ZXA10 C300 V1.2.5P3 Patch Release Notes - 20170320
ZXA10 C300 V1.2.5P3 Patch Release Notes - 20170320
5P3 Patch
Release Notes
ZXA10 C300 V1.2.5P3 Patch Release Notes
Gaominghu
V1.1 2015/12/25 Added new patch(No.27- No.30)
i
TABLE OF CONTENT
TABLE OF CONTENTS.......................................................................................................... 2
FIGURES 12
TABLES 12
1.1 Update Table.......................................................................................................13
1.2 Information Table................................................................................................23
2 Patch List........................................................................................................... 29
2.1 SCXLV125P3T13_r0.pat, SCXMV125P3T13_r0.pat, SCXNV125P3T13_r0.pat,
SMXAV125P3T13_r0.pat....................................................................................29
2.1.1 PON service configuration memory leaks............................................................29
2.1.2 Singapore telecom LACP interrupt problem........................................................30
2.1.3 Add/remove GTGHG line card memory leaks.....................................................31
2.1.4 Port-location configuration only first ONU normal problem on NMS....................31
2.1.5 Bandwidth utilization of uplink always 0...............................................................31
2.1.6 Line card HWONLINE using first external clock..................................................33
2.1.7 SMXA type 2&3 sub-card 10G uplink down problem...........................................34
ZXAN(config)#show card....................................................................................................34
ZXAN(config)#show updatecpld.........................................................................................34
2.1.8 SMXA power consumption Display abnormal......................................................34
ZXAN#show power..............................................................................................................34
ZXAN#show power..............................................................................................................35
2.2 SCXLV125P3T13_r1.pat, SCXMV125P3T13_r1.pat, SCXNV125P3T13_r1.pat,
SMXAV125P3T13_r1.pat....................................................................................35
2.2.1 ACL-based traffic mirror, remote traffic mirror failed............................................35
ZXAN(config)#interface gei_1/22/1.....................................................................................35
G3-Koszalin(control-panel)#...............................................................................................36
G3-Koszalin(control-panel)#...............................................................................................36
View under Shell that the task XponCtrl of the main control card was suspended.......37
SMXA>ti 161 38
Execute cmd line 370 code 91828 error:' port-location format airtel vport 1 '...............40
Execute cmd line 386 code 91828 error:' port-location format airtel vport 1 '...............40
Execute cmd line 398 code 91828 error:' port-location format airtel vport 1 '...............40
Execute cmd line 413 code 91828 error:' port-location format airtel vport 1 '...............40
Execute cmd line 431 code 91828 error:' port-location format airtel vport 1 '...............40
Interpt Error cmd line 228:' onu profile sip VOIP dial-plan VOIP660DialPlan onu.........40
2.2.8 It prompted error when creating SIP profile on NMS...........................................40
2.2.9 The main control card SMXA’s global igmp disable didn’t take effect..................40
2.2.10 Problem when NMS deleted xconnect entries.....................................................41
2.2.11 SCXN adopted extended VLAN, services were blocked after active / backup
changeover.........................................................................................................41
ZXAN(config-if)#show load-balance...................................................................................41
ZXAN(config)#interface gei_1/19/1.....................................................................................42
2.2.12 SCXN 1588 bind VLAN, configuration was lost after active / backup changeover
42
2.2.13 SCXN configures extended VLAN, yellow packets were dropped.......................42
Get info: 43
FileName:/display.c.............................................................................................................43
ProcessName:InterptProc1.................................................................................................43
Free info: 43
FileName:0x043
RegSets: 51
Layer 1 addr:[0x0]0x0..........................................................................................................52
SN Bind: disable......................................................................................................63
Device ID: 63
Description: ONU-7:15.................................................................................................63
Stack chain: 66
2.8.8 HW HG8120 ONU was off-line in Tie Ling, Liaoning province.............................66
2.9 GTGOAV125P3T13_r2.pat, GTGOCV125P3T13_r2.pat,
GTGODV125P3T13_r2.pat, GTGOEV125P3T13_r2.pat,
GTGOGV125P3T13_r2.pat, GTGHGV125P3T13_r2.pat....................................67
2.9.1 ONU failed to learn MAC addresses....................................................................67
2.9.2 Services of yellow packets of the third-party ONU were abnormal......................67
2.9.3 After Ip-source-guard was enabled, confused data caused ONU services
blocked................................................................................................................68
2.9.4 DGi alarm was frequently reported......................................................................68
2.9.5 LOID mistaken binding problem..........................................................................68
2.9.6 Optical power collection caused suspension of tXponCtrlSub0 task....................69
Basic-info: 71
Material-Number:.................................................................................................................71
Diagnostic-info:................................................................................................................... 72
Alarm-thresh:....................................................................................................................... 72
2.13 SMXAV125P3T13_r5.pat, HUTQV125P3T13_r0.pat,
HUTQBV125P3T13_r0.pat, GUCDV125P3T13_r0.pat........................................73
2.13.1 Uplink Optical Transceiver Temperature Alarm...................................................73
Basic-info: 73
Material-Number:.................................................................................................................74
Diagnostic-info:................................................................................................................... 74
Alarm-thresh:....................................................................................................................... 74
2.14 SCXLV125P3T13_r5.pat.....................................................................................75
2.14.1 SCXL+GTGOG Lose Packets in the Downstream Direction................................75
2.15 GTGOAV125P3T13_r3.pat GTGODV125P3T13_r3.pat......................................75
2.15.1 F601 Unable to Connect Online..........................................................................75
2.16 SCXLV125P3T13_r6.pat, SCXMV125P3T13_r6.pat, SCXNV125P3T13_r6.pat,
SMXAV125P3T13_r6.pat....................................................................................76
2.16.1 ARP Proxy Answer Error.....................................................................................76
2.16.2 802.1X Packets Extracted to CPU By Default, Causing SNMP Packet Loss.......76
2.16.3 SNMP Ping Failure..............................................................................................76
2.16.4 Snmp V3 Ping Failure.........................................................................................76
2.17 GTGOEV125P3T13_r4.pat, GTGOAV125P3T13_r4.pat,
GTGOCV125P3T13_r4.pat, GTGODV125P3T13_r4.pat,
GTGOGV125P3T13_r4.pat, GTGHGV125P3T13_r4.pat....................................77
2.17.1 New Service Cannot Learn MAC After the PON Service Profile is Replaced......77
2.17.2 GEMPORT Rate Limit cpb=pbs Causes Service Interruption..............................77
2.17.3 “Failure” Returned When GEMPORT ID is 129...................................................78
2.17.4 GTGOA Port With an Even Number Does Not Show MAC.................................78
2.17.5 9806mib Synchronization Error...........................................................................78
2.17.6 Powered-off F607 (HGU) Under GTGHG Cannot Re-Register............................78
2.17.7 ONU Cannot Register Under GTGOA.................................................................79
2.17.8 Huawei ONU Cannot Work..................................................................................79
2.17.9 OLT Fiber Break and Power-off Alarm Optimization............................................79
2.17.10 GPON Line Card tod Distribution Error................................................................79
2.17.11 EPONN Card Suspends and “txponnpproc suspend” Prompted Under Shell......80
2.18 SCXLV125P3T13_r7.pat, SCXMV125P3T13_r7.pat, SCXNV125P3T13_r7.pat,
SMXAV125P3T13_r7.pat....................................................................................80
2.18.1 ARP Proxy Answer Error.....................................................................................80
2.18.2 Task Suspension Caused by Telnet Screen Width Too Large.............................81
2.18.3 Last Port Cannot be Obtained by Standard mib getnext......................................81
2.18.4 Multiple oid Variable Binding When Performing GetBulk Operations Causes
Dead Loop...........................................................................................................81
2.18.5 EMS Cannot Display “patch-saved” Information..................................................82
2.18.6 GDFO Port Rate is Displayed as 100M When It Is Configured Into 1000M.........82
2.18.7 SNMP Operation Causes Process Suspension...................................................82
2.18.8 “show gpon remote-onu mac” Causes Memory Out-of-Boundary........................82
2.18.9 “Failure” Is Returned When Downloading a Patch With the Same Name as the
Patch That Is Illegally Deleted.............................................................................83
