;;
- esac
- ;;
- esac
+ if [ "$n_type" = "pre" ]; then
+ if [ "$n_op" = "stop" ]; then
+ ocf_log debug "$OCFS2_UUID: ignoring pre-notify for
stop."
+ return $OCF_SUCCESS
+ fi
+ else
+ # TODO: This should be safe, but I don't dare yet.
+ # if [ "$n_op" = "start" ]; then
+ # ocf_log debug "$OCFS2_UUID: ignoring post-notify for
start."
+ # return $OCF_SUCCESS
+ # fi
+
+ # remove unames from notify_stop_uname; these have been
+ # stopped and can no longer be considered active.
+ for UNAME in "$n_stop"; do
+ n_active="${n_active//$UNAME/}"
+ done
+ fi
+ # These will be started/have been started. Prepare to
+ # communicate with them.
+ n_active="$n_active $n_start"
ocf_log debug "$OCFS2_UUID - notify: $n_type for $n_op - active on
$n_active"
- if [ "$n_type" != "post" ]; then
- ocf_log debug "$OCFS2_UUID: ignoring pre-notify."
- return $OCF_SUCCESS
- fi
-
local n_myself=${HA_CURHOST:-$(uname -n | tr A-Z a-z)}
ocf_log debug "$OCFS2_UUID: I am node $n_myself."
@@ -443,25 +415,9 @@
ocf_log err "$OCFS2_CLUSTER_ROOT/node/$entry:
failed to link"
# exit $OCF_ERR_GENERIC
fi
-
- if [ "$entry" = "$n_myself" ]; then
- OCFS2_DO_MOUNT=yes
- ocf_log debug "$OCFS2_UUID: To be mounted."
- fi
;;
esac
done
-
- # (3)
- # For now, always unconditionally go ahead; we're here, so we
- # should have the fs mounted. In theory, it should be fine to
- # only do this when we're activating ourselves, but what if
- # something went wrong, and we're in the membership but don't
- # have the fs mounted? Can this happen? TODO
- OCFS2_DO_MOUNT="yes"
- if [ -n "$OCFS2_DO_MOUNT" ]; then
- Filesystem_start
- fi
}
#
------------------------------
_______________________________________________
Linux-ha-cvs mailing list
[email protected]
http://lists.community.tummy.com/mailman/listinfo/linux-ha-cvs
End of Linux-ha-cvs Digest, Vol 30, Issue 72
********************************************
