SDH Alarms all

Tejas Networks India Ltd., 2006, Proprietary Information

Tejas NetworksSDH Alarms


Organisation of Slides

SDH section hierarchy SDH objects, nomenclature Downstream and Upstream Alarms understanding rules RS alarms MS alarms HP / LP alarms Description of Alarms Alarm Masking and Suppressed Secondary Alarms Alarm propagation examples


There are four sections Regenerator Section (RS), Multiplex Section (MS), Higher Order Path Section (HP), and Lower Order Path Section (LP)

RS is a part (section) of the optical fibre network, within which RSOH part of SDH frame is NOT opened

MS is a part (section) of the optical fibre network, within which MSOH part of SDH frame is NOT opened

HP is a part (section) of the optical fibre network, within which higher order VC part of SDH frame is NOT opened (it may be opened only for interpreting HOPOH)

LP is a part (section) of the optical fibre network, within which lower order VC part of SDH frame is NOT opened (it may be opened only for interpreting LOPOH)

SDH Section Hierarchy


SDH Section Hierarchy (contd.)

Points to Remember: Without opening RS, one can not do operation with MS and/or open MS Without opening MS, one can not do operation with HP and/or open HP Without opening HP, one can not do operation with LP and/or open LP

Consequences So, for Tejas nodes, even if one is making a VC4 level pass-through (an operation with HP

without opening it), he/she is opening MS & therefore terminating the MS One can change any HPOH field (e.g., J1 transmitted trace) only when one is opening HP (e.g., VC12

level cross-connect exists on AU4 mapping), but not when HP is not disturbed (e.g., VC4 level pass-through on AU4 mapping)

Points to Remember: For Tejas nodes, for AU4 mapping, one can make VC4 and VC12/VC11 level and not VC3 level

pass-through for E1/DS1 trafficConsequences

If in a STM-1 node, multiple (say, 18) E1/DS1 traffic have to be passed-through with some other trafficadded/dropped from that node, one has to make multiple (18) VC12 level pass-through


Section Hierarchy (examples)

Example 1

Example 2

ADM 1 ADM 2Reg.

MS

RS RS

ADM 1 ADM 2 ADM 3

RSRS

MSMS (STM 1)

(VC4)


Example 3 (for AU4 mapping only)

Section Hierarchy (examples) ( contd.)

DA B CVC12 VC12

E1 E1

VC12VC4

RS

MS

RS RS

MS MS

HPHP

LP



Example 4a (for STM1 capacity & AU4 mapping only)

H

A B

C

D

E

F

G

#1 E1 between A & E#2 E1 between F & H

E3 between F & G

E1

E1VC12VC12

E1

E1

VC 12

VC 12E3

E3 VC 3

VC3

Reg.

RS A-B, B-C, C-D, D-E, F-B, C-G, E-HMS A-B, B-C, C-E, F-B, C-G, E-H

HP A-B, B-C, C-E LP A-E

HP F-B, B-C, C-GLP F-G

HP F-B, B-C, C-E, E-HLP F-H



Example 4b (for STM4 capacity & AU4 mapping only)

H

A B

C

D

E

F

G

#1 E1 between A & E#2 E1 between F & H

E3 between F & G

E1

E1VC12VC12

E1

E1

VC 12

VC 12E3

E3 VC 3

VC3

STM # 1

STM # 2

----- VC 4

Reg.

STM # 2

Within STM # 1

STM # 1

RS A-B, B-C, C-D, D-E, F-B, C-G, E-HMS A-B, B-C, C-E, F-B, C-G, E-H

HP A-E LP A-E

HP F-C, C-HLP F-H

HP F-C, C-GLP F-G


SDH objects, nomenclature

3 different kinds of objects: STM port (STM-1 / STM-4 / STM-16) AU (AU-3 / AU-4 / AU-4-4c / AU-4-16c) Higher-order object

(present even if no HO cross-connect) TU (TU-11 / TU-12 / TU-2 / TU-3) Lower-order object