2.18.10 Patch Activation Causes SCXL Card Cannot Enable MFF Due to Memory Out-
of-boundary.........................................................................................................83
2.18.11 Loopback Address Can Be Pinged but Cannot Be Telnet...................................83
2.19 GTGOEV125P3T13_r5.pat, GTGOAV125P3T13_r5.pat,
GTGOCV125P3T13_r5.pat, GTGODV125P3T13_r5.pat
GTGOGV125P3T13_r5.pat, GTGHGV125P3T13_r5.pat....................................84
2.19.1 IGMP Traffic Slow and Overtime.........................................................................84
2.19.2 New F607 Registration Slow...............................................................................84
2.19.3 F822 Cannot Connect Online..............................................................................85
2.19.4 OMCI Alarm Memory Leakage............................................................................85
2.19.5 GTGHG 140900/140901 PCB Forbids CPLD Upgrade.......................................86
2.20 SCXLV125P3T13_r22.pat SCXNV125P3T13_r22.pat.......................................86
2.20.1 Administrator create ONU profile failure in NMS.................................................86
2.21 GTGOE.FW(V1.2.5P3T20)..................................................................................86
2.21.1 Gemport rate limit failture problem......................................................................86
2.22 GTGOGV125P3T13_r22.pat...............................................................................86
2.23 GTGOEV125P3T13_r22.pat GTGHGV125P3T13_r22.pat..................................87
Command Format:...............................................................................................................91
Command Function:............................................................................................................ 91
3.3 Rollback Steps....................................................................................................92
Command Format:...............................................................................................................92
Command Function:............................................................................................................ 92
Command Format:...............................................................................................................93
Command Function:............................................................................................................ 93
3.4 Upgrading Confirmation.......................................................................................94
Command Format:...............................................................................................................94
ZXAN#show patch-saved....................................................................................................94
Command Format:...............................................................................................................94
Command Function:............................................................................................................ 94
ZXAN#show patch-running.................................................................................................95
3.5 Update Precautions.............................................................................................95
FIGURES
TABLES
GTGODV125P3T13_
r26.pat
GTGOEV125P3T13_r
26.pat
GTGOGV125P3T13_
r26.pat
GTGHGV125P3T13_
r26.pat
2 Patch List
Symptom: When used onu profiles, remote management can cause problems of task
suspending. And the remote management of related configuration under repeated
configuration cause memory leaks. Because the code is a bug, if at the time of abnormal
branch will appear just allocated memory is not released, this will lead to memory leaks,
can use the memory will be less and less. Similar to the following code, and can lead to
memory allocated to avlEntry not released:
if(gponrmIsVirtualOnu(pOntPos))
{
return ERR_GPON_RECORD_NONEXT;
}
avlEntry->ontPos.slotId = pOntPos->slotId;
avlEntry->ontPos.oltId = pOntPos->oltId;
avlEntry->ontPos.ontId = pOntPos->ontId;
arg.p = &avlEntry->ontPos;
return avlInsert(&root->root, &avlEntry->avlNode, arg, root->compare);
}
Because this place multiple instances or multiple types only insert a AVL node, but
multiple instances, or more than one type of configuration every time apply
GponRm_AddOntNodeToAVL calls will be alloced a memory, and could eventually
because of repeated node index and no insert AVL tree, which leads to memory leaks;
Many command line or MIB can trigger operation, there are more similar functions.
Note: If those memory had leaked before the patch has been actived, there is no way to
recovery, only by restart main contrl board or by swap to restore.
Symptom: When Singapore telecom tested the V1.2.5P3 version, it found that LACP will
appear constantly aggregation and deaggregation, led to service interruption, finally
found root cause: the LACP protocol packets were discarded. since the LACP protocol
module didn't receive the LACP protocol packets, so it deaggregated, after a while,
receive the packets, it was aggregated. It was a period of time again.
Note: none.
Symptom: if add/remove GTGHG line card once, it was leak 16K memory, again and
again to add delete line card can use less memory.
Note: If those memories had leaked before the patch has been activated, there is no
way to recovery, only by restart main control board or by swap to restore.
Symptom: Thailand project, configure the number of ONU, it can only choose the first
related configuration ONU, when configurate ONU port-location function on NMS, other
ONUs are not appeared in the list. for example,in the 1/2/1 interface, the first ONU port-
location can be select, but the second ONU port-location can not be select ,it was
empty. if you remove the ONU 1/2/1:1 and 1/2/1:2 port-location function, the behind
ONU 1/2/1:3 will be normal.
Note: none.
Symptom: the bandwidth utilization of uplink always was 0, even in the case of traffic.the
comman as follow:
Description is none
Duplex full
scramble payload-enable
Input:
Dropped :0 Fragments :0
Jabber :0 MacRxErr :0
IncorrectVlanDrop: 0
Action: Due to the output format was error, fix the bug.
Note: Those patches which date was 2015-03-13 were some problem, they were
deprecated.
Symptom: When CICK or CICG used first external clock as the clock source, under
some combinations, some line cards may be appear HWONLINE state, as follow:
1. SCXL/SCXM/SCXMC+CICK, some PCB version and CPLD version, the CICK will be
HWONLINE.
1. SCXL+CICK, the SCXL PCB is Neither 080600 nor 080601,the CICK will be
HWONLINE.
2. SCXL+CICK, the SCXL PCB is Neither 080600 nor 080601,if the line cards were
sensitivity of the 2M PPS, they will be HWONLINE, such as FTGKB and ETGOD.
3. SCXM+CICK, the SCXM CPLD version is less than V1.3(not incude V1.3), the CICK
will be HWONLINE.
4. SCXM+CICG, if the line cards were sensitivity of the 2M PPS, they will be
HWONLINE, such as FTGKB and ETGOD.
6. SCXMC+CICG, if the line cards were sensitivity of the 2M PPS, they will be
HWONLINE, such as FTGKB and ETGOD.
Root cause: the cleaning function was called the initial function
Note: Only SCXM and SCXL possible problems, SCXN, SMXA there is no problem.
Symptom: The SMAX main control board, which PCB version was 131200 and CPLD
version was V1.4, was reboot, the 10G uplink port of types 2&3 sub-card has been down
state.the service was interrupt.Replace the new version of the CPLD can solve, or use
this patch. The PCB and CPLD versioin as follow:
ZXAN(config)#show card
Rack Shelf Slot CfgType RealType Port HardVer SoftVer Status
-------------------------------------------------------------------------------
1 1 3 SMXA SMXA 3 131200 V1.2.5P3 INSERVICE
1 1 4 SMXA 3 OFFLINE
ZXAN(config)#show updatecpld
Rack Shelf Slot RealType NextResetUpdate LastUpdateResult Version
-----------------------------------------------------------------------------
1 1 3 SMXA NO SUCCESS V1.4
Symptom: when show power,some time was normal, some time was abnormal, as
follows:
ZXAN#show power
------------------------------------------------------------
Rack Shelf Slot Voltage(V) Current(A) Power(W)
------------------------------------------------------------
1 1 3 N/A. N/A. 55
1 1 4 N/A. N/A. 1
ZXAN#show power
------------------------------------------------------------
Rack Shelf Slot Voltage(V) Current(A) Power(W)
------------------------------------------------------------
1 1 3 N/A. N/A. 55
1 1 4 N/A. N/A. 6694
Symptom: Adopt the ACL mode to configure traffic mirror, but it didn’t take effect, and
unable to get mirroring effect. Because ports of the drive and service mux interface are
inconsistent, ACL-based traffic mirror configuration failed.
Configuration:
ZXAN(config)#acl hybrid number 300
ZXAN(config-hybd-acl)#rule 1 permit any any any any vlan-id eq 100
ZXAN(config-hybd-acl)#rule 2 permit any any any any vlan-id eq 200
ZXAN(config)#interface gei_1/22/1
Note: If the problem has appeared before patched, the configuration should be deleted
first before patched, and then load the patch once again, after activated, re-configure it.
Symptom: CPU queue rate-limit configuration is inconsistent with the value saved,
shown as below:
G3-Koszalin(control-panel)#cpu queue 5 512
G3-Koszalin(control-panel)#
G3-Koszalin(control-panel)#show cpu queue
------------------------------------------------------------------------
QueueId Ratelimit
0 256
1 256
2 256
3 256
4 256
5 512
6 256
7 256
G3-Koszalin(control-panel)#
G3-Koszalin(control-panel)#show run | begin control-
control-panel
packet-limit all 1000
cpu queue 6 512
!
Note: NA.
Symptom: When Winter-Time / Summer Time were converted on March 10, C300 had a
problem, without switching, we detected the configuration and we found the Summer
Time setting had a problem. It started from the second Sunday in March, ended the first
Sunday in November, but they were wrong shown on the OLT and EMS, some sites
became Saturday, some became Friday. When the fault happened, the command
changed, the correct setting should be Sunday, but some became Friday and some
became Saturday somehow.
clock summer-time recurring DST 2 friday mar 2:0:0 1 friday nov 2:0:0 60. Finally
confirm: when setting via mib, NE processing was normal, but when getting via mib, NE
returned value isn’t converted according to mib interface’s definition, causing actual
returned value to be one day in advance than mib’s defined value. When configured as
Sunday on NMS, it will be one day in advance and becomes Saturday when querying, if
NMS configures once again, when querying via NMS, the value got will be Friday again.
Above fault will appear.
