Andre et al,
B3 errors can be from dirty fiber. B3 errors can also be from a broken
POS card. B3 is a byte, and is most definitely not set by telco
equipment, at least not in the original poster's situation. This is by no
means intended to be a full introduction to sonet, but it will hopefully
help.
SONET link (logical breakdown):
PTE--------------------------------------PTE (B3)
LTE----------LTE-------------LTE---------PTE (B2)
STE---STE----STE-----STE-----STE---STE---STE (B1)
PTE=Path Terminating Equipment (Router)
LTE=Line Terminating Equipment (MUX)
STE=Section Terminating Equipment (REGEN)
B1 and B2 are each bytes contained in the Section and Line overheads of
the SONET frame. B3 is a byte contained within the Path overhead (which
is contained within the SPE (synchronus payload envelope)).
Quick breakdown of sonet frame:
<--3 bytes--|-----------------87 bytes-------------->
+---+---+---+---------------------------------------+ -
| (B1) | | |
| Section | Synchronus | |
| Overhead | Payload | 9
+-----------+ Envelope | |
| Line | | Rows
| Overhead | | |
| (B2) | | |
+-----------+---------------------------------------+ -
SPE (synchronus payload envelope):
<--3 bytes--|------------84 bytes--------->
+-----------+-----------------------------+ -
| | | |
| STS Path | Data Payload | 9
| Overhead | (IE: IP packets) | Rows
| (B3) | | |
+-----------+-----------------------------+ -
So anyways, now you have an idea of what a sonet frame looks like. There
are many seperate control and telemetry bytes in the various overheads
that I didn't mention, but it is sufficient to know that they are there,
along with the B1,B2,B3 bytes.
B3 byte:
+---+---+---+---+---+---+---+---+
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
+---+---+---+---+---+---+---+---+
Each bit is an even parity bit caclulated over each respective bit of the
payload. (eg: bit 1 is even parity for all bit 1's in the payload).
So, the B3 byte is roughly a parity check of the data payload of the SPE
frame. The B1 and B2 bytes are also parity checks of their respective
payloads.
Additionally, the B1 and B2 bytes are regenerated at every device
encountered on their respective layer. (ie: B2 byte is recalculated
everytime the sonet frame traverses a mux (LTE), B1 byte is recalculated
verytime the sonet frame traverses a regen (STE))
What this boils down to:
If you see the number of B3 errors incrementing, but the B2 and B1 errors
remain steady, then the problem is in the telco network, beyond the first
LTE. If B3 and B2 are incrementing, but not B1, the problem is beyond the
first STE. This information is helpful when trying to help the telco
isolate the fault.
I hope someone finds all of the above helpful.
--phil
On Tue, 19 Sep 2000, Andre Dieball wrote:
Hi
B3 errors are definetly NOT from a dirty fiber ....
B3 is a bit which is set in the SONET frames (i believe), so the B3
comes from your carrier or the POS card is broken.
What is your carrier? I know that there was a problem with DWDM
equipment in the past at our carrier. There was a failure in their
bandwidth manager, which causes that the value of the "B3 bit" as
swapped without any reason.
Do you also see these B3 threshold exceeds on "the other end" of the
circuit?
Have you had a look in the Bugwatcher on CCO?
Rgds
Andre
> -----Original Message-----
> From: tishi@ziplink.net [mailto:tishi@ziplink.net]
> Sent: Dienstag, 19. September 2000 21:46
> To: cisco-nsp@puck.nether.net
> Subject: SONET-4-ALARM everyday
>
>
> Hi all,
>
> I am getting syslog entries listed below everyday.
> I had a TAC case but all I got from CISCO was "it's a dirty fiber".
> If that is the case, how a dirty fiber affects us everyday
> almost at the same time? I contacted a SONET provider,
> they told me that "we do not see any hit". I also see lots of
> PATH errors.
>
> any suggestions?
>
> -tishi
>
> >>>>
> Sep 14 04:44:55 EDT: %SONET-4-ALARM: POS5/0/0: B3 BER
> exceeds threshold, TC alarm declared
> Sep 14 04:46:44 EDT: %SONET-4-ALARM: POS5/0/0: B3 BER below
> threshold, TC alarm cleared
> Sep 15 04:44:56 EDT: %SONET-4-ALARM: POS5/0/0: B3 BER
> exceeds threshold, TC alarm declared
> Sep 15 04:46:45 EDT: %SONET-4-ALARM: POS5/0/0: B3 BER below
> threshold, TC alarm cleared
> Sep 16 04:44:57 EDT: %SONET-4-ALARM: POS5/0/0: B3 BER
> exceeds threshold, TC alarm declared
> Sep 16 04:46:46 EDT: %SONET-4-ALARM: POS5/0/0: B3 BER below
> threshold, TC alarm cleared
> Sep 17 04:44:59 EDT: %SONET-4-ALARM: POS5/0/0: B3 BER
> exceeds threshold, TC alarm declared
> Sep 17 04:46:48 EDT: %SONET-4-ALARM: POS5/0/0: B3 BER below
> threshold, TC alarm cleared
> Sep 18 04:44:59 EDT: %SONET-4-ALARM: POS5/0/0: B3 BER
> exceeds threshold, TC alarm declared
> Sep 18 04:46:48 EDT: %SONET-4-ALARM: POS5/0/0: B3 BER below
> threshold, TC alarm cleared
> Sep 19 04:45:00 EDT: %SONET-4-ALARM: POS5/0/0: B3 BER
> exceeds threshold, TC alarm declared
> Sep 19 04:46:49 EDT: %SONET-4-ALARM: POS5/0/0: B3 BER below
> threshold, TC alarm cleared
> <<<<
>
>
--phil
This archive was generated by hypermail 2b29 : Sun Aug 04 2002 - 04:12:17 EDT