Yes, one of the board is weaker and generates an average of 30 HW per day but until it does not cause of dropping board, this situation is acceptable, because it does not make much impact on hashing speed.
My problem miner started out this way, but it has progressively gotten worse. Right now I have disconnected the board that throws the errors and letting it run to see how long the miner goes before issues appear again. If the board with the HW errors is the cause of my boards dropping then this should prove that out. I tried the one board at a time testing and sadly I couldn't reproduce it. So now I am more or less going at this "backwards" and removing the suspects as they poke their heads out of the holes. I can tell you this isn't a lot of fun.
How many L3s do you have Longsnowsm, or put it another way, what % of cards have failed or showing signs of failing? Just the 1 so far?
On the L3's so far it is one miner that has crashed and lost boards repeatedly. It appears that it is likely the one board that throws the hardware errors that is causing this issue. I have been running it since Saturday with the one board disconnected and it is working fine so far. If by the end of the week it has remained stable then I will be shipping this one board back to Bitmain.
I have another L3 miner that reports some hardware errors on one board, but so far it has not escalated into a problem.
I have heard of others now that seem to be having some issues now as well. Some think it is temps and are setting fan speeds to fixed speeds at like 80%. It is board 3 that seems to be getting warmer than the others so people are trying to work around it. So far I haven't messed with the fan speeds, but I did notice my problem miner becomes a bigger problem as the ambient temps get warmer. However in my case the problem exists regardless of temps, but it becomes more persistent when it gets warmer.