Note: NA
Symptom: In Vietnam, on other versions, XponCtrl task was suspended caused by Mib
access conflict, and V1.2.5P3 version also has the hidden risk. Patches were needed to
solve it. Exception on other versions: NE XponCtrl task was suspended, show gpon onu
state command failed, that is:
View under Shell that the task XponCtrl of the main control card was suspended.
SMXA>task XponCtrl(0x56357c38) suspend. (Error=0x2)
tt 161
task: pid:161
task call func info:
only back traced one func addr: 0x10f683a8 name: getLastOfflineRecord
SMXA>task XponCtrl(0x56357c38) suspend. (Error=0x2)
SMXA>ti 161
Note: If the problem has appeared, you need to restart to be able to recover by adopting
active / backup changeover mode. After patched, first perform active / backup
changeover, and then restart the original main control card.
Symptom: When it was SCXL as the main control, if the uplink port adopted trunk, and
at the same time, enable the remote mirror, when the trunk port was destination port, it
would lead to endless loop for NE Protocol task, causing NE detached.
Note: Just SCXN had the problem, if the problem has appeared, you need to restart to
be able to recover.
Symptom: Perform self-test in the Institute that other versions appeared once, after the
active card was switched, the backup card failed to be standby. Through the code
walkthrough, there were loopholes.
Note: NA
2.2.7 Failed to load after the configuration in sip profile was saved and
restarted
Symptom: In GPON, if dial-plan was configured in sip profile, it would cause problems
for dial-plan and app-srv/accesscode parameters to generate scripts, after saved and
restarted, the dial-plan/app-srv/accesscode configuration in sip profile would fail to load.
Error information listed as blow may appear:
Execute cmd line 119 code 65733 error:'task add'
Interpt Error cmd line 132:' '
Interpt Error cmd line 165:'port-location rackno 1 frameno 1'
Interpt Error cmd line 228:' onu profile sip VOIP dial-plan VOIP660DialPlan
onuInterpt Error cmd line 307:'p2p-onu'
Execute cmd line 370 code 91828 error:' port-location format airtel vport 1 '
Execute cmd line 386 code 91828 error:' port-location format airtel vport 1 '
Execute cmd line 398 code 91828 error:' port-location format airtel vport 1 '
Execute cmd line 413 code 91828 error:' port-location format airtel vport 1 '
Execute cmd line 431 code 91828 error:' port-location format airtel vport 1 '
Interpt Error cmd line 228:' onu profile sip VOIP dial-plan VOIP660DialPlan onu
Symptom: When creating SIP profile on NMS, it always prompted error, because the NE
wasn’t compatible with original practices, it needed to deliver only 1 of 4 index table
entries in the MIB table originally to be able to succeed, but the NE modified it as strict
inspection to need 4 index, otherwise, it would return error; because the NE wasn’t
compatible with original practices, causing the profile to fail to be created.
Note: NA.
2.2.9 The main control card SMXA’s global igmp disable didn’t take effect
Symptom: When the user configured C320 global igmp disable, the multicast protocol
packets from the user port were still extracted to the CPU, that is, when global igmp
disable was configured, the OLT multicast service was transmitted transparently, in this
case, multicast service was blocked. Because when igmp disable was configured,
ETGOB adopted aggregated inline port PON line card, the drive of the SMXA main
control card configured multicast protocol packet extracting rule for the line card’s
aggregation port, causing upstream multicast protocol packets of the line card to be
extracted to the CPU, but not to be forwarded to the uplink port, multicast service was
abnormal. Condition for the fault: It needs to configure ETGOB or other PON line cards,
and at the same time, the inline port of the line card is aggregated.
Action: When global disable was configured, IGMP protocol packets wouldn’t be
extracted.
Note: Just SMXA had the problem, while other main control cards are normal. After the
patch is activated and effective, deactivate it, after in global igmp enable and igmp
disable, the patch is still effective; to make the patch not effective, you need to configure
igmp enable first, and then activate the patch, and igmp disable again.
Symptom: xconnect from the user port to uplink port and customer xconnect could be
created successfully, but failed to be deleted. xconnect entries couldn’t be deleted; user
xconnect was normal, and can be added and deleted.
Note: NA.
2.2.11 SCXN adopted extended VLAN, services were blocked after active /
backup changeover
Symptom: After extended VLAN was configured on the SCXN main control card,
execute the command: ZXAN(config)#ex-switch vlan 200 to configure VLAN service,
after active / backup changeover, services would be blocked. In active / backup mode,
first configure extended VLAN, and configure same VLAN on the uplink port, etc., and
then perform active / backup changeover, services will be blocked.
ZXAN(config-if)#show load-balance
Current load_balance setting: Disable --- Active / backup mode, not load sharing
mode
ZXAN(config)#ex-switch vlan 200 --- Configure extended VLAN
ZXAN(config)#interface gei_1/19/1
ZXAN(config-if)#sw vlan 1000 tag --- Configure same VLAN on the uplink port
If the uplink port VLAN is configured first, and then configure extended VLAN, services
won’t be affected after active / backup changeover.
Note: Just SCXN main control card had the problem, if the problem has appeared
before, you need to delete the extended VLAN manually, and then re-configure it.
2.2.12 SCXN 1588 bind VLAN, configuration was lost after active / backup
changeover
Symptom: After active / backup changeover, the clock FPGA interface’s VLAN lost,
causing L2 messages unable to be switched to FPGA, and 1588 messages unable to be
forwarded normally.
EC No.: 613003118989.
Note: Just SCXN main control card had the problem, if the problem has appeared
before, you need to delete the extended VLAN manually, and then re-configure it. To
delete the patch, strictly execute: First delete the configuration, and then deactivate the
patch, re-configure it after the patch is activated again.
Symptom: If the command ex-switch was used to configure extended vlan and enable
color mapping, the yellow messages forwarded to FPGA would be directly dropped, that
is, CFI=1 messages would be directly dropped:
ZXAN(config)#ex-switch vlan 3000
ZXAN(config)#qos cfi-to-drop cfi1 medium
Note: Just SCXN main control card had the problem, if the problem has appeared
before, you need to delete the extended VLAN manually, and then re-configure it.
Symptom: In Lithuania, there was one OLT on which only one user could log
simultaneously via telnet; later, log in the NE’s shell, it was found that the maximum
memory block of the NE had only 192 bytes, but memory fragments reached 560, 000.
After the fault was recovered, observe the NE memory changes, it was confirmed that
10001 bytes memory was leaked, and the leaked call chain:
Leaked service trigger causes, executing the following command will lead to:
Note: If the problem has appeared, you need to restart to be able to recover by adopting
active / backup changeover mode. After patched, first perform active / backup
changeover, and then restart the original main control card. Abnormal TACACS+
messages caused memory out of boundary
Symptom: When the Tacacs server sent specific abnormal messages to the NE, or the
NE received some abnormal tacacs messages, it would cause memory out of boundary,
when it did happen, it might cause other tasks suspended, and if a certain task’s
memory was out of boundary by tacacs, the task might be suspended; if the NMS task
was out of boundary, it might cause NMS detached.
Note: If suspension problem has appeared, that is, out of boundary might have
happened, you need to restart to be able to recover, or perform active / backup
changeover to recover. It is better execute active / backup changeover twice after
patched, to completely eliminate hidden risks.
Fault Description: After the version was updated to V1.2.5P3 in Indonesia, the main
control task MsanCfg was suspended, its task stacking information:
Ma0x8f0aab8 (SysCtrlProcess): Check task id=0x8f043f8, name=MsanCfg
suspend!
0x8f0aab8 (SysCtrlProcess): MP check task id=0x8f043f8, name=MsanCfg
suspend!
sanCfg
1df4760 vxTaskEntry +68 : ScheEntry ()
1572a8c ScheEntry +114: 15702f0 ()
15705a0 rosPShowAll +670: 1570034 (8f043f8)
1570210 rosPShowAll +2e0: MsCtrl_CfgTaskEntry ()
43097c MsCtrl_CfgTaskEntry+288: MsCtrl_ShowCmdProcess ()
430320 MsCtrl_ShowCmdProcess+64 : MsCtrl_CmdProcess ()
430198 MsCtrl_CmdProcess+a8 : GponSrv_CliCmdHandler (81010906, 172e9c08,
172e9370)
91b17c GponSrv_CliCmdHandler+a4 : CfgGponOnuRegParaParseFunction
(172e9c08, 172e9370)
c402d0 CfgGponOnuRegParaParseFunction+104: GponsrvCli_Cfg (172e9c08,
172e9370, 8b44e2c)
c52128 GponsrvCli_Cfg +90 : PonsrvCli_Cfg (172e9c08, 172e9370, 8b44e2c)
c12540 PonsrvCli_Cfg +29c: GponCliCfg_ExecAccess (8b44e2c, 8b44a50,
8b44a58)
c1228c GponCliCfg_ExecAccess+d8 : XponCtrl_ExecAccess (924, 8b44adc, 34c,
8b44a50, 8, 8b44a58)
8cf06c XponCtrl_ExecAccess+1b8: GponSrv_ExecHandler (924, 8b44adc,
8b44a50, 8b449c4)
7fa08c GponSrv_ExecHandler+188: SetGponOnuBaseInfoFunction (8b44adc,
8b44a50, 8b449c4)
After above are configured, and then delete ONU, or re-add vlan-filter-mode iphost and
onu-vlan iphost, maybe there will appear tasks suspended.
