Technical Blog Post
Abstract
nco_p_generic_tmf814_2_0 probe unable to subscribe to Ericsson SOEM notification
Body
Previously, it was reported that users of probe-nco-p-generic-tmf814-2_0 usually encountered a connection problem when subscribing to notification of the Ericsson SOEM EMS. This was observed in the probe debug log shown below
...
2015-06-30T15:24:51: Debug: D-JPR-000-000: Subscribing for notifications
2015-06-30T15:24:51: Debug: D-JPR-000-000: Subscribe notification consumer...
2015-06-30T15:24:51: Debug: D-JPR-000-000: Populating event channel...
2015-06-30T15:24:51: Debug: D-JPR-000-000: Set consumer admin...
2015-06-30T15:24:51: Debug: D-JPR-000-000: Set proxy push supplier...
2015-06-30T15:24:51: Error: E-JPR-000-000: Exception when subscribing to notification : org.omg.CORBA.UNKNOWN: vmcid: 0x0 minor code: 0 completed: Maybe
2015-06-30T15:24:51: Information: I-JPR-000-000: Endpoint subscribed
2015-06-30T15:24:51: Debug: D-JPR-000-000: com.ibm.tivoli.netcool.omnibus.probe.framework.ProbeRunner.connect EXITING
2015-06-30T15:24:51: Debug: D-JPR-000-000: com.ibm.tivoli.netcool.omnibus.probe.framework.ProbeRunner.resetRetry ENTERING
2015-06-30T15:24:51: Debug: D-JPR-000-000: com.ibm.tivoli.netcool.omnibus.probe.framework.ProbeRunner.resetRetry EXITING
2015-06-30T15:24:51: Information: I-JPR-000-000: Probe connected
2015-06-30T15:25:51: Debug: D-JPR-000-000: com.ibm.tivoli.netcool.omnibus.probe.framework.HeartbeatTask.run ENTERING
2015-06-30T15:25:51: Information: I-JPR-000-000: Checking Probe Heartbeat
2015-06-30T15:25:51: Debug: D-JPR-000-000: com.ibm.tivoli.netcool.omnibus.probe.framework.HeartbeatTask.run EXITING
2015-06-30T15:25:51: Information: I-JPR-000-000: DISCONNECT 'Lost subscription'
2015-06-30T15:25:51: Debug: D-JPR-000-000: com.ibm.tivoli.netcool.omnibus.probe.framework.ProbeRunner.resetForRetry ENTERING
2015-06-30T15:25:51: Debug: D-JPR-000-000: com.ibm.tivoli.netcool.omnibus.probe.framework.ProbeRunner.haltScheduledTasks ENTERING
2015-06-30T15:25:51: Debug: D-JPR-000-000: com.ibm.tivoli.netcool.omnibus.probe.framework.ProbeRunner.haltScheduledTasks EXITING
2015-06-30T15:25:51: Debug: D-JPR-000-000: Scheduled resync time = 0 : Time now = 1435670751436
2015-06-30T15:25:51: Debug: D-JPR-000-000: com.ibm.tivoli.netcool.omnibus.probe.framework.ProbeRunner.disconnectProbe ENTERING
2015-06-30T15:25:51: Information: I-JPR-000-000: Probe disconnecting
2015-06-30T15:25:51: Information: I-JPR-000-000: Disconnecting TMF814...
2015-06-30T15:25:51: Information: I-JPR-000-000: Probe disconnected
...
From IBM ORB debug log, the above error message (highlighted in red) indicated that when probe was requesting TMF814 API obtain_notification_push_supplier, the EMS replied with SYSTEM_EXCEPTION
OUT GOING:
Request Message
Date: June 30, 2015 3:24:51 PM CEST
Thread Info: ProbeRunner
Local Port: 41218 (0xA102)
Local IP: 10.127.44.88
Remote Port: 27889 (0x6CF1)
Remote IP: 10.127.44.88
GIOP Version: 1.2
Byte order: big endian
Fragment to follow: No
Message size: 160 (0xA0)
--
Request ID: 20
Response Flag: WITH_TARGET
Target Address: 0
Object Key: length = 64 (0x40)
004F4F00 FE504F41 FE457665 6E744368
616E6E65 6C466163 746F7279 504F41FE
00000000 00000000 493F35DF 134EC7F4
00000000 00000000 00000000 00000003
Operation: obtain_notification_push_supplier
Service Context: length = 2 (0x2)
Context ID: 1229081874 (0x49424D12)
Context data: length = 8 (0x8)
00BD0001 16000001
Context ID: 17 (0x11)
Context data: length = 2 (0x2)
0002
Data Offset: a2
0000: 47494F50 01020000 000000A0 00000014 GIOP............
0010: 03000000 0000BDBD 00000040 004F4F00 ...........@.OO.
0020: FE504F41 FE457665 6E744368 616E6E65 .POA.EventChanne
0030: 6C466163 746F7279 504F41FE 00000000 lFactoryPOA.....
0040: 00000000 493F35DF 134EC7F4 00000000 ....I?5..N......
0050: 00000000 00000000 00000003 00000022 ..............."
0060: 6F627461 696E5F6E 6F746966 69636174 obtain_notificat
0070: 696F6E5F 70757368 5F737570 706C6965 ion_push_supplie
0080: 7200BDBD 00000002 49424D12 00000008 r.......IBM.....
0090: 00BD0001 16000001 00000011 00000002 ................
00A0: 0002BDBD BDBDBDBD 00000002 ............
IN COMING:
Reply Message
Date: June 30, 2015 3:24:51 PM CEST
Thread Info: RT=3:P=690676:O=0:TCPTransportConnection[addr=10.127.44.88,port=27889,local=41218]
Local Port: 41218 (0xA102)
Local IP: 10.127.44.88
Remote Port: 27889 (0x6CF1)
Remote IP: 10.127.44.88
GIOP Version: 1.2
Byte order: big endian
Fragment to follow: No
Message size: 56 (0x38)
--
Request ID: 20
Service Context: length = 0 (0x0)
Reply Status: SYSTEM_EXCEPTION
Exception ID: IDL:omg.org/CORBA/UNKNOWN:1.0
Minor code: 0(0x0)
Completion status:MAYBE
0000: 47494F50 01020001 00000038 00000014 GIOP.......8....
0010: 00000002 00000000 0000001E 49444C3A ............IDL:
0020: 6F6D672E 6F72672F 434F5242 412F554E omg.org/CORBA/UN
0030: 4B4E4F57 4E3A312E 30000000 00000000 KNOWN:1.0.......
0040: 00000002
This is because it was found that (and later confirmed by Ericsson) SOEM EMS TMF814 implementation only supported STRUCTURED_EVENT. Different from the GA-ed version of the probe; probe-nco-p-generic-tmf814-2_0 which only supported SEQUENCE_EVENT. Therefore, in probe-nco-p-generic-tmf814-2_1 (testfix), a new probe property; NotificationClientType has been introduced to support both SEQUENCE_EVENT and STRUCTURED_EVENT. By default, this property has a value of 'SEQUENCE_EVENT'. Since this EMS only support Structure event, user need to enable and change the property value to 'STRUCTURED_EVENT' after apply the testfix.
UID
ibm11082271