[drbd-mc] heartbeat status failed

[Ralf Prengel]

Rasto Levrinc schrieb:

> 
> Which Heartbeat version do you have? Version =< 2.1.3 is supported at the
> moment, 2.1.4 and partly 2.99.3 with pacemaker is supported in the current
> git.

lxclus01:~ # rpm -qa|grep heart
heartbeat-resources-2.99.3-1.6
libheartbeat2-2.99.3-1.6
heartbeat-2.99.3-1.6
heartbeat-ldirectord-2.99.3-1.6
heartbeat-common-2.99.3-1.6
lxclus01:~ #



> 
> Try crm_mon, cibadmin -Q on all server nodes, if it does not work, you would
> have to debug that first.

============
Last updated: Thu Feb 26 13:24:34 2009
Current DC: lxclus02 (1af59558-d077-4574-8944-68517f83840b)
Version: 1.0.2-c02b459053bfa44d509a2a0e0247b291d93662b7
2 Nodes configured.
0 Resources configured.
============

Node: lxclus01 (403cba20-bb7f-4dd2-a1c8-686e532367a8): online
Node: lxclus02 (1af59558-d077-4574-8944-68517f83840b): online

lxclus01:~ # cibadmin -Q
<cib validate-with="pacemaker-1.0" crm_feature_set="3.0.1"
have-quorum="1" admin_epoch="0" epoch="16" num_updates="3"
cib-last-written="Thu Feb 26 10:37:15 2009"
dc-uuid="1af59558-d077-4574-8944-68517f83840b">
  <configuration>
    <crm_config>
      <cluster_property_set id="cib-bootstrap-options">
        <nvpair id="cib-bootstrap-options-dc-version" name="dc-version"
value="1.0.2-c02b459053bfa44d509a2a0e0247b291d93662b7"/>
      </cluster_property_set>
    </crm_config>
    <nodes>
      <node id="403cba20-bb7f-4dd2-a1c8-686e532367a8" uname="lxclus01"
type="normal"/>
      <node id="1af59558-d077-4574-8944-68517f83840b" uname="lxclus02"
type="normal"/>
    </nodes>
    <resources/>
    <constraints/>
  </configuration>
  <status>
    <node_state id="403cba20-bb7f-4dd2-a1c8-686e532367a8"
uname="lxclus01" crmd="online" join="member"
crm-debug-origin="do_state_transition" shutdown="0" ha="active"
in_ccm="true" expected="member">
      <lrm id="403cba20-bb7f-4dd2-a1c8-686e532367a8">
        <lrm_resources/>
      </lrm>
      <transient_attributes id="403cba20-bb7f-4dd2-a1c8-686e532367a8">
        <instance_attributes
id="status-403cba20-bb7f-4dd2-a1c8-686e532367a8">
          <nvpair
id="status-403cba20-bb7f-4dd2-a1c8-686e532367a8-probe_complete"
name="probe_complete" value="true"/>
        </instance_attributes>
      </transient_attributes>
    </node_state>
    <node_state id="1af59558-d077-4574-8944-68517f83840b"
uname="lxclus02" ha="active" in_ccm="true" crmd="online"
crm-debug-origin="do_state_transition" join="member" expected="member"
shutdown="0">
      <lrm id="1af59558-d077-4574-8944-68517f83840b">
        <lrm_resources/>
      </lrm>
      <transient_attributes id="1af59558-d077-4574-8944-68517f83840b">
        <instance_attributes
id="status-1af59558-d077-4574-8944-68517f83840b">
          <nvpair
id="status-1af59558-d077-4574-8944-68517f83840b-probe_complete"
name="probe_complete" value="true"/>
        </instance_attributes>
      </transient_attributes>
    </node_state>
  </status>
</cib>
lxclus01:~ #


> 
> Sometimes there is problem with mgmtd daemon. When you stop the heartbeat
> and mgmtd is still running, kill it and then start the heartbeat again.

ok
this seems to be the problem.
mgmtd is not running

Feb 26 13:29:02 lxclus01 mgmtd: [32237]: ERROR: Cannot sign on with
heartbeat
Feb 26 13:29:02 lxclus01 mgmtd: [32237]: ERROR: REASON:
Feb 26 13:29:02 lxclus01 mgmtd: [32237]: ERROR: Can't initialize
management library.Shutting down.(-1)
Feb 26 13:29:19 lxclus01 mgmtd: [32243]: info: G_main_add_SignalHandler:
Added signal handler for signal 15
Feb 26 13:29:19 lxclus01 mgmtd: [32243]: debug: Enabling coredumps
Feb 26 13:29:19 lxclus01 mgmtd: [32243]: WARN: Core dumps could be lost
if multiple dumps occur.
Feb 26 13:29:19 lxclus01 mgmtd: [32243]: WARN: Consider setting
non-default value in /proc/sys/kernel/core_pattern (or equivalent) for
maximum supportability
Feb 26 13:29:19 lxclus01 mgmtd: [32243]: WARN: Consider setting
/proc/sys/kernel/core_uses_pid (or equivalent) to 1 for maximum
supportability
Feb 26 13:29:19 lxclus01 mgmtd: [32243]: info: G_main_add_SignalHandler:
Added signal handler for signal 10
Feb 26 13:29:19 lxclus01 mgmtd: [32243]: info: G_main_add_SignalHandler:
Added signal handler for signal 12
Feb 26 13:29:19 lxclus01 heartbeat: [32195]: WARN: Client [mgmtd] pid
32243 failed authorization [no default client auth]
Feb 26 13:29:19 lxclus01 heartbeat: [32195]: ERROR:
api_process_registration_msg: cannot add client(mgmtd)
Feb 26 13:29:19 lxclus01 mgmtd: [32243]: ERROR: Cannot sign on with
heartbeat
Feb 26 13:29:19 lxclus01 mgmtd: [32243]: ERROR: REASON:
Feb 26 13:29:19 lxclus01 mgmtd: [32243]: ERROR: Can't initialize
management library.Shutting down.(-1)

Any hints how to solve this on Suse 11.1?


Thanks for your help.

-- 
i.A. Ralf Prengel
Customer Care Manager
Comline AG
Hauert 8
D-44227 Dortmund/Germany

Fon +49231 97575- 904
Fax +49231 97575- 905
EMail ralf.prengel at comline.de

Die CeBIT steht dieses Jahr unter dem Motto " Wer die Welt bewegt, bewegt sich hier". Aus diesem Grund finden Sie auch Comline vom 03.03.2009 bis 08.03.2009 auf der CeBIT in Halle 9,  Stand C60. Nutzen Sie die Gelegenheit sich das Angebot von Comline auf der CeBIT anzusehen und/ oder vereinbaren Sie Termine für eine persönliche Beratung.

www.comline.de 
Vorstand Stephan Schilling, Erwin Leonhardi 
Aufsichtsrat Dr. Franz Schoser (Vorsitzender) HR Dortmund B 14570 USt.-ID-Nr. DE 124727422