Note: The patch must cooperate to use with r0; if the suspension problem appears, you
need to perform active / standby change-over twice to completely restore or reboot the
NE. When patching, if there is no problem temporarily, we suggest performing active /
standby change-over twice after patched if there is host or other configuration, making
the two main control cards run once again, to completely eliminate hidden dangers.
2.5 SCXLV125P3T13_r3_indonesia.pat
Fault Description: After the version was updated to V1.2.5P3 in Indonesia, the main
control task MsanCfg was suspended, its task stacking information:
Ma0x8f0aab8 (SysCtrlProcess): Check task id=0x8f043f8, name=MsanCfg
suspend!
0x8f0aab8 (SysCtrlProcess): MP check task id=0x8f043f8, name=MsanCfg
suspend!
sanCfg
1df4760 vxTaskEntry +68 : ScheEntry ()
1572a8c ScheEntry +114: 15702f0 ()
15705a0 rosPShowAll +670: 1570034 (8f043f8)
1570210 rosPShowAll +2e0: MsCtrl_CfgTaskEntry ()
43097c MsCtrl_CfgTaskEntry+288: MsCtrl_ShowCmdProcess ()
430320 MsCtrl_ShowCmdProcess+64 : MsCtrl_CmdProcess ()
430198 MsCtrl_CmdProcess+a8 : GponSrv_CliCmdHandler (81010906, 172e9c08,
172e9370)
94e158 gponDeleteOmciHostBridgePortData+50 :
gponFreeOmciHostBridgePortCfgBuf (8b43e54)
95b750 gponFreeOmciHostBridgePortCfgBuf+184:
gponFreeOmciIpHostVlanFilterModeCfgBuf (8b43e54)
95c948 gponFreeOmciIpHostVlanFilterModeCfgBuf+160: avlDelete ()
1e877ac avlDelete +130: avlRebalance ()
After above are configured, and then delete ONU, or re-add vlan-filter-mode iphost and
onu-vlan iphost, maybe there will appear tasks suspended.
Action: Modify AVL tree and Special modifications for Indonesia scenario.
Note: The patch must cooperate to use with r0 and Only for Indonesia scenario; if the
suspension problem appears, you need to perform active / standby change-over twice to
completely restore or reboot the NE. When patching, if there is no problem temporarily,
we suggest performing active / standby change-over twice after patched if there is host
or other configuration, making the two main control cards run once again, to completely
eliminate hidden dangers.
2.6.1 The card status was in configuring, and the occupancy of the
backup main control CPU was 99% all the time
1. Failed to add ONU Failure reason: device operation failed (the card is in
Configuring status);
2. Logged in the OLT and found that the card status was abnormal: the status of
GTGOG cards in Slot 8 and 15 was in configing.
3. Use show processor to view that the occupancy of the backup main control card’s
CPU was 99%, and the memory utilization rate of the active main control card and
backup main control card was about 75%.
4. All the users under Slot 8 and 15 were offline, and it was invalid to deliver card
restarting command via CLI.
Active main control card’s PPMgt task was blocked all the time, stacking information:
[ROS10]:tt 0x8f0b348
1df56b0 vxTaskEntry +68 : 1574018 ()
1574018 ScheEntry +6b4: 1571020 (8f0b348)
15711fc rosPShowAll +2e0: 11ae54 ()
11ae54 PPMgt +208c: Ppmgt_unitstate_proc ()
118c88 Ppmgt_unitstate_proc+188: Ppmgt_pre_proc ()
111d38 Ppmgt_pre_proc +304: send_noconfig_alarm_or_restore ()
12aa84 send_noconfig_alarm_or_restore+4c : PRWGCircleSync ()
15419c PRWGCircleSync +2a8: copy ()
1e2b998 copy +80 : 1e2b688 ()
1e2b6b0 cd +cc : stat ()
1e0f8cc stat +28 : open ()
1e14b58 open +14 : 1e14b68 ()
Action: Because when synchronizing active / backup files, if the backup main control
card’s nfs file system wasn’t finished, it would be abnormal when synchronizing. Use the
patch to modify current problem.
Note: After the problem has appeared, maybe restarting won’t be effective, you need to
execute reset_card command under the main control’s serial port to recover.
Symptom: Disabling the summer-time function via NMS could just disable the summer-
time function on the active main control card, the summer-time configuration on the
backup main control card wouldn’t be deleted, and it was still valid, after active / backup
changeover, the summer-time function could be enabled once again.
Note: NA.
Symptom: Protocol task of the main control card was suspended, suspension file’s
contents:
RegSets:
r0 = f7 sp = 8bd4180 r2 = 0 r3 = 6e278f0
r4 = 8bd4188 r5 = 0 r6 = 0 r7 = 0
r8 = 0 r9 = 400000 r10 = 2cdd848 r11 = 6e278f0
r12 = 0 r13 = 0 r14 = 0 r15 = 0
r16 = 0 r17 = 0 r18 = 0 r19 = 1a43a1e8
r20 = 8bd41dc r21 = 8bd41d0 r22 = 8bd41c4 r23 = 1a43a030
r24 = 8bd49e8 r25 = 0 r26 = 6e278f0 r27 = 8bd41f4
r28 = 6e278f0 r29 = 20189c31 r30 = 1 r31 = 8bd424c
msr = b032 lr = 42cd80 ctr = 2b8f4c pc = 42cdc8
cr = 20000044 xer = 0
Task stacking:
Stack chain:
Layer 0 addr:[0x42cdc8]MsPortGetIfEntry_CommonPort + 0x90
Layer 1 addr:[0x0]0x0
Layer 2 addr:[0x24b97f8]if_cmd_show_interface_undirect_eth_ifId+ 0x164
Layer 3 addr:[0x24ba008]if_cmd_show_interface_undirect_eth+ 0x2e8
Layer 4 addr:[0x1cf9e28]if_cmd_show_interface + 0x3e0
Layer 5 addr:[0x1cdc458]ip_oam_handler + 0x3d4
Layer 6 addr:[0x13b0fa8]DistribInProtocol + 0x4e8
Layer 7 addr:[0xc1564]ProtocolEntry + 0x1a30
According to analysis, the ifid transmitted was abnormal, and would be out of boundary,
then add protection in the patch.
Note: NA.
2.6.4 After version upgrade, the acl extend configuration would lose
Symptom: For the project in Czech Republic, after the version was upgraded from
V1.2.3P3 to V1.2.5P3, extend acl configuration would lose. The format of 1.2.3P3’s
extend acl saved to startrun.dat is:
Because in 1.2.5P3 version, the command’s format was modified as acl extended, at the
same time, 1.2.5P3 version didn’t support to match acl extend command ambiguously,
that is, acl extend failed to be identified as acl extended, causing the command loading
to fail, and the command to lose.
Action: Fixed bug, acl extend could be matched as acl extended, and could load
successfully.
Note: After the version was upgraded V1.2.5P3, and the patch was activated, you
needed to restart to make it effective, that is, if there were NEs configuring acl extend
command, after the patch was activated, you needed to restart to make it effective, or
re-create acl extend manually.
2.6.5 VLAN rate-limit of the main control caused suspension of the PPMgt
task
Symptom: When the version was upgraded to C300V1.2.5P3 version, the PPMgt task
was suspended, through our analysis, we confirmed that is was caused by VLAN rate-
limit of the main control, and its stacking:
[ROS10]:tt 0x8f0b348
1df56b0 vxTaskEntry +68 : ScheEntry ()
1573a78 ScheEntry +114: 15712dc ()
157158c rosPShowAll +670: 1571020 (8f0b348)
15711fc rosPShowAll +2e0: 11a16c ()
11a16c PPMgt +13a4: 118d00 ()
118da8 Ppmgt_unitstate_proc+2a8: PPMgt_Add_Board ()
Note: After the problem appeared, after patched, you need to perform active / backup
changeover to recover.
2.6.6 The dhcp ip hash conflict caused users unable to get IP addresses
Symptom: For the project in Latvia, one ONU failed to be called, and there was the
following print under the serial port:
On the existing network, after related IP data was cleared manually, the ONU dialed
successfully. Further analyze: dhcp relay adopted standard mode in Latvia, the ip hash
table shouldn’t save data, and at the same time, there weren’t failures of deletion; Under
the offer response of the existing network’s server, the code processing of present NEs
had loopholes: among the 5 IPs responded by the server, only one IP could apply
normally, while other four might be left in the table entries, causing later failure of
allocating these IPs to PCs by the server once again.
