ERS 8600 TMUX Parity Question

[msbutton]

I knew to the ERS8600 arena but have been tasked to find out the definition of what the Tmux parity.

We have a situation that when one CPU is inserted into the Chasis the system runs fine. When we insert a secondary CPU, it runs for approximately 24 hours (until peak time) and then the system crashes.

What we have found is that the TMUX 6 Parity errors are extremely high when the two CPU's are inserted. When only one CPU is running it increments at a VERY slow rate, 10 a day.

But we don't know what the errors are or what the various columns mean.

Any help would be appreciated.

Here is the snapshot:

fabric 1
tmux 1 Parity 0 0 0 0
History 0 0 0 0 0 0 0 0
tmux 4 Parity 0 0 0 0
History 0 0 0 0 0 0 0 0
tmux 6 Parity 0 0 51 0
History 0 0 0 0 0 0 0 0
tmux 7 Parity 0 0 0 0
History 0 0 0 0 0 0 0 0
tmux 10 Parity 0 0 0 0
History 0 0 0 0 0 0 0 0

 

[andy88]

With dual SF cards inserted, both card's switch fabrics are used. The TapMux ASICs are the main interface between the I/O module and the backplane, they provides two connections, one to each switch fabric.

The CPU monitors I/O traffic on the backplane and if an error is detected the software will reset the SWIP (Switch Fabric ASIC) and all TapMux ASICs.

You should see an error in the log file similar to this if that happens;

WARNING Task=tChasServ FAD Mis-Align detected, SWIP Reset Status=8”

or

HwCheck: Fad CRR Failed, Reset swip

The parity errors you are seeing are a symptom of a hardware error on the backplane. From the info you have provided it is the TapMux in slot 6 that has an issue.

[highlight #EF2929]tmux 6[/highlight] Parity 0 0 51 0

I would recommend you replace the SF in slot 6