(present only if LO cross-connect exists) Nomenclature STM-1 chassis slot port (these fields are product specific) AU-4 chassis slot port STM # 1 AU-3 chassis slot port STM # K (for AU-3 mapping) TU-3 chassis slot port STM # K (for AU-4 mapping) TU-2 chassis slot port STM # K L TU-12 chassis slot port STM # K L M (M = 1 to 3) TU-11 chassis slot port STM # K L M (M = 1 to 4)

Note: STM # = 1 (for STM-1)= 1 to 4 (for STM-4) like that, K = 1 to 3, L = 1 to 7


Downstream & Upstream

Downstream direction for a fault conditionAlong the direction of fault condition received

OR Towards the Back-plane of the node receiving fault condition

Upstream direction for a fault conditionOpposite of the direction of fault condition received

OR Away from the Back-plane of the node receiving fault condition

Downstream & Upstream direction for a node not fixedDepends on direction of fault condition (abbreviated as FC)

ADM 1 ADM 2 ADM 3

FC 1Downstream

Upstream

FC 2

Upstream

Downstream


Alarm Understanding Rules

Rule 1

Rule 2

FC 1Alarm reported

Alarm reportedFC 1

ADM 1 ADM 2

ex. a

ADM 1 ADM 2

ex. b

Alarms reported are alarms received

Alarms are reported on SDH Objects


Alarm Understanding Rules (contd.)

Rule 3

ADM 1 ADM 2

ex.

3a. No Object => No Alarms reported

FC on TU12 (1-1-1)

NO TU12(1-1-1)

3b. Object Mismatch => No Alarms reported

FC on TU12 (1-1-1)

TU11(1-1-1)

ADM 1 ADM 2

ex.

Note:These two examples are not possible for AU object

WHY?

See slide 9

NO Alarm reported for FC on TU12 (1-1-1)

NO Alarm reported for FC on TU12 (1-1-1)



Rule 44a. No PT XC => No Alarms pass-through

FC on AU4 (1)

NO VC4PT (1)

Alarm reported for FC on AU4 (1)

FC on TU12 (1-1-1)

ADM 1 ADM 2 ADM 3

ex. a

ADM 1 ADM 2 ADM 3

ex. b

NO Alarm pass-through

NO VC12PT (1-1-1)

NO Alarm pass-throughNO Alarm reported for FC on TU12 (1-1-1)



4b. Bigger PT XC => No Alarms reported & Alarm pass-through

FC on TU12 (1-1-1)Alarm pass-through for

FC on TU12 (1-1-1)

NO Alarm reportedfor FC on TU3 (1)

VC4

ADM 1 ADM 2 ADM 3

ex. a STM-1 links

4c. Smaller PT XC => No Alarms reported (always ??) &Alarm pass-through but on smaller object

FC on TU3 (1)VC12

(1-1-1)

NO Alarm reportedfor FC on TU12 (1-1-1)

ADM 1 ADM 2 ADM 3

ex. b STM-1 links

Alarm pass-through for FC on TU12 (1-1-1)

What if Same size PT XC ?


RS Alarms

RS alarms are those, which can be reported even by a pure Regenerator(who has privilege of opening (interpreting & rewriting) only RSOH)

LOS (Loss of Signal)based on whole RSOH

LOF (Loss of Frame)based on A1, A2 bytes

TIM (Trace Identifier Mismatch)based on J0 byte

SF (Signal Fail)based on B1 byte

SD (Signal Degrade)based on B1 byte

D3D2D1F1E1B1J0A2A1

RSOH bytes

Note: The order in which the alarms are written is important,as we will see later while discussing Alarm masking


MS Alarms

MS alarms are those, which can be reported by a Add-Drop Multiplexer, irrespective ofcross-connect configuration

(who has privilege of opening (interpreting & rewriting) RSOH, MSOH, AU pointers plusopening HOPOH(s) / TU Pointers / LOPOH(s) depending upon cross-connect configuration)

AIS (Alarm Indication Signal)reported based on K2 byte -- bits 6,7,8

SF (Signal Fail)based on B2 bytes

SD (Signal Degrade)based on B2 bytes

RDI (Remote Defect Indication)based on K2 byte -- bits 6,7,8

MSOH bytes

K2K1B2D6D5D4D9D8D7

E2M1S1D12D11D10

Note 1: The order in which the alarms are written is important, we will see later while discussing Alarm maskingNote 2: MS-AIS is also called Line-AIS or AIS on STM port

