similar to: o2net patch that avoids socket disconnect/reconnect

Currently, o2net connects to a node on hb_up and disconnects on hb_down and net timeout. It disconnects on net timeout is ok, but it should attempt to reconnect back. This is because sometimes nodes get overloaded enough that the network connection breaks but the disk hb does not. And if we get into that situation, we either fence (unnecessarily) or wait for its disk hb to die (and sometimes hang

Modification from V1 to V2: 1. Use atomic ops instead of spin_lock in timer. 2. Add some comments when querying connect_expired work. These comments are copied form Zach's mail.;) Currently, o2net connects to a node on hb_up and disconnects on hb_down and net timeout. It disconnects on net timeout is ok, but it should attempt to reconnect back. This is because sometimes nodes get

This patch separates o2net and o2quo from knowing about one another as much as possible. This is the first in a series of patches that will allow userspace cluster interaction. Quorum is separated out first, and will ultimately only be associated with the disk heartbeat as a separate module. To do so, this patch performs the following changes: * o2hb_notify() is added to handle injection of

The following series of patches comprises the bulk of our outstanding changes for Ocfs2. Aside from the usual set of cleanups and fixes that were inappropriate for 2.6.25, there are a few highlights: The '/sys/o2cb' directory has been moved to '/sys/fs/o2cb'. The new location meshes better with modern sysfs layout. A symbolic link has been placed in the old location so as to

Currently, o2net_accept_one() is allowed to accept a connection from listening node itself, such a fake connection will not be successfully established due to no handshake detected afterwards, and later end up with triggering connecting worker in a loop. We're going to fix this by treating such connection request as 'invalid', since we've got no chance of requesting connection

The only non-fix here is Joel's new configfs convenience macros, but nobody is using them yet, so I think the patch is safe. By the way, these patches (as usual) are all rebased on top of your latest tree. I think that since the vast majority of ocfs2 and configfs patches are self-contained and within a small area of the kernel that this should probably be fine. If you feel otherwise however,

This is a forward port for net_proc.c from 1.2.

Fix ocfs2 to use the page fragment allocator rather than kzalloc in order to allocate the buffers for the handshake message and keepalive request and reply messages. Switch from using sendpage() to using sendmsg() + MSG_SPLICE_PAGES so that sendpage can be phased out. Signed-off-by: David Howells <dhowells at redhat.com> cc: Mark Fasheh <mark at fasheh.com> cc: Joel Becker <jlbec

Fix ocfs2 to use the page fragment allocator rather than kzalloc in order to allocate the buffers for the handshake message and keepalive request and reply messages. Slab pages should not be given to sendpage, but fragments can be. Switch from using sendpage() to using sendmsg() + MSG_SPLICE_PAGES so that sendpage can be phased out. Signed-off-by: David Howells <dhowells at redhat.com>

Fix ocfs2 to use the page fragment allocator rather than kzalloc in order to allocate the buffers for the handshake message and keepalive request and reply messages. Switch from using sendpage() to using sendmsg() + MSG_SPLICE_PAGES so that sendpage can be phased out. Signed-off-by: David Howells <dhowells at redhat.com> cc: Mark Fasheh <mark at fasheh.com> cc: Joel Becker <jlbec

Fix ocfs2 to use the page fragment allocator rather than kzalloc in order to allocate the buffers for the handshake message and keepalive request and reply messages. Slab pages should not be given to sendpage, but fragments can be. Switch from using sendpage() to using sendmsg() + MSG_SPLICE_PAGES so that sendpage can be phased out. Signed-off-by: David Howells <dhowells at redhat.com>

Signed-off-by: Christoph Hellwig <hch@lst.de> Index: linux-2.6/fs/ocfs2/cluster/tcp.c =================================================================== --- linux-2.6.orig/fs/ocfs2/cluster/tcp.c 2007-05-06 13:51:17.000000000 +0200 +++ linux-2.6/fs/ocfs2/cluster/tcp.c 2007-05-17 15:00:14.000000000 +0200 @@ -261,14 +261,12 @@ out: static void o2net_complete_nodes_nsw(struct o2net_node

All, Please review the patches backported to 1.4 from mainline. Sunil

Replace wait_event_interruptible() with wait_event() in o2net_send_message_vec(). This is because this function is called by the dlm that expects signals to be blocked. Fixes oss bugzilla#1126 http://oss.oracle.com/bugzilla/show_bug.cgi?id=1126 Signed-off-by: Sunil Mushran <sunil.mushran at oracle.com> --- fs/ocfs2/cluster/tcp.c | 7 ++----- 1 files changed, 2 insertions(+), 5

Hi all, I have a problem with cpu affinity in my R-2.11.1 installation compiled against ATLAS running on a Linux (Ubuntu 10.04) cluster under GridEngine. I wish to use Grid Engine's core binding feature to bind user processes into the number of cores they request on the cluster, thus preventing badly behaved multi-threaded libraries from consuming more cores than requested. An example of

From: Tom Rix <trix at redhat.com> This is a upcoming change to clean up a new warning treewide. I am wondering if the change could be one mega patch (see below) or normal patch per file about 100 patches or somewhere half way by collecting early acks. clang has a number of useful, new warnings see

From: Tom Rix <trix at redhat.com> This is a upcoming change to clean up a new warning treewide. I am wondering if the change could be one mega patch (see below) or normal patch per file about 100 patches or somewhere half way by collecting early acks. clang has a number of useful, new warnings see