Note: If the problem has appeared, after the patch is activated, you need to perform
active / backup changeover once to recover.
Symptom: Configure hostname via CLI, the maximum length of the characters was 32
bytes, but at most 64 bytes could be configured via mib interface. If MIB interface is
adopted to configure hostname and its length is more than 32 bytes, after the
configuration is saved, after rebooted, the hostname will be lost. For example:
hostname SQ-SHUANGFENGSI-ZTE-FTTH-C300-1
Note: NA.
Symptom: when changing NMS ONU ID to driver ONU ID, the 127 will be as a invalid
ID, that could lead to accidentally delete, mishandle the ONU data, which the driver onu
id was 127.
Note: none.
Symptom: When the line card was inservice, CLI command show process, or to use line
card shell command spy, found out some task total percent is high, because for the
processing of the PON MAC driver alarm task have long scanning. For example, after
executing spy command as follow:
Symptom: when test one multicast service, run the script, after repeated many times ,not
delete the multicast configuration(flow 255), the multicast service was not working. clear
ONU all configuration and restart, it was still not working,only deleted onu and added
onu again, or delete the multicast flow 255 and reconfigured the multicast flow 255,the
multicast was work. Due to the repeated the clear or remove mvlan, the line card did
NOT check the number of vlans, lead to no protection.
Action: fixed the bug, added the flow of 255 operation of protection.
Note: none.
Symptom: Singapore telecom test GTGOE line card, when enable IP SOURCE GUARD,
in the process of protocol interaction, upward all GUA IPv6 address is not allowed to
pass through, the realization of the current version is only discarded GUA IPv6 address
for 2001:: /16, need to modify for addresses 2001::/3.
Symptom: Shenyang mobile V1.2.5 P3 version upgrade, the line card multiple tasks
were suspending, its tasks and stacks information were as follows:
-> i
NAME ENTRY TID PRI STATUS PC SP ERRNO DELAY
---------- ------------ -------- --- ---------- -------- -------- ------- -----
tExcTask excTask 6ffe7e8 0 PEND 857218 6ffe6c8 0 0
……
mac_config Configuratio 6415a00 102 SUSPEND 77afe90 6a02270 3d0004 0
……
tDrvIntHnd DRV_IntHandl 5e23d80 106 SUSPEND 77afe90 6a02270 1c0001 0
……
STATE_MACHIState_Machin 610f800 120 SUSPEND 77afe90 6a02270 0 0
KEY_EXCHANGKey_Exchange 6059168 120 SUSPEND 77afe90 6a02270 0
0
Ploam_DS Ploam_Ds_Tas 5fd0dc8 120 SUSPEND 77afe90 6a02270 1c0001
0
……
mac_performp_perf_task 5ea87b0 120 SUSPEND 77afe90 6a02270 3d0004
0
… ….
olsTask 6f91a4 65ca708 150 SUSPEND 77afe90 6a02270 0 0
……
almMaskTaskalmInfoRepor 65c64f8 235 SUSPEND 77afe90 6a02270 0 0
……
muxTask macMuxTimerT 65dab28 240 SUSPEND 7d7e04 65da618 0 0
……
tRosIdle rosIdleTask 69a93a0 255 READY 68a4c 69a9300 0 0
value = 0 = 0x0
tt 0x65dab28
7da8ec vxTaskEntry +68 : macMuxTimerTask ()
61ba80 macMuxTimerTask+f0 : checkOnuBipCounterHandleAlm ()
641f44 checkOnuBipCounterHandleAlm+284: checkSfiSdiAlmStatus ()
6d4b34 checkSfiSdiAlmStatus+270: 6d47a8 ()
6d4858 setOnuSfiSdiAlmThreshold+418: recvAlmInfoFromOtherModule (65da9e0)
6ca868 recvAlmInfoFromOtherModule+28 : almInfoHandle (65da9e0)
6cdb48 almInfoHandle +6c : reportAlmToOtherModule (65da9e0)
6cd2a4 reportAlmToOtherModule+54 : handleOnuAlarmReport (65da9e0)
6d7384 handleOnuAlarmReport+34 : handleAndReportOnuAlarmAssert (65da9e0)
6d7a80 handleAndReportOnuAlarmAssert+40 : handleInShieldRangeOnuAlm
(65da9e0)
6d7928 handleInShieldRangeOnuAlm+280: reportOnuAlarmAssert (65da9e0)
6d7690 reportOnuAlarmAssert+2c : reportAlarmToSerModule (65da9e0)
6cd118 reportAlarmToSerModule+2b0: MuxlayLogOnuMsg (5, d)
708330 MuxlayLogOnuMsg+c4 : MuxlayvLogMsg ()
Note: If there were already suspending, after the patch was active,it need to restart the
line card.
Symptom: when onu was registered by LOID, these onu will not be working. Because
the LOID and LOIDPW cannot report to auth module, onu’s state can't move syncMib,
and finally can’t move working state. Because it was found out that the task sent
message to itself, to the normal processing will not be able to perform, which will not be
able to get the LOID and LOIDPW from onu, therefore, authentication is not able to
complete, the ONU can't working.
Note: only for GTGOA and GTGOD line card, other line cards were normal.
Symptom: SCXM/SCXN + GTGOG, when the swap was from slot 10 to slot 11, the
service may interrup second level. Because of the GTGOG inner-port which was
corresponding slot 11 was disable, after swap completed, the port was enabled, it
affected the service recovery time.
Note: only for GTGOG line card, other line cards were normal..
Symptom: GTGOG pcb version was 120301, two otdr optical transceivers were online at
the same time,one was in the first four interface,other was in the last four interface (for
example, the first was plugged PON 1, the second was plugged PON 5), the last optical
transceiver could read failed. GTGOG PCB version was 140600, the optical transceiver
may get error information, such as the manufacturer information was Hisense ,it may be
HHsense.
Note: only for GTGOG line card,other line cards were normal..
Symptom: ONU was typec protection,after swap, need to change the active or standby
state, the active and standby state will to be deal with different process, since the code
is a bug, not modify its state before the swap, when after swap, the standby ONU will be
active, but it was still standby, so the ONU will be offline(the state will be from O4 to O2),
and then comes back online(the state will be from O2 to O5), the service may be
interrupt long time.
Note: only for GTGOG and GTGHG line cards, other line cards were normal. The typec
refers to our own typec on the command line, is refers to the two optical transceiver and
one PON MAC, rather than ITU typec, the ONU is only 9806...
Symptom: In GTGHG line card,the first eight PON interface and the last eight PON
interface had more than one ONU at the same time watch on the same multicast, if one
of the user left the multicast, all user of the eight PON interface had watching the same
multicast will also be affected, multicast will interrupt. Such as:
If the user A left the multicast 225.1.1.1, then the user B and user C multicast 225.1.1.1
interrupt, user D, E, F is not affected; If the user F left, the user D and E multicast
225.1.1.1 interrupt, user A, B, C are working. The user A, B, C group in any one to leave
the other users multicast interrupt, another D, E, F group were not affected; Same user
D, E, F group any one to leave the other users multicast interrupt, another A,B,C group
were working.
Fault Description: Under the engineering environment of Wen Zhou Telecom, adopt
V1.2.5P3-version GTGHG and HW’s ONU (HG8245, HG8120C) to interconnect, the
ONU was on-line and off-line frequently, because the OLT side monitored the ONU’s
TIWi alarm, and then set the ONU as Deactive, and then it was on-line once again.
Later, via communicating with the third party, we knew that when HW interconnects with
its own OLT, its DOWi and TIWi thresholds are loosened a bit, with its threshold set as 8
and 16, our OLTs still monitor to generate alarms according to standard 4 and 8,
therefore, TIWi alarm is generated.
Action: When the PON port becomes the working port, the ONU is normal, report once
again, the alarm disappears.
Note: If the alarm has appeared before patched, it can be recovered only after you wait
for next change-over, or it will be normal automatically after the ONU is on-line and off-
line once. Just comprise GTGOC, GTGOE, GTGOG and GTGHG line cards.
Fault Description: When testing GTXOG in the Institute, the task was suspended during
the startup process of the line card. Because of the task order problem when starting the
line card, it probably will cause suspension of line cards, caused them unable to be
started normally. Through code walkthrough, other line cards have the same situation,
just a matter of probability, therefore, all line cards are modified. If the line card has
already been started normally, the problem won’t exist.
Note: If the line card appears the problem, it needs to patch first before activating and
rebooting.