MS-RDI is also called Line-RDI or RDI on STM port


HP / LP Alarms

HP / LP alarms are those, which can be reported by a Add-Drop Multiplexer, havingHO / HO & LO object (LO object => LO cross-connect)

(who has privilege of opening (interpreting & rewriting) RSOH, MSOH, AU Pointers plusat least interpreting HOPOH(s) / opening (interpreting & rewriting) RSOH, MSOH, AU Pointers, HOPOH(s), TU Pointers plus at least interpreting LOPOH(s)depending upon cross-connect configuration)

HP-AIS reported based on H1, H2 bytes

HP-LOP (Loss of Pointer) based on H1, H2 bytes

HP-UNEQ (unequipped) based on C2 byte

HP-TIM based on J1 byte

HP-SF based on B3 byte

HP-SD based on B3 byte

HP-RDI based on G1 byte -- bit 5

Note 1: Same as beforeNote 2: HP-Alarm is also

called AU-Alarmor Alarm on AU

LP-Alarm is also called TU-Alarmor Alarm on TU

K3F3H4F2G1C2B3J1

N1

HOPOHbytes

H1, H2, H3 AUPointer bytes


HP / LP Alarms (contd.)

LP-AIS reported based on V1, V2 bytes

LP-LOP based on V1, V2 bytes

LOM (Loss of Multiframe) based on H4 byte bits 7,8

HP-PLM / SLM (Payload / Signal Label Mismatch)based on C2 byte LP-UNEQ based on V5 byte bits 5,6,7

LP-TIM based on J2 byte

LP-SF based on V5 byte bits 1,2

LP-SD based on V5 byte bits 1,2

LP-RDI based on V5 byte -- bit 8

LP-PLM / SLM based on V5 byte bits 5,6,7

Note 1: Same as beforeNote 2: Whole of this slide assumes

TU2/TU12/TU11 for LP. If thereis TU3 with AU4 mapping, thenalso it is LP but Pointers & POHbytes will be like HO

K4N2J2V5

LOPOH bytes

V1, V2, V3 TU Pointer bytes


Description of Alarms

LOSReceived power is less than Laser receiver sensitivity (All bits interpreted as 0)

ADM 1 ADM 2

ex. TxRxRxTx

LOSTx off / misconnectivity Rx off / misconnectivityFiber Cut

Received power is less than Laser receiver sensitivity

(Low power transmitted, Span is longer than specified, Fiber gets deformed etc. etc.)

LOFAnything other than F6 28 (Hex) in any (?) of the A1 A2 bytes (within a STM frame)

-- for consecutive 5 frames (625 Ms) OOF (Out of Frame) clearing 2 frames-- for consecutive 24 frames (3 ms) LOF clearing 24 frames

Note: Prolonged LOS => LOF, but not always LOF => LOS(this fact will be used as one of the Alarm Masking logic later)

LOS clears when 2 consecutive framing patterns are received & no new LOS condition is detected


Description of Alarms (contd.)

TIM (J0)Received J0 trace (1/16 byte(s)) != Expected J0 trace (1/16 byte(s))

Note: For both SF & SD, alarm clearing threshold is 1 decade lower than generation threshold, e.g., Gen. Thr. is 1 in 1000 or higher => Clg. Thr. is 1 in 10000 or lower

SF (B1/B2/B3/V5)Equivalent BER exceeds alarm generation threshold ( 1 in 10 / 1 in 10 / 1 in 10 )3 4 5

5 9SD (B1/B2/B3/V5)Equivalent BER exceeds alarm generation threshold ( 1 in 10 to 1 in 10 )

P1

P2

A B C

Rx trace = C to B

Rx trace = A to B

Tx trace = A to B

Exp trace = A to B

Tx trace = C to BExp trace = C to B



Generation of AIS & RDI Upon Receiving traffic affecting RS alarm, a Reg.

generates AIS towards downstream side(all 1 in whole STM frame)

Upon Receiving traffic affecting RS alarm, a ADMgenerates MS-AIS towards downstream side

