Hello,
I'm having an interesting issue where I have two different MOD5234 boards running the same program, but one board is having periodic delays in TCP transmission. I am swapping the two boards into the exact same hardware and using the same cables and power source.
The program is just a simple loop sending a short message to a TCP server then waiting for a response, repeating forever. The problem I'm seeing is that about 4 times a minute, one of the boards will have a ~0.5s delay in between transmitting the ACK to the server response and sending its next message (I'm using Wireshark to snoop ethernet activity). With the other board, I only observed this happen twice in 10 minutes. I've repeated the tests several times with similar results each time. I've also seen the problematic board miss incoming packets from the server entirely, though that's not as repeatable and could be software, as I witnessed it while running a different program.
The boards do look slightly different, and I'm wondering if there perhaps are any known issues with ethernet on some MOD5234 batches? The boards are both Rev 1.7, but the problematic board has the text "88 94V0" underneath the revision, whereas the better board says "88E 94V0".
Alternately, I suppose I could have damaged the board somewhere along the way, but it seems odd to me that any damage would have the effect of occasionally delaying ethernet but not disabling it. I'm software though, so my knowledge of that stuff is pretty limited. Is that possible?
Any insight that might be offered would be greatly appreciated. Thanks!
-David