Fault Description: For Singapore Telecom, when PW authentication was used, and
enable sn-bind disable at the same time. Afte3r the ONU was on-line normally for a
while (at least more than 3 minutes), power it off or insert / extract the ONU, and then
the ONU was on-line once again, services would be abnormal, and the ONU might be
on-line and off-line repeatedly, mainly because ONU off-line alarms were not cleared. If
there appear LOSi, or repeated unknown in query status, that is, show gpon onu detail-
info gpon-onu_1/x/x:x, there will appear:
Action: When the ONU is on-line and ONU ID sets up mapping, clear previous alarm
information.
Note: If the problem has already appeared, you need to make the ONU on-line once
again, for example, insert / extract optical fiber, or execute shutdown/no shutdown. All
line cards have similar situation.
2.8.5 Coexistence of Ipv6 and ipv4 was the problem that arp messages
were dropped at downlink
Fault Description: When ip-source-guard was enabled, and IP+MAC mode was
configured, if ipv4 and ipv6 got the IP address at the same time, downlink ipv4 arp
would be blocked, and would be dropped inside the GTGOE, if just ipv4’s IP address
was got, there would be no problem. If only IP filter mode was adopted, there would be
no problem, either. Adopt the following commands to view the filtering mode adopted by
current ip-source-guard:
Note: If the problem has already appeared before patched, after patched, you need to
unbind the IP, and re-acquire IP address, and then it will be normal. Only GTGOE has
modified the problem.
2.8.6 Loopback ping but there was packet loss under Linkstar scenario
Fault Description: In engineering, use linkstar to test: one PON port is connected with
three ONUs, of them, the UNI ports of two ONUs adopt network cables to connect, and
then ping packets on the third ONU, at this time, there was packet loss occasionally.
Networking diagram:
Because the uplink side will deliver flood packets downwards (for example, NMS
initiates to query other IP’s arp packets or unknown flood packets, or MAC addresses
whose destination address is the same as ONU3 ping packet), and upwards via the
looped onu1 and onu2 loopback, maybe it will affect the ping packets on ONU3, causing
packet loss.
Action: When anti-spoofing is enabled, SW part of the line card is set as Uplink Port
Priority; when anti-spoofing is disabled, shut down Uplink Port Priority, to restore it as
being able to migrate.
Note: The patch is customized for linkstar, just for the line card GTGOG and GTGHG.
You need to enable the anti-spoofing function to make it take effect, and at the same
time, the anti-spoofing function and ONU loopback detection function are mutually
excluded: either enable the anti-spoofing function, or enable the loopback detection
function.
Fault Description: In Bao Shan, Yunan province, there appeared the alarm on NMS that
the card’s software hasn’t run, all the PON ports on the line card had LOS alarm,
services would be interrupted temporarily, and NMS alarms:
Note: Just GTGOG and GTGHG line cards had the problem.
Fault Description: After the version was updated to V1.2.5P3 in Tie Ling, Liaoning
province, interconnect with HW’s ONU (HG8120C), the ONU was on-line and off-line
frequently, because the OLT side monitored the ONU’s SFi alarm, and then set the ONU
as Deactive, and then it was on-line once again, other ONUs had no similar situations.
V1.2.3P3 and V1.2.5P3 versions change in terms of SFi threshold, improving it higher.
-3
Because the original (standard widest threshold) is modified into -5, maybe the type of
the HW’s ONU had a large error code, modify the SFi threshold back to 10-3 via the
patch.
Note: GTGOC, GTGOE, GTGOG and GTGHG cards all have the same situation.
Symptom: In Colombia, ETB had the problem that the GTGOG line card’s ONUs failed
to learn MAC addresses, and services were blocked. View the number of the MAC
addresses learned internally, and miscalculate it as already reaching the maximum
value, but actually, there was no or few MAC addresses, therefore, it didn’t learn MAC
addresses any longer, but directly drop flows, which caused services blocked.
Note: Just GTGOG and GTGHG cards have the problem. When patch was active you
need to execute reset card, it became to take effect.
Symptom: After Da Lian Unicom was upgraded to V1.2.5P3 version, service flows of the
third-party ONU were blocked, and we analyzed and found that it was because the flows
transmitted by the ONU were all yellow packets (that is, CFI bit position was 1), while
the color sensitivity mode was configured at the OLT side, at the same time, it was
configured as non-coupling and EBS weight as 0, in this way, the OLT side directly
dropped all yellow packets, which caused services blocked.
Action: Modify green packets and yellow packets as being configured as coupling mode,
and at the same time, set EBS weight as non-0.
Note: Just GTGOG and GTGHG cards have the problem. When patch was active you
need to execute reset card, it became to take effect.
Symptom: In S1 version, after IP-source-guard was enabled, if you batched delete and
add ONUs repeatedly, finally services of some ONUs would be abnormal; via our
analysis, because when deleting repeatedly, IP-source-guard configuration data was
confused, causing the IP addresses of some ONUs to be inconsistent with actual IPs.
Symptom: In engineering, due to the problem of the ONU’s power adapter for linkstar,
the ONU reported DGi alarm constantly via PLOAM messages mistakenly, a great
number of PLOAM messages impacted the normal processing of other ONUs, affected
other ONUs to be on-line normally, causing other ONUs on the line card to be affected.
Action: When monitoring similar ONU, deactivate the ONU, and forbid it to be on-line in
a period of time.
Note: NA.
Symptom: After Chong Qing Telecom was upgraded to V1.2.5P3 version, part of ONUs
failed to register, and the LOIDs of ONUs were bound mistakenly. Via final packet
capturing and ONU, it was confirmed that the problem was related to the ONU, and it
needed OLT to avoid.
Action: Add the function: when the OLT side receive same LOIDs within a rather short
period (less than 3s), it thinks the last ONU’s LOID is correct, while previous LOID is the
LOID in its cache the ONU reserves before and it’s wrong, and it needs to be dropped.
Note: NA.
Symptom: In V1.2.3P3 version, the third-party NMS regularly collected ONU’s receiving
optical power, which caused suspension of GPON line card’s tXponCtrlSub0 task, and
its stacking information:
tt 0x6916308
5e02a4 vxTaskEntry +68 : 34263c ()
342664 XponCtrl_NP_SyncMsgprocess+1c8: 342530 ()
342624 XponCtrl_NP_SyncMsgprocess+188: XponCtrl_NP_AsyncMsgprocess ()
342468 XponCtrl_NP_AsyncMsgprocess+a8 : Xpon_NpAsyncMsgProcess ()
33b7f0 Xpon_NpAsyncMsgProcess+198: XPonNpMsgHandler ()
33aba4 XPonNpMsgHandler+f8 : TransceiverGetRxPower (6914e76, 86303a0)
32db04 TransceiverGetRxPower+28 : GetRxPowerMsgHandle (6914e76, 86303a0)
32e748 GetRxPowerMsgHandle+54 : 32e250 (6914e82, 86303a8)
32e318 GetGponTransceiverInfoFromDriver+374: mux_IoCtrl ()
5d0248 mux_IoCtrl +10 : mux_IoCtrl_Distribute ()
5d0228 mux_IoCtrl_Distribute+74 : 4b2e84 (44, 6914cec, 6914cf8, 0)
4b2f10 macMuxTimerTask+2bc: getOnuRxPowerOltSide (6914cec, 6914cf8)
4a31c4 getOnuRxPowerOltSide+1bc: getOnuMinStatPowerVal ()
Because the input parameters were abnormal, and there was no protection inside,
causing data out of boundary, which led to suspension of tasks. V1.2.5P3 also had the
same problem.
Note: NA.
Symptom: After Shan Dong Telecom was upgraded to V1.2.5P3 version, part of ONUs
failed to register. Symptoms: its state migrated from logging -> losi repeatedly. Seen
from internal analysis at the OLT side, after the ONU migrated to O5 state, it didn’t
receive upstream PLOAM messages, and then deactivate the ONU, and then it was on-
line again. At the ONU side, its state was O2->O3->O4->O2, sometimes, it could reach
O5, and then it received the Deactive message from the OLT, its state migrated to O2.
Symptom: DST tested V1.2.5 P3T13 version, when configurating the VLAN XConnect,
the PTP and L2CP packet were NOT transparent, and at the same time, it can be
learning the MAC address. The previous version for protocol processing was as follow:
1 About PTP packets, they were default transferred to CPU in upstream, NOT transfer to
main contrl board, It was NOT configure. downstream is according to rule which was the
configurated service-port.
2 About L2CP packets, they were forwarded directly to the main contrl board without any
processing, It was NOT configure. downstream is according to rule which was the
configurated service-port.
Action: For protocol packet NOT transparent, Automatically adopt by vlan xconnect
command as follow:.
When vlan xconnect was configured on one port, PTPs and L2CPs packet on the port
were transfered according to the rules of service-port, the service-port which is
configured vlan xconnect flow were not learn MAC address. If there is no configuration
any xconnect on the port, the behavior as before.
