Hi,
I have a 2 node HB 2.1.3 cluster running on CentOS 5. I just upgraded
the passive node to CentOS 5.4, but the heartbeat packages did not
change:
heartbeat-stonith-2.1.3-3.el5.centos
heartbeat-2.1.3-3.el5.centos
heartbeat-pils-2.1.3-3.el5.centos
Now, when I try to start HB on the node, it reports that it is starting,
but the daemons never actually start:
r...@fc-fmcln02:~$ service heartbeat start
logd is already running
Starting High-Availability services:
2010/05/04_18:02:53 INFO: Resource is stopped
2010/05/04_18:02:53 INFO: Resource is stopped
[ OK ]
r...@fc-fmcln02:~$ ps aux | grep heartbeat
root 6117 0.0 0.0 3916 696 pts/0 S+ 18:02 0:00 grep
heartbeat
Logs say:
May 4 18:02:53 fc-fmcln02 heartbeat: [6112]: info: Version 2 support:
false
May 4 18:02:53 fc-fmcln02 heartbeat: [6112]: WARN: logd is enabled but
logfile/debugfile is still configured in ha.cf
May 4 18:02:53 fc-fmcln02 heartbeat: [6112]: info:
**************************
May 4 18:02:53 fc-fmcln02 heartbeat: [6112]: info: Configuration
validated. Starting heartbeat 2.1.3
May 4 18:02:53 fc-fmcln02 heartbeat: [6113]: info: heartbeat: version
2.1.3
May 4 18:02:53 fc-fmcln02 heartbeat: [6113]: info: Heartbeat
generation: 1208455483
Running /usr/lib/heartbeat/heartbeat -d 1000 shows:
heartbeat[6122]: 2010/05/04_18:04:00 ERROR: Cannot shmget for process
status: Invalid argument
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(keepalive,1)
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(deadtime,10)
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(warntime,5)
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(initdead,120)
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(udpport,694)
heartbeat: udpport setting must precede media statementsheartbeat[6122]:
2010/05/04_18:04:00 debug: add_option(bcast,eth1)
heartbeat[6122]: 2010/05/04_18:04:00 debug:
add_option(auto_failback,off)
heartbeat[6122]: 2010/05/04_18:04:00 debug:
add_option(node,fc-fmcln01.corp.follett.com)
heartbeat[6122]: 2010/05/04_18:04:00 debug:
add_option(node,fc-fmcln02.corp.follett.com)
heartbeat[6122]: 2010/05/04_18:04:00 info: respawn directive: hacluster
/usr/lib/heartbeat/ipfail
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(use_logd,yes)
heartbeat[6122]: 2010/05/04_18:04:00 info: Enabling logging daemon
heartbeat[6122]: 2010/05/04_18:04:00 info: logfile and debug file are
those specified in logd config file (default /etc/logd.cf)
heartbeat[6122]: 2010/05/04_18:04:00 debug:
add_option(logfile,/var/log/hb.log)
heartbeat[6122]: 2010/05/04_18:04:00 debug:
add_option(debugfile,/var/log/heartbeat-debug.log)
heartbeat[6122]: 2010/05/04_18:04:00 debug: uid=hacluster, gid=<null>
heartbeat[6122]: 2010/05/04_18:04:00 debug: uid=hacluster, gid=<null>
heartbeat[6122]: 2010/05/04_18:04:00 debug: uid=<null>, gid=haclient
heartbeat[6122]: 2010/05/04_18:04:00 debug: uid=root, gid=<null>
heartbeat[6122]: 2010/05/04_18:04:00 debug: uid=<null>, gid=haclient
heartbeat[6122]: 2010/05/04_18:04:00 debug: Beginning authentication
parsing
heartbeat[6122]: 2010/05/04_18:04:00 debug: 16 max authentication
methods
heartbeat[6122]: 2010/05/04_18:04:00 debug: Keyfile opened
heartbeat[6122]: 2010/05/04_18:04:00 debug: Keyfile perms OK
heartbeat[6122]: 2010/05/04_18:04:00 debug: 16 max authentication
methods
heartbeat[6122]: 2010/05/04_18:04:00 debug: Found authentication method
[sha1]
heartbeat[6122]: 2010/05/04_18:04:00 info: AUTH: i=1: key = 0x8c52d78,
auth=0x5c6228, authname=sha1
heartbeat[6122]: 2010/05/04_18:04:00 debug: Outbound signing method is 1
heartbeat[6122]: 2010/05/04_18:04:00 debug: Authentication parsing
complete [1]
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(cluster,linux-ha)
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(hopfudge,1)
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(baud,19200)
heartbeat: baudrate setting must precede media
statementsheartbeat[6122]: 2010/05/04_18:04:00 debug:
add_option(hbgenmethod,file)
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(realtime,true)
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(msgfmt,classic)
heartbeat[6122]: 2010/05/04_18:04:00 debug:
add_option(conn_logd_time,60)
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(log_badpack,true)
heartbeat[6122]: 2010/05/04_18:04:00 debug:
add_option(syslogmsgfmt,false)
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(coredumps,true)
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(crm,false)
heartbeat[6122]: 2010/05/04_18:04:00 info: Version 2 support: false
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(autojoin,none)
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(uuidfrom,file)
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(compression,zlib)
heartbeat[6122]: 2010/05/04_18:04:00 debug:
add_option(compression_threshold,2)
heartbeat[6122]: 2010/05/04_18:04:00 debug:
add_option(traditional_compression,no)
heartbeat[6122]: 2010/05/04_18:04:00 debug:
add_option(max_rexmit_delay,250)
heartbeat[6122]: 2010/05/04_18:04:00 debug: Setting max_rexmit_delay to
250 ms
heartbeat[6122]: 2010/05/04_18:04:00 debug:
add_option(record_config_changes,on)
heartbeat[6122]: 2010/05/04_18:04:00 debug:
add_option(record_pengine_inputs,on)
heartbeat[6122]: 2010/05/04_18:04:00 debug:
add_option(enable_config_writes,on)
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(memreserve,6500)
heartbeat[6122]: 2010/05/04_18:04:00 debug: add_option(debug,1)
heartbeat[6122]: 2010/05/04_18:04:00 WARN: logd is enabled but
logfile/debugfile is still configured in ha.cf
heartbeat[6122]: 2010/05/04_18:04:00 info: **************************
heartbeat[6122]: 2010/05/04_18:04:00 info: Configuration validated.
Starting heartbeat 2.1.3
heartbeat[6122]: 2010/05/04_18:04:00 debug: HA configuration OK.
Heartbeat starting.
My /etc/ha.d/ha.cf has not changed either. It is the same on node1:
keepalive 1 # Interval between heartbeat (HB)
packets.
deadtime 10 # How quickly HB determines a dead node.
warntime 5 # Time HB will issue a late HB.
initdead 120 # Time delay needed by HB to report a dead
node.
udpport 694 # UDP port HB uses to communicate between
nodes.
bcast eth1 # Which interface to use for HB packets.
auto_failback off # Auto promotion of primary node upon return
to cluster.
node fc-fmcln01.corp.blah.com # Node name must be same as uname -r.
node fc-fmcln02.corp.blah.com # Node name must be same as uname -r.
respawn hacluster /usr/lib/heartbeat/ipfail
# Specifies which programs to run at startup
use_logd yes # Use system logging.
logfile /var/log/hb.log # Heartbeat logfile.
debugfile /var/log/heartbeat-debug.log # Debugging logfile.
Can someone help me determine why HB dies immediately after I start it?
How can I further troubleshoot this issue?
Thanks!
Josh
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
