MGC-100 Sync Errors

[zuke123456]

Hi all,

I have a problem I'm hoping someone can help with. I am getting
r-sync loss i-syncvideo loss and r-symc loss errors. I read the other
thread on the problem and it turned out to be the clock source.
Mine might be different. I am getting these errors intermittently.
when it happens I get errors across all sites at the same time
no matter where they are dialed in from. I only see the errors
when in transcoding mode. when in switching mode the video
freezes. Since the problem was with all sites dialed in i started with
the phone company. They read their smart jack and said the errors
were coming from my side. I then called polycom. The tech told me
to put the bridge in a local loop. I looped one port on the net-4 card to the other. I still get the errors even in the loopback. I have tried
moving the cards to different slots with no success. I even reloaded the bridge software. I am running 4.61.2 When the bridge goes into this state their is nothing i can do about it. it just goes away on its own. I currently have 2 transcoding conferences running. they are using different audio and MUX cards and both have errors. Also
one is in the loopback and one is isdn so i am also using different ports on the net-4 card. I'm not sure what to do.

any suggestions would be appreciated

Thanks

[tonyw1962]

Sounds like your clock source is wigging out. Where are you getting your timing from?

[Border3899]

You could also try reloading software to the individual Net cards versus the MCU as a whole.

[jcicc]

Just a thought in my admittedly limited expereince on the MGC. Have you tried upgrading your software. The current software level is 6.x; It's possible that some issues may have been recognized and patched in later releases than what you are running.

[WOtech]

loopback proves internal issues. try moving the i/o and front card to another slot, retest. if that fails upgrade. if you still have issues call youre service provider to replace both cards..6.03 was pretty stable for the most part over 320.

[tdstew]

I agree with the clocking comment. How many PRI's are you using? Are you using NFAS?

[BPVTC]

If you’re timing off of that specific Net card and modifications have been made to the clock source, make sure your HARD BOOT the MCU. Has the Telco stated all is fine?

[zuke123456]

Thanks for the replies. The TELCO says everything is fine on their end.
they read their smart card and said all the errors were comming from me. I had someone come to look at it. as usual when someone comes it works fine. He typed something at the dos prompt and the
problem went away. I believe he reloaded the software to the card.
He said if i removed the card and put it back it should reload also. Well my problem came back today so I tried removing the cards
and reinstall them. It didn't work. does anyone know the dos command to reload the software?

thanks

[shel_103]

Hi,

I have the same problem with my MGC 100! I even swapped out the chassis and the problem cleared for a couple of months. Now it's happens about every month and a hard reboot fixes it. The errors occur on all transcoded ISDN sites in every conference. IP sites are not affected. We're running software version 6.03 so I don't think software is your problem. I also had the errors while in loopback mode.

I am the Telco in this situation and the MGC sits 20 feet away from our DMS switch that provides the PRIs to our bridge. I was told it was clocking too but we have reduntant clock sources and the problem occurred with clocking from both locations. Also, a clocking problem would cause issues with other equipment and customers here. I tested every PRI with our Tbird and there weren't any errors to loops on the MCU end of the PRIs. If the PRIs were taking erros that would show up in our DMS switch as well.

The only thing I didn't swap out previously was the control module -because the one they sent me was not pre-configured and wouldn't take the boot disk. Initially, moving my net 8 card to another slot corrected the problem so that's why we swapped out the chassis. Did you try moving slots? But again, the issue has re-surfaced in my bridge but is correctable now, which is different than when this happened initially. I would love to know what's causing this as well.

[zuke123456]

I did try moving the slots. I would run the loop overnight, move the slots and run the loop again. I did get less errors when I spread out the cards in the chassis. Sence the problem was intermittent and nothing else worked I thought I might have a heat problem. Usually when I get the errors I can do nothing to fix it. It just goes away on its own. I had sucess one time by removing the net4 card booting up without it then reinstalling it. That didn.t work this time. I have 2 PRI lines so I switched the clock source to a diffrent line. so far it is working but as you know the problem comes and goes. I'll let my loop run over the weekend and post back on monday.

[VuPorts]

If you have a problems with the hardware of the MGC100 or 50's, I have plenty of parts available. DAVID

[zuke123456]

All,

I was able to borrow a net-2 card from another bridge. When installed
the problem went away. I then reinstalled my net-4 card and the errors came back. I have more testing to do but it seems like the net-4 card is the problem.

[Unclefester]

On our 100 we do not use a net-2/4/8 card and are recieving the same sync error problems as discussed in previous posts. I see the errors on my serial connections. Some of them take very few errors and others too many to count. For example serial connection up for 5 hours yesterday 12,670.00 V-sync loss. We have changed out GPS timing source and the problem continues. The sync issues come and go, some days a circuit will run great others it's unusable. We have switched out circuit paths, codecs, cables to isolate the issue, and nothing.

[abramsd]

To be specific, I get this error ONLY when connecting a TANDBERG 880 E5.3 at 384K regardless if the VC is VS,TRANS or CP. I have tried using a video setting of Auto, H261 and H263.

Any thoughts useful thoughts on how to make this work on the bridge, obviously replacement of customers equipment is not an option since we only provide the Bridging.

###

Post Post

Visual Inspection:
TANDBERG view of POLYCOM from MGC = Polycom Video
POLYOCM view of TANDBERG from MGC = Blue Screen


The issue may be in the H.245 Annex D Object Identifier Assignments. Indicated by the lack of the Annex D being recognized in the CapEx.

T880:
g722_64k - 40
h261 - 297600 Bps
CIF at 30 fps
QCIF at 30 fps
Annex D

MCU:
g722_64k - 40
h261 (Transmit) - 313300 Bps
CIF at 30 fps
QCIF at 30 fps

Referencing the ITU H.245 Reccomendation PDF at:
http://www.itu.int/rec/T-REC-H.245-200605-I/en

(Ironically starting on page 245)

There are 13 "Standardized Versions Defined" (hence the word 'standard'?)

The LOGGER files on the MGC show this for the H261 (I think these are BOOLEAN values 0=False 1=True). There is a disparity in the stillImageTransmission which is maybe why I am seeing the error. As well there is an disparity in the qcifMPI values.

----- VIDEO cap: h261CapCode -----

header.direction = Receive
header.type = VIDEO
header.roleLabel = Role People
header.capCode = h261CapCode
qcifMPI = 1
cifMPI = 1
temporalSpatialTradeOffCapability = 0
stillImageTransmission = 1
maxBitRate = 3133 (313300 Bps)

----- VIDEO cap: h261CapCode -----

header.direction = Transmit
header.type = VIDEO
header.roleLabel = Role People
header.capCode = h261CapCode
qcifMPI = -1
cifMPI = 1
temporalSpatialTradeOffCapability = 0
stillImageTransmission = 0
maxBitRate = 3133 (313300 Bps)
=========================

###