slots are failing in 24000 after the frimware upgrade 5.3.x

[Anonymous]

EMC CE was onsite to upgrade the switch from 5.3.0-->5.3.2. After the fw upgrade, CP in slot 5 reported as fault and the CE replaced it and then loaded the 5.3.2 code on it. Then blade is slot 3 reported as faulted and the CE reseated it to see if it would clear the error and then replaced it. Then blade in port 10 fault and the reseated and replaced it. Now the new CP in slot 5 is reporting as faulty. CE then called into the lab and I saw that from the supportsave hadump reported both CPs as ok, just hadisabled. Slotshow reported is as faulty with rsn: 21. From this I saw that the CP failed some internal port initialization (just info messages though). I had the CE do an haenable, CPs ended up being sync'ed up but slotshow reported it as faulty. I had the CE poweron/poweroff the blade to see where it was failing. But still is showing faulty

[Anonymous]

I was going through the release notes and one bug that has been fixed in 5.3.1 seems to be very close to what you describe.

Although you are on 5.3.2 and should not be affected by this, maybe its a good idea to compare your problem with the bug description.

DEFECT000208020 High Summary: On 2G platforms, hafailover then portdisable/portenable of

E-ports causes port blade fault with tx parity error

Symptom: Under heavy IO going across ISLs, perform an hafailover and

then disable each E-port and then enable each E-port. Then a blade faults

with error 21. This only impacts 2G platforms; 2G switch may be faulted

and/or rebooted under the same circumstance.

Feature: FOS Software

Function: ASIC Driver

Probability: Low

Risk of Fix: Low

Service Request # : 130111

Reported in Release: FOS5.3.0

[Anonymous]

anyhow you have to replace the blade.It happens often while doing FOS upgrade. If it is due to Bug then you have to change FOS.