Note: NOT enable 1588 and XConnect functions on the same port at the same time.
2.11 FTGKBV125P3T14_r20_dst.pat
Symptom: FTGKB_LX. MVR P3T14 version, only when the xconnect was configuration
on one port, the PTP and L2CP can be translated with the rules of service-port, if there
is no configuration xconnect, there is still old behavior in P3T13 version, the PTP directly
discarded, and L2CP VLAN is not translated directly forwarding. It was NOT satisfied
with DST scenario.
2.12 FTGKAV125P3T13_r0.pat
2.12.1 Reading txPower of the uplink port’s optical modules was abnormal
Symptom: The GDFO uplink ports 17/1 and 17/2 were off. View that the txPower was
“no signal”, and the optical interfaces were unavailable. After the fibers of the two optical
interfaces were cut over to 19/3 and 20/3, it was normal, and then re-insert two optical
modules on 17/1 and 17/2, you could see that the states of the optical modules were
abnormal. It was confirmed on-site that the optical modules using WTD also had the
same problem.
00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00
Register-Data :
5f 00 d3 00 5a 00 d6 00 8d cc 74 04
88 a3 79 2b 9c 40 01 f4 75 30 03 e8
1f 07 02 76 18 a5 03 1a 18 a5 00 1f
15 f7 00 32 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 3c c4 c6 df
c2 a2 96 b9 01 00 00 00 00 7f ff e1
01 00 00 00 01 00 00 00 10 00 00 bf
2c 58 7e e8 0d 28 09 31 00 00 00 00
00 00 02 f8 00 40 00 00 00 40 00 00
00 00 00 00 00 00 00 00
Diagnostic-info:
RxPower : -80.000 (dbm) TxPower : No Signal
Bias-Current : 6.688 (mA) Laser-Rate : 13(100Mb/s)
Temperature : 44.406 (c) Supply-Vol : 3.250(v)
Alarm-thresh:
RxPower-Upper : 3 (dbm) RxPower-Lower : -34(dbm)
TxPower-Upper : 9 (dbm) TxPower-Lower : -14(dbm)
Bias-Upper : 131(mA) Bias-Lower : 0 (mA)
Voltage-Upper : 7 (v) Voltage-Lower : 0 (v)
Temperature-Upper: 90 (c) Temperature-Lower: -45(c)
Action: .
Note: .
Register-Data :
64 00 f6 00 5f 00 fb 00 8c a0 75 30
88 b8 77 24 9c 40 03 e8 88 b8 05 dc
18 a6 03 e8 13 94 04 eb 18 a6 00 4f
13 94 00 64 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 3f 80 00 00
00 00 00 00 01 00 00 00 01 00 00 00
01 00 fc 00 01 00 00 00 00 00 00 5e
1c d0 7f 90 23 58 08 90 00 00 00 00
00 00 02 f8 00 40 00 00 00 40 00 00
00 00 00 00 00 00 00 00
Diagnostic-info:
RxPower : -80.000 (dbm) TxPower : -6.592(dbm)
Bias-Current : 18.096 (mA) Laser-Rate : 13(100Mb/s)
Temperature : -995.281 (c) Supply-Vol : 3.267(v)
Alarm-thresh:
RxPower-Upper : 3 (dbm) RxPower-Lower : -34(dbm)
TxPower-Upper : 9 (dbm) TxPower-Lower : -14(dbm)
Bias-Upper : 131(mA) Bias-Lower : 0 (mA)
Voltage-Upper : 7 (v) Voltage-Lower : 0 (v)
Temperature-Upper: 90 (c) Temperature-Lower: -45(c)
Action: Identify the model of the specified optical transceiver module and process the
fault by software.
Note: None.
2.14 SCXLV125P3T13_r5.pat
Symptom: For SCXL+GTGOG, the inner port uses the trunk mode, and the main control
card uses VLAN+Source MAC+destination MAC, while the line card uses
SVLAN+Source MAC. Due to different hash algorithm, the PPPoE packets from the
same ONU are transmitted to two ports. The line card trunk has a bug and drops the
traffic received on one port. Meanwhile, the line card does not support modifying the
hash algorithm; therefore we need to modify the hash algorithm into SVLAN+Source
MAC on the main control card.
Action: Modify the hash algorithm of the trunk into SVLAN+Source MAC when
SCXL+GTGOG are used.
Note: None.
2.15 GTGOAV125P3T13_r3.pat
GTGODV125P3T13_r3.pat
Symptom: The F601 connected under the GTGOA card connects online and offline
repeatedly at an interval of four minutes. The reason is because LOS. The fault occurs
when the ONU registers by LOID and SN BIND is disabled.
Action: When the ONU registers by LOID and migrates to SYNCMIB, increase
OLT/ONU LOS deletion logic.
Note: None.
Symptom: After the ARP proxy is enabled, it answers ARP requests on the network
side, and the ARP requests sent by other network equipment will receive wrong answer
messages.
Action: Forbid the ARP proxy from answering packets on the network side.
Note: None.
Symptom: By default, 802.1X packets are extracted to the CPU, which causes SNMP
packet loss.
Action: Modify that the 802.1X packets are not extracted to the CPU by default.
Note: None.
Symptom: SNMP ping from the EMS to the C300 fails. The fault is caused by SNMP
buffering packet memory leakage due to SNMP CallList concurrent access.
Note: None.
Note: None.
2.17 GTGOEV125P3T13_r4.pat,
GTGOAV125P3T13_r4.pat,
GTGOCV125P3T13_r4.pat,
GTGODV125P3T13_r4.pat,
GTGOGV125P3T13_r4.pat,
GTGHGV125P3T13_r4.pat
2.17.1 New Service Cannot Learn MAC After the PON Service Profile is
Replaced
Symptom: New service cannot learn MAC after the PON service profile is replaced.
When replacing the profile, the new VPORT inserting into other VPORTs of existing
profile will cause mapping relation disorder between GEMPORT and VPORT (BP), the
new GEMPORT will be deleted, and the service will be interrupted.
Symptom: Services are interrupted when gemport rate limit cpb=pbs, the reason is
because ebs=0 when pbs is less than or equals to cbs.
Action: Modify GEMPORT rate limit configuration, and configure ebs into the larger
value between cbs and 10 when pbs is less than or equals to cbs.
Symptom: “Failure” is returned and the service is interrupted when GEMPORT ID is 129
2.17.4 GTGOA Port With an Even Number Does Not Show MAC
Symptom: MAC is displayed by running the “show mac-real-time” command, but not
displayed by running the “show mac” command.
Symptom: The 9806 is persistently in mib synchronization state and cannot enter
working state.
Action: Modify the OMCI mib synchronization overtime timer from “t0” to “3*t0”. By
default, t0 is 100 seconds. After the patch is installed, the default overtime is 300
seconds. t0 can be modified via the serial port.
Note: None.
Symptom: In multiple Heilongjiang offices, ONUs connected under the GTGHG card are
disconnected and unable to register. Through analysis, we found that the reason is
because the discovery time window latency of the 16 PON ports under GTGHG exceeds
the timer.
Symptom: ONUs are unable to register. It is reported that the ONU SN has been
registered; therefore the ONUs cannot re-register.
Symptom: At the end of July, Hebei Shijiazhuang Xingtang Unicom reported that some
Huawei ONUs connected under the C300V1.2.5P3 GTGOE line card cannot connect
online and are in “LOS--LOGGING—SYNCMIB” state. The fault can be recovered by
running the “shutdown/no shutdown” command.
Action: Find the last LOS ONU, judge whether it has power-off alarms, and judge
whether the interval between the power-off time and LOS time is shorter than 10S.
Optimize the alarm logic.
Note: None.
Symptom: Tod frame skip occurs when performing tod clock testing on the 1588
platform.
Action: Avoid the situation that 1pps and superframe value are inconsistent. Only allow
tod distribution when the two values are consistent.
Symptom: Memory application fails when xPON ONU authentication is being processed,
causing task suspension.
Note: None.
Both OLT1 and OLT2 are under the same Switch and the same VLAN.
Ping from the network gateway to ONU2 fails (ONU1 can ping the network
gateway).
OLT1 disables MFF. Wait for a while, ONU2 can ping the network gateway.
Under normal situations, MFF forwards ARP packets to the user side when it
cannot answer. But now, it is found that C300 V1.2.5P3 forwards the packets to the
network side.
After OLT2 receives the packets, ONU2 MAC is drifted to the uplink port, and later,
the upstream traffic sent by ONU2 is interrupted (OLT2 enables MAC anti-drifting
and upstream port priority).
Through analyzing the packet transmission function, we found that the port sends
packets in both upstream and downstream direction to the member ports within the
aggregation group.
Action: Forbid the ARP proxy answer the packets sent from the network side.
Note: None.
