Ethernet down ?

Running Ultramonkey 3 on RHEL AS 3.0 Update 4 ( 2.4.21-27.EL)
It has run fine for months. But it failed over a few days ago.
Still try to find it out why.

There are info in ha-log and messages

[root@pdmail1man log]# tail ha-log
heartbeat: 2005/09/29_23:24:25 info: Current arena value: 135168
heartbeat: 2005/09/29_23:24:25 info: MSG stats: 0/8646457 ms age 770
[pid1561/HBREAD]
heartbeat: 2005/09/29_23:24:25 info: ha_malloc stats: 0/172929140  252/0
[pid1561/HBREAD]
heartbeat: 2005/09/29_23:24:25 info: RealMalloc stats: 1468 total malloc
bytes. pid [1561/HBREAD]
heartbeat: 2005/09/29_23:24:25 info: Current arena value: 135168
heartbeat: 2005/09/29_23:24:25 info: These are nothing to worry about.
heartbeat: 2005/09/30_01:26:45 info: Link pdmail2man:eth0 up.
heartbeat: 2005/09/30_01:26:55 info: Link pdmail2man:eth0 dead.
heartbeat: 2005/09/30_02:44:09 info: Link pdmail2man:eth0 up.
heartbeat: 2005/09/30_02:44:19 info: Link pdmail2man:eth0 dead.

but ping is fine
[root@pdmail1man log]# ping pdmail2man
PING pdmail2man (172.16.20.190) 56(84) bytes of data.
64 bytes from pdmail2man (172.16.20.190): icmp_seq=0 ttl=64 time=0.082 ms
64 bytes from pdmail2man (172.16.20.190): icmp_seq=1 ttl=64 time=0.080 ms

on pdmail1man, /var/log/messages
Sep 29 23:24:25 pdmail1man heartbeat[1544]: info: Current arena value:
135168
Sep 29 23:24:25 pdmail1man heartbeat[1544]: info: These are nothing to worry
about.
Sep 30 01:26:45 pdmail1man heartbeat[1544]: info: Link pdmail2man:eth0 up.
Sep 30 01:26:45 pdmail1man ipfail[1573]: info: Link Status update: Link
pdmail2man/eth0 now has status up
Sep 30 01:26:55 pdmail1man heartbeat[1544]: info: Link pdmail2man:eth0 dead.
Sep 30 01:26:55 pdmail1man ipfail[1573]: info: Link Status update: Link
pdmail2man/eth0 now has status dead
Sep 30 01:26:55 pdmail1man ipfail[1573]: info: Asking other side for ping
node count.
Sep 30 01:26:55 pdmail1man ipfail[1573]: info: Checking remote count of ping
nodes.
Sep 30 01:26:55 pdmail1man ipfail[1573]: info: No giveup timer to abort.
Sep 30 02:44:09 pdmail1man heartbeat[1544]: info: Link pdmail2man:eth0 up.
Sep 30 02:44:09 pdmail1man ipfail[1573]: info: Link Status update: Link
pdmail2man/eth0 now has status up
Sep 30 02:44:19 pdmail1man heartbeat[1544]: info: Link pdmail2man:eth0 dead.
Sep 30 02:44:19 pdmail1man ipfail[1573]: info: Link Status update: Link
pdmail2man/eth0 now has status dead
Sep 30 02:44:19 pdmail1man ipfail[1573]: info: Asking other side for ping
node count.
Sep 30 02:44:19 pdmail1man ipfail[1573]: info: Checking remote count of ping
nodes.
Sep 30 02:44:19 pdmail1man ipfail[1573]: info: No giveup timer to abort.
Sep 30 04:20:05 pdmail1man ipfail[1573]: info: Ping node count is balanced.

on pdmail2man, /var/log/ha-log
heartbeat: 2005/09/29_12:25:02 info: Link 172.16.20.1:172.16.20.1 up.
heartbeat: 2005/09/29_12:25:02 info: Status update for node 172.16.20.1:
status ping
heartbeat: 2005/09/29_12:25:02 info: Local status now set to: 'active'
heartbeat: 2005/09/29_12:25:02 info: Starting child client
"/usr/lib/heartbeat/ipfail " (500,500)
heartbeat: 2005/09/29_12:25:02 info: Starting "/usr/lib/heartbeat/ipfail "
as uid 500  gid 500 (pid 1819)
heartbeat: 2005/09/29_12:25:02 info: remote resource transition completed.
heartbeat: 2005/09/29_12:25:02 info: remote resource transition completed.
heartbeat: 2005/09/29_12:25:02 info: Local Resource acquisition completed.
(none)
heartbeat: 2005/09/29_12:25:02 info: Initial resource acquisition complete
(T_RESOURCES(them))
heartbeat: 2005/09/30_04:20:06 info: Link pdmail1man:eth0 up.
heartbeat: 2005/09/30_04:20:16 info: Link pdmail1man:eth0 dead.
on pdmail2man, /var/log/messages
Sep 29 12:28:08 pdmail2man kernel: mtrr: type mismatch for fc000000,800000
old: uncachable new: write-combining
Sep 29 12:28:08 pdmail2man kernel: mtrr: type mismatch for fc000000,800000
old: uncachable new: write-combining
Sep 29 13:05:01 pdmail2man ipfail[1819]: info: Ping node count is balanced.
Sep 30 01:27:06 pdmail2man ipfail[1819]: info: Ping node count is balanced.
Sep 30 02:44:30 pdmail2man ipfail[1819]: info: Ping node count is balanced.
Sep 30 04:20:06 pdmail2man heartbeat[1790]: info: Link pdmail1man:eth0 up.
Sep 30 04:20:06 pdmail2man ipfail[1819]: info: Link Status update: Link
pdmail1man/eth0 now has status up
Sep 30 04:20:16 pdmail2man heartbeat[1790]: info: Link pdmail1man:eth0 dead.
Sep 30 04:20:16 pdmail2man ipfail[1819]: info: Link Status update: Link
pdmail1man/eth0 now has status dead
Sep 30 04:20:16 pdmail2man ipfail[1819]: info: Asking other side for ping
node count.
Sep 30 04:20:16 pdmail2man ipfail[1819]: info: Checking remote count of ping
nodes.
Sep 30 04:20:16 pdmail2man ipfail[1819]: info: No giveup timer to abort.


Here is /etc/ha.d/ha.cf

debugfile /var/log/ha-debug
logfile /var/log/ha-log
logfacility     local0
keepalive 1
deadtime 10
warntime 10
initdead 30
auto_failback off
udpport 694
bcast   eth0 eth2               # Linux
mcast eth2 225.0.0.1 694 1 0
node pdmail1man
node pdmail2man
ping 172.16.20.1
respawn hacluster /usr/lib/heartbeat/ipfail


-- 
Ultra Monkey - http://www.ultramonkey.org/
To UNSUBSCRIBE, email to [email protected], with a body:
unsubscribe ultramonkey-users [email protected]
where "[email protected]" is YOUR email address.