Hi All,

We confirmed that SIGTERM was not handled in hbagent used in Pacemaker.

Environment is the next environment.

  * RHEL6.4(kernel-2.6.32-358.23.2.el6.x86_64)
  * net-snmp-libs-5.5-49.el6_5.1.x86_64
  * hp-snmp-agents-9.50-2564.40.rhel6.x86_64
  * net-snmp-utils-5.5-49.el6_5.1.x86_64
  * net-snmp-5.5-49.el6_5.1.x86_64
   * It is understood that net-snmp included in RHEL is slightly different from 
the original.

Even if the application of hbagent received SIGTERM, hbagent did not stop.
The problem seems to be taking place by connect processing in the 
netsnmp_unix_transport processing.

----------------------
(gdb) bt
#0  0x0000003c9b0e97a0 in ?? ()
#1  0x000000351ca725ec in netsnmp_unix_transport (addr=0x7fff138e8700, 
local=<value optimized out>) at snmpUnixDomain.c:398
#2  0x000000351ca72e60 in netsnmp_unix_create_tstring (string=<value optimized 
out>, local=0, default_target=<value optimized out>) at snmpUnixDomain.c:452
#3  0x000000351ca68923 in netsnmp_tdomain_transport_full 
(application=0x351e841a0d "agentx", str=0x0, local=0, default_domain=<value 
optimized out>, default_target=0x0) at snmp_transport.c:443
#4  0x000000351e8210d4 in subagent_open_master_session () at 
mibgroup/agentx/subagent.c:768
#5  0x000000351e821561 in agentx_reopen_session (clientreg=67, clientarg=<value 
optimized out>) at mibgroup/agentx/subagent.c:876
#6  0x000000351ca636f4 in run_alarms () at snmp_alarm.c:263
#7  0x0000000000406285 in main (argc=<value optimized out>, argv=<value 
optimized out>) at hbagent.c:1669
(gdb) f 1
#1  0x000000351ca725ec in netsnmp_unix_transport (addr=0x7fff138e8700, 
local=<value optimized out>) at snmpUnixDomain.c:398
398             rc = connect(t->sock, (struct sockaddr *) addr,
(gdb) l
393                 return NULL;
394             }
395             memcpy(t->remote, addr->sun_path, strlen(addr->sun_path));
396             t->remote_length = strlen(addr->sun_path);
397     
398             rc = connect(t->sock, (struct sockaddr *) addr,
----------------------

Someone know the outbreak of the similar problem?
And, someone know the workaround?

 * And we doubt the bug of Kernel and investigate it.

Best Regards,
Hideo Yamauchi.


------------------------------------------------------------------------------
_______________________________________________
Net-snmp-users mailing list
Net-snmp-users@lists.sourceforge.net
Please see the following page to unsubscribe or change other options:
https://lists.sourceforge.net/lists/listinfo/net-snmp-users

Reply via email to