Symptom: Use the PuTTY.exe software to telnet the NE and configure Column into 600,
and choose “Forbid resizing completely”, the main control card suspends during the
telnet process.
Note: None.
Symptom: Performing GetBulk operations when multiple oid variables are bound causes
process dead loop and CPU utilization too high.
Note: If the card is in dead loop, reboot the card and then activate the patch.
Symptom: The EMS cannot display “patch-saved”. After the patch is installed, the EMS
displays the patch version stored in the file system.
Note: None.
Symptom: GDFO port rate is displayed as 100M when it is configured into 1000M, and
the port cannot join the aggregation group.
Action: Modify the port working rate logic, and update the actual rate when it is
configured into mandatory. Modify the software logic.
Note: None.
Symptom: After being upgraded to V1.2.5P3T13, two NEs are suspended at Shaanxi
Telecom. It is because the NE is upgraded but the EMS is not, and C300 is managed by
the C220 EMS. Through analyzing the crash files and code, we found that the problem
is caused by incomplete protection of shelf and card.
Action: Increase the validity check of the SNMP index value, check the shelf and card
range, and return error in case of exception.
Note: If the NE task is suspended, reboot the card and activate the patch.
Symptom: The “show gpon remote-onu mac causes” command causes memory out of
range.
Note: If memory out-of-boundary occurs, reboot the card and activate the patch.
Symptom: When the patch version is illegally deleted, “failure” will return when
downloading the patch version with the same name.
Note: None.
2.18.10 Patch Activation Causes SCXL Card Cannot Enable MFF Due to
Memory Out-of-boundary
Symptom: Patch activation causes memory out-of-range, and causes the SCXL card
cannot enable the MFF function.
Note: None.
2.19 GTGOEV125P3T13_r5.pat,
GTGOAV125P3T13_r5.pat,
GTGOCV125P3T13_r5.pat,
GTGODV125P3T13_r5.pat
GTGOGV125P3T13_r5.pat,
GTGHGV125P3T13_r5.pat
Symptom: The speed that the IGMP traffic transmitted to the OLT uplink port is slow.
Symptom: After upgrading from V1.2.3P3 to V1.2.5P3T13, new F607 that uses LOID
registration mode needs five minutes to connect online. The provisioning process is as
follows: connect the F607, query the UNCFG configuration, and modify the ONU LOID
and CLI LOID the same. The fault is caused by:
The LOID configured via the CLI is A. According to the general provisioning
process, connect the ONU (the ONU LOID is B), the OLT obtains the unconfigured
information (LOID) and place it into the aging table.
Modify the ONU LOID into B. According to the current logic, obtaining the UNCFG
information including LOID will activate the ONU frequently. When the aging time
comes (five minutes), the ONU is re-activated, the OLT obtains the new LOID, and
the ONU registers.
Action: Modify the patch according to the provisioning process. UNCFG ONU does not
rely on the aging time. Each time when the OLT discovers UNCFG-ONU, it obtains
LOID.
Symptom: After upgrading to V1.2.5P3T13, the F822 MDU connected under the
GTGOE card at some Jinan Telecom sites is unable to connect online. The OLT fails to
obtain the PW from the OLT, causing the registration process is unable to be completed.
The problem cannot be solved after rebooting the card or line card. Through further
research, we found that the OLT is unable to obtain the PW when the ONU is allocated
with ONUID of 3 or 5. Because the RANGING TIME message received by the ONU
does not enter O5; therefore the PW is not reported. However, the ONU can report the
PW and register normally if its ONUID is 7 or 10. Because the problem cannot be solved
within a short period, the OLT needs to increase the prevention methods.
Action: Delete the MAC ONU when failing to obtain the PW. Delete the SN and ONU
binding relationship, skip the ONUID that causes registration failure, and allocate from
ONUID+1. When reaching the maximum parameter threshold 127, allocate (1). If
ONUID 1 to 127 are all allocated, allocate ONUID 0.
Note: Affect the GTGOG/GTGHG cards. The GTGOE card has the problem, but it has
solved by the r2 patch.
Through analysis, we found that the OMCI alarm processing has memory leakage.
When the OMCI processes the alarms, it removes the meid when storing the firstly
reported alarm, causing each alarm applies memory and triggering memory leakage.
Note: Affect all GPON line cards. Reboot the card to solve the problem.
Symptom: The old CPLD version working with 140900/140901 PCB will cause the card
unable to work.
Action: Upgrading the CPLD is not allowed if the PCB version is 140900/140901.
Symptom: When NMS configure ONU profile, the index of OID is 0, and the length of
index is larger than 1. NE will return “NO_CREATION ” error.
Note: None.
2.21 GTGOE.FW(V1.2.5P3T20)
2.22 GTGOGV125P3T13_r22.pat
Symptom: After configuring rate limit for vport, the rate of FTP service is not stable.
Action: Enable a special mode of QOS by patch.
Note: After active the patch, it will work after rebooting the card. This patch is only for TRUE of
Thailand
2.25 GTGOEV125P3T13_r11.pat
Symptom: :After updating C300V1.2.5P3 in Portugal, multicast services of part of GPON
ONUs were blocked, and fail to restore, you need to delete multicast under the remote management
and re-configure it to be able to restore the service.
Action: Use this patch to make sure the multicast VLAN is down to remote ONU when the line
card is rebooted.
Attention:When you change the mode of smartgroup,you must no the port from the smartgroup
first.After changing the mode ,you add the port again.
2.29 SCXLV125P3T13_r27.pat
Attention: If main board has suspended, you should active the patch and reboot the OLT.
2.30 SCXNV125P3T13_r20.pat
Attention: If main board has suspended, you should active the patch and reboot the OLT. Please
deactive and delete old patch before downloading and active the new patch
2.31 GTGOEV125P3T13_r20.pat
Confirm the version of the NE and which patches you need to load.
Study related patch specifications carefully, to view whether there are special
requirements.
1. Configure related FTP server, and the configuring process is the same as NE
version upgrading.
2. Download a patch.
Command Format:
Command Function:
Download the specified patch file into the main control card. The status of the patch
downloaded successfully is deactivated, of them, patchname is the name of the
patch file, with its value range as 1-32 characters. After downloaded successfully,
you can view the patch information via show patch-saved.
Execution results:
Examples
1, Download the patch of the main control card
ZXAN#download patch scxlv125p3t13_r0.pat
Downloading from host(10.63.196.193)
Transfering file SCXLV125P3T13_r0.pat ...
......[Successfully]
Command Format:
patch activate [patchname]
Command Function:
Activate the patch of the main control card: The main control card activates the specified
patch to make it effective, if there is a standby one, activate the specified patch of the
standby one at the same time;
Activate the patch of the line card: Specify the patch name, and activate all the line
cards supporting the patch inside the shelf.
Active / standby synchronization: The command is executed on the active card first,
after executed successfully, it will continue to be executed on the standby card
automatically.
Execution results:
De-activate a patch
Command Format:
patch deactivate patchname
Command Function:
De-activate the patch which has been effective. Of them, patchname is the name of the
patch file, with its value range as 1-32 characters.
De-activate the patch of the main control card: The main control card de-activates the
specified patch to make it effective, if there is a standby one, de-activate the specified
patch of the standby one at the same time;
De-activate the patch of the line card: Specify the patch name, and de-activate all the
line cards supporting the patch inside the shelf.
Active / standby synchronization: The command is executed on the active card first,
after executed successfully, it will continue to be executed on the standby card
automatically.
Execution results:
Delete a patch
Command Format:
delete patch [patchname]
Command Function:
Delete the specified patch file, if the status of the patch is Activated, it is not allowed to
delete. Of them, patchname is the name of the patch file, with its value range as 1-32
characters.
Active / standby synchronization: The command is executed on the active card first,
after executed successfully, it will continue to be executed on the standby card
automatically.
Execution results:
Confirm to delete?[yes/no]:y
Start deleting file
deleting SCXLV125P3T13_r0.pat ..
[Successfully]
After upgrade, you can judge whether the patch is upgraded successfully via inquiring
the status of the patch.
Command Format:
show patch-saved
Command Function:
Show the patch files saved on the active equipment, including patches of the main
control card and the line card. If there is a standby one, show the patch information
saved on the standby one at the same time.
Active / standby synchronization: The command is executed on the active card first,
after executed successfully, it will continue to be executed on the standby card
automatically.
Execution results:
ZXAN#show patch-saved
Patch infomation on master board
Loc FileName PatchTag BuildTime PatchLen AdminState
------------------------------------------------------------------------------------------
Command Format:
Command Function:
Show the status of patches which have been activated already, including the main
control card and the line card.
Execution results:
ZXAN#show patch-running
Loc FileName PatchTag OperateTime PatchState
-------------------------------------------------------------------------------
1/1/10 scxlv125p3t13_r0.pat 2.0 2001-01-09 01:17:58 ACTIVE