[Linux-cluster] qdisk votes not in cman
Hell, Robert
Robert.Hell at fabasoft.com
Fri Aug 31 06:52:12 UTC 2007
Hi,
for me it seems that qdiskd cannot successfully register in cman.
I think I found the reason for your problem in the source (libcman.c):
#define MAX_CLUSTER_NAME_LEN 16
...
int cman_register_quorum_device(cman_handle_t handle, char *name, int
votes)
{
struct cman_handle *h = (struct cman_handle *)handle;
char buf[strlen(name)+1 + sizeof(int)];
VALIDATE_HANDLE(h);
if (strlen(name) > MAX_CLUSTER_NAME_LEN)
{
errno = EINVAL;
return -1;
}
memcpy(buf, &votes, sizeof(int));
strcpy(buf+sizeof(int), name);
return info_call(h, CMAN_CMD_REG_QUORUMDEV, buf,
strlen(name)+sizeof(int), NULL, 0);
}
==> try using a shorter qdiskd device name than
"/dev/mpath/1HITACHI_750403220023p1" (< 16 chars).
Regards,
Robert
________________________________
From: linux-cluster-bounces at redhat.com
[mailto:linux-cluster-bounces at redhat.com] On Behalf Of Dave Costakos
Sent: Freitag, 31. August 2007 01:45
To: linux-cluster at redhat.com
Subject: [Linux-cluster] qdisk votes not in cman
I have a cluster running RHEL5 U1 beta with two nodes and a quorum disk
(at least it's in the cluster.conf file). However, when I run cman_tool
nodes and cman_tool status, I don't see the qdisk votes represented.
I've confirmed that qdiskd is indeed running. I'd appreciate it if
someone could help point me to my mistake here.
Thanks in advance.
Relevant info (too much??)
uname -rm: 2.6.18-36.el5xen x86_64
/etc/redhat-release: Red Hat Enterprise Linux Server release 5.1 Beta
(Tikanga)
rpm -q cman: cman-2.0.70-1.el5
CMAN/Qdisk Info:
[root at test-ibm-3650 ~]# cman_tool nodes
Node Sts Inc Joined Name
1 M 2256 2007-08-30 16:36:38 test-ibm-3650.qualcomm.com
2 M 2264 2007-08-30 16:36:38 test-ibm-3550.qualcomm.com
[root at test-ibm-3650 ~]# cman_tool status
Version: 6.0.1
Config Version: 18
Cluster Name: xen_cluster
Cluster Id: 31668
Cluster Member: Yes
Cluster Generation: 2264
Membership state: Cluster-Member
Nodes: 2
Expected votes: 2
Total votes: 2
Quorum: 2
Active subsystems: 9
Flags:
Ports Bound: 0 11 177
Node name: test-ibm-3650.qualcomm.com
Node ID: 1
Multicast addresses: 239.192.123.48
Node addresses: 10.45.5.7
[root at test-ibm-3650 ~]# service qdiskd status
qdiskd (pid 8994) is running...