(all 1 in whole STM frame minus RSOH)& generates MS-RDI towards upstream side

(in K2 byte b6 -- b8 set as 110)

Upon Receiving traffic affecting HP alarm, a ADMgenerates AU-AIS towards downstream side

(all 1 in whole AU)& generates HP-RDI towards upstream side

(in G1 byte b5 set as 1)



Note: 1) For generating MS-AIS / AU-AIS / TU-AIS, the ADM need not be a term. equip. for MS / HP / LP2) Upon receiving MS-AIS / AU-AIS / TU-AIS also, the ADM generates MS-AIS / AU-AIS / TU-AIS

towards downstream & generates MS-RDI/HP-RDI/LP-RDI towards upstream3) Some alarms are by default traffic affecting or non traffic affecting, whereas

some alarms can be made traffic affecting by user action

Bytes and bits involved in Reception for RDIs remain unchanged

Upon Receiving traffic affecting LP alarm, a ADMgenerates TU-AIS towards downstream side

(all 1 in whole TU)& generates LP-RDI towards upstream side

(in G1 byte b5 set as 1 for TU3 || in V5 byte b8 set as 1 for TU2/12/11)

Reception of AIS & RDI (condition should persist for consecutive 3 to 5 frames)

Reception for MS-AIS in K2 byte b6 -- b8 received as 111for AU-AIS All 1 in H1, H2 bytes (for TU3 AIS also)for TU-AIS All 1 in V1, V2 bytes (TU2/12/11)



Example of generation of AIS, RDI

ADM

Any traffic affecting RS Alarm or MS-AIS (Rx)MS-AIS (Gen)

MS-RDI

Any traffic affecting HP Alarm or AU-AIS (Rx)AU-AIS (Gen)

HP-RDI

Any traffic affecting LP Alarm or TU-AIS (Rx)TU-AIS (Gen)

LP-RDI

Example of reception of TU-AIS, LP-RDI

ADM 1 ADM 2 ADM 3E1 E1

VC12 VC12 VC12

TU-AIS (Rx)

LP-RDI (Rx)

Any traffic affecting RS/HP/LP Alarm



AU/TU-LOP (AU-LOP is not reported in Tejas nodes, as always valid AU pointer values are sent)8/9/10 consecutive invalid AU/TU pointers received or8/9/10 consecutive NDF (New Data Flag) received (other than in a concatenation indicator)

E4 E4VC4 VC4

AU-LOP

AU-LOP

(cleared when 3 equal valid pointers received)

E1 E1VC12 VC12

TU-LOP

TU-LOPADM 1 ADM 2 ADM 3

Ex.

ADM 1 ADM 2 ADM 3

Ex.



HP/LP-UNEQAll 0 in C2 byte for at least 5 frames (for AU4/AU3/TU3)000 in V5 byte, bits 5,6,7 for at least 5 multi-frames (for TU2/12/11)

ADM 1 ADM 2

ex. AU Sig. Label UNEQuipped

UNEQuipped AU Sig. Label

AU has no XC

AU has no XC

HP-UNEQ

HP-UNEQ

ADM 1 ADM 2

ex. E1

VC12AU has no XC


AU Sig. Label TUG-structured

HP-UNEQ



LOMMultiframe information not recovered from H4 byte (bits 7,8) for 1 to 5 ms

(i.e., 2 to 10 multi-frames)

TIM (J1/J2) (Default action is to Ignore TIM)Concept is like TIM (J0), but

a) Remember Section Hierarchy Tx trace (J1/J2) can not be edited within a HP/LP

Note: LOM is an alarm concerning LP, but inferred from HOPOH byte so, it will be reported on a HO object

DA B CVC12 VC12

E1 E1

VC12VC4

Tx trace can be edited for J0, J1, J2 all

Tx trace can be edited for J0 only

Tx trace can be edited for J0, J1 only

b) All trace lengths are now 16/64 bytes



HP/LP-PLM (SLM) (Default action is to Report PLM, but no Downstream AIS)Mismatch in own and received signal label

in C2 byte for at least 5 frames (for AU4/AU3/TU3)in V5 byte, bits 5,6,7 for at least 5 multi-frames (for TU2/12/11)

ADM 1 ADM 2

ex.