syslog qdisk messages:
Aug 30 16:37:06 test-ibm-3650 qdiskd[8852]: <info> Quorum Daemon
Initializing
Aug 30 16:37:07 test-ibm-3650 qdiskd[8852]: <info> Heuristic: '/bin/ping
test-ibm-3550.qualcomm.com -c1 -t1' UP
Aug 30 16:37:07 test-ibm-3650 qdiskd[8852]: <info> Heuristic: '/bin/ping
test-ibm-3650.qualcomm.com -c1 -t1' UP
Aug 30 16:37:26 test-ibm-3650 qdiskd[8852]: <info> Initial score 2/2
Aug 30 16:37:26 test-ibm-3650 qdiskd[8852]: <info> Initialization
complete
Aug 30 16:37:26 test-ibm-3650 qdiskd[8852]: <notice> Score sufficient
for master operation (2/2; required=1); upgrading
Aug 30 16:37:38 test-ibm-3650 qdiskd[8852]: <info> Assuming master role
Aug 30 16:38:11 test-ibm-3650 qdiskd[8994]: <info> Quorum Daemon
Initializing
Aug 30 16:38:12 test-ibm-3650 qdiskd[8994]: <info> Heuristic: '/bin/ping
test-ibm-3550.qualcomm.com -c1 -t1' UP
Aug 30 16:38:12 test-ibm-3650 qdiskd[8994]: <info> Heuristic: '/bin/ping
test-ibm-3650.qualcomm.com -c1 -t1' UP
Aug 30 16:38:32 test-ibm-3650 qdiskd[8994]: <info> Initial score 2/2
Aug 30 16:38:32 test-ibm-3650 qdiskd[8994]: <info> Initialization
complete
Aug 30 16:38:32 test-ibm-3650 qdiskd[8994]: <notice> Score sufficient
for master operation (2/2; required=1); upgrading
Aug 30 16:38:44 test-ibm-3650 qdiskd[8994]: <info> Assuming master role
cluster.conf
<?xml version="1.0" ?>
<cluster alias="xen_cluster" config_version="20" name="xen_cluster">
<fence_daemon clean_start="0" post_fail_delay="0"
post_join_delay="20"/>
<clusternodes>
<clusternode name=" test-ibm-3650.qualcomm.com
<http://test-ibm-3650.qualcomm.com> " nodeid="1" votes="1">
<fence>
<method name="1">
<device
name="rsa-test-ibm-3650"/>
</method>
</fence>
</clusternode>
<clusternode name=" test-ibm-3550.qualcomm.com
<http://test-ibm-3550.qualcomm.com> " nodeid="2" votes="1">
<fence>
<method name="1">
<device
name="rsa-test-ibm-3550"/>
</method>
</fence>
</clusternode>
</clusternodes>
<cman/>
<fencedevices>
<fencedevice agent="fence_rsa"
hostname="rsa-test-ibm-3650" ipaddr="rsa-test-ibm-3650"
login="Administrator" name="rsa-test-ibm-3650" passwd="R0cknR011"/>
<fencedevice agent="fence_rsa"
hostname="rsa-test-ibm-3550" ipaddr="rsa-test-ibm-3550"
login="Administrator" name="rsa-test-ibm-3550" passwd="R0cknR011"/>
</fencedevices>
<rm>
<failoverdomains>
<failoverdomain name="vm_failover" ordered="0"
restricted="0">
<failoverdomainnode
name="test-ibm-3650.qualcomm.com" priority="1"/>
<failoverdomainnode name="
test-ibm-3550.qualcomm.com" priority="1"/>
</failoverdomain>
</failoverdomains>
<resources>
<clusterfs device="/dev/cluster_vg/xen_volume"
force_unmount="0" fsid="13127" fstype="gfs" mountpoint="/xen"
name="xen"/>
</resources>
<vm autostart="1" domain="vm_failover" exclusive="0"
name="dave-xen1" path="/xen/vm-config" recovery="restart"/>
<vm autostart="1" domain="vm_failover" exclusive="0"
name="dave-xen2" path="/xen/vm-config:/etc/xen" recovery="relocate"/>
</rm>
<quorumd device="/dev/mpath/1HITACHI_750403220023p1"
interval="2" log_level="7" min_score="1" tko="10" votes="2">
<heuristic interval="2" program="/bin/ping
test-ibm-3550.qualcomm.com -c1 -t1" score="1"/>
<heuristic interval="2" program="/bin/ping
test-ibm-3650.qualcomm.com -c1 -t1" score="1"/>
</quorumd>
</cluster>
--
Dave Costakos
mailto: david.costakos at gmail.com <mailto:david.costakos at gmail.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/linux-cluster/attachments/20070831/5db098c3/attachment.htm>
More information about the Linux-cluster
mailing list