AU Sig. Label TUG-structured


TUG-ST

UNEQ

TUG-ST

UNEQ

HP-PLM (SLM)

HP-PLM (SLM)

E1VC12

AU has no XC

Asynch. C4

TUGSTE1

VC12 VC12VC4E1

ADM 1 ADM 3

ex.

ADM 2

Asynch. C4

TUGST

Asynch. C4

TUGST

Asynch. C4

TUGST

HP-PLM (SLM) on all ports


Masking of Alarms

Why?Do not want to crowd the alarm reporting page ( and thereby confuse the user) with those alarms, not required for unearthing the root causeWhen? (The logics)

Logic 1 (when the alarms are related)if ( FC1 ==> FC2 but FC2 =/=> FC1 )

then ( Mask FC2 in presence of FC1 )Note: When FC1 clears, FC2 may or may not clear in the later case FC2 will be reported now

ex. 1a) LOS ==> LOF but LOF =/=> LOS 1b) LOS ==> HP-UNEQ but HP-UNEQ =/=> LOS2) AU-AIS reported because of MS/AU-AIS generated

==> HP-RDI andTU-AIS & LP-RDI(s) reported (if TU object(s) are there) but not vice-versa

4) AU/TU-AIS reported ==> AU/TU-LOP but not vice-versa

3) HP-UNEQ because of no XC at other end==> TU-LOP(s) (if TU object(s) are there) but not vice-versa


Masking of Alarms (contd.)

Note: When FC1 clears, FC2 will be reported

Logic 2 (when the alarms are not related)if ( FC1 has higher priority than FC2 )

then ( Mask FC2 in presence of FC1 )ex. 1) AU/TU-LOP has higher priority than HP/LP-UNEQ

(if one is not getting the starting location of VC, how to look at what is happening within VC)2) HP/LP-TIM, if action is chosen as Report TIM, Downstream AIS (i.e. traffic affecting)

has higher priority than HP/LP-RDI(first correct received problem, then only look for problem in other direction)3) HP/LP-TIM has higher priority than HP/LP-PLM(first correct mis-connection, then see signal label problem within correct correction)

4) HP/LP-UNEQ has higher priority than HP/LP-TIM (even if traffic affecting)(what to gain by correcting mis-connection, if even after that traffic can not be carried)


Secondary Suppressed Alarms (SSA)

AIS and RDI are secondary alarms they are indications, not root causes

E1VC12 VC12VC12

E1ADM 1 ADM 3

ex.

ADM 2

These alarms on a pass-through node is normally not reported in the main alarm page,they are reported in a separate page called suppressed secondary alarms page

These alarms on a path terminating node is reported in the main alarm page as terminating alarmsAU-AIS and HP-RDI are not suppressed, even for pass-through nodes, for Tejas products

Traffic affecting FC TU-AIS (terminating)

LP-RDI (terminating) LP-RDI (SSA)


Alarm Propagation Examples

For every example, Assumption(s) is/are stated Root Cause(s) is/are stated Diagrammatic representation is made (OFCs are shown in cyan) Alarm(s) generated / condition(s) generated for reporting alarms is/are

shown in black Alarm(s) existing at a port is/are shown in red

Alarm(s) masked at a port is/are covered with

Alarm(s) reported at secondary supprressed alarm page is/are shown in pink, italicised

Note(s), whenever required is/are mentioned in green


Alarm Propagation Examples (contd.)

Example 1

A B

Assumption: AU-4 Mapping on both ports Root Cause: NO XConnect on both ports

AU4 Signal Label Unequipped

HP-RDI

HP- UNEQ

HP-RDI


HP- UNEQ

HP-RDI

HP-RDI

Note: 1) if AU-3 mapping, then what happens?2) In newer version of Tejas software, UNEQ is not reported for this root cause



HP-RDI

HP- UNEQ


Signal Label TUG-structure

HP-SLM

HP-RDI

TU-LOP

Example 2Assumption: AU-4 Mapping on both ports, Root Cause: NO XConnect on the port of B

A BE1

VC12

Invalid TU Pointer value

LP-RDI

Note: LP-RDI is not reported on B (See Rule 3a)

HP-SLM default action is report SLM, no downstream AIS


LOS

MS-AISAU-AISTU-AIS

MS-RDIHP-RDILP-RDI


VC-12 VC-12

E1 E1

A CB (Reg.)

Example 3Assumption: AU-4 Mapping on both ports of A & C Root Cause: Fiber cut in the link from A to B

AIS

MS-RDIHP-RDILP-RDINote: The Reg. can not generate any RDI

Actually at C, AU-AIS & TU-AIS conditions are also received


LOS

MS-RDIHP-RDILP RDI


MS-AIS

LP RDI MS-RDIHP-RDI

E1 E1

VC-12 VC-12

A CB

Example 4Assumption: AU-4 Mapping on all ports Root Cause: Fiber cut in the link from A to B

VC-12

ADM B VC-12 PT

TU AIS

Note: Only TU-AIS is reported on Node C (See Rule 4c)

LP RDI

LP-RDI on B is SSA


LOS

MS-RDIHP-RDILP RDI


MS-AIS

LP RDI MS-RDIHP-RDI

E1 E1

VC-12 VC-12

A CB

Example 5Assumption: AU-4 Mapping on all ports Root Cause: Fiber cut in the link from A to B

VC-4

ADM B VC-4 PT

Note: Only AU-AIS is reported on Node C (See Rule 4c)LP-RDI on B is not reported (See Rule 3b)

AU AISTU AIS


Invalid TU Pointers (1-1-2)

TU-LOP (1-1-2)A DCB

E1(2)

VC-12 (1-1-2)

Example 6Assumption: AU-4 Mapping on all ports Root cause: NO XConnect on B, C & D for (1-1-2)

E1 (1)E1 (1)

VC-12 (1-1-1)

LP RDI (1-1-2)

Note: Why E1(1) is shown?LP-RDI is not reported on B (See Rule 3a)




TU-LOP (1-1-2)

LP RDI (1-1-2)

Note: LP-RDI at node B is secondary suppressedTU-AIS at node A is reported as terminating alarm


VC-12 (1-1-2)

A DCB

Example 7Assumption: AU-4 Mapping on all ports Root cause: NO XConnect on C & D for (1-1-2)

E1 (1)E1 (1)

VC-12 (1-1-1)

E1(2)

VC-12 (1-1-2)TU-AIS (1-1-2)

TU AIS (1-1-2)

LP RDI (1-1-2)

LP-RDI(1-1-2)



TU-LOP (1-1-2)

LP RDI (1-1-2)

Note: K-L-M value need not remain same throughout a particular LP, alarms will be reported accordingly on different objects


TU-AIS (1-1-2)

TU AIS (1-1-2)

LP RDI (1-1-2)

LP-RDI(1-1-2)

VC-12 (1-1-2)

A DCB

Example 8Assumption: AU-4 Mapping on all ports Root cause: NO XConnect on C for (1-1-2)

E1 (1)E1 (1)

VC-12 (1-1-1)

E1(2)

VC-12 (1-1-2) E1(2)

VC12(1-1-2)


TU-LOP (1-1-2)

LP RDI (1-1-2)


Invalid TU Pointers(1-1-1)

TU-LOP(1-1-1)

LP-RDI(1-1-1)

Note: LP-RDI from A is not reported on B (See Rule 3b).Why assumption on SLM?


A CB

VC-12(1-1-1)VC-4 VC-12(1-1-2)

VC-12(1-1-2)E1 (1)

E1 (2)E1(2)

Example 9Assumption: AU-4 Mapping on all ports, Root cause: NO XConnect on C for (1-1-1)

VC4 PT at node B,For each port, HP-SLM default action is ignore SLM


LOS

MS-RDIHP-RDI

LP RDI

TU AIS

LP RDI

MS-AIS

LP RDI MS-RDIHP-RDI


VC-12VC-12

VC-12

E1 E1

A CB

D

Example 10 (with SNCP)Assumption: AU-4 Mapping on all ports Root cause: Fiber-cut in the link from A to B

W A-B-C, P A-D-C

VC-12

Note: SNCP is always uni-directional & forTejas, it is 1+1


Thank YouThank YouThank YouThank You

SDH Alarms all

Documents