ESP32-S3 dev board wroom-1 - pins 8 and 9 shorted

[lafar6502]

Hi, I'm trying to put a simple project together for ESP32-S3 wroom-1 board, but accidentally discovered that pins 8 and 9 are shorted.
Pin 8 i'm using for I2C, and pin 9 - for an encoder button.
When i was trying to handle the input from the encoder button, i discovered the board is resetting whenever the button is pressed. Then, after some measurements i discovered there's a short circuit / zero resistance connection between pins 8 and 9 on the board (the ESP32 module was disconnected entirely from the rest of the circuit).
I suspect this might have something to do with the device reset - as the error log mentioned something about I2C transaction broken (so i assume the encoder button on pin 9 was messing up I2C on pin 8).
Is this possible and can be fixed with some software configuration? Or is my ESP32 broken? Unfortunately i only have 1 microcontroller and can't compare with anything else.

Thanks
R

[lafar6502]

Looks like some electric damage to pin 9, button works OK on another pin and doesn't cause the board to reset.
Pin 8 works fine with I2C, as long as pin 9 is not used.
Must be electrostatic discharge because i haven't connected anything with voltage to it.

[lafar6502]

Today i continued work on the device and just now discovered another pair of GPIOs shorted - pin 1 and pin 2.
I don't know if they were OK originally because i never checked resistance between them - just found out when i was checking for shorts between contacts on the board i'm making.
So, is it something that happened only to me, or are the GPIOs in ESP-32 particularly easy to damage during soldering or assembly?

EDIT: one more pair found - pins 5 and 6 are shorted too. To be clear, i'm testing this on ESP32-Wrover board disconnected and taken out from the socket, so it's not an interference from the rest of my circuit. Im gradually enabling more functions of my board and this is when i discover problems, i did not test every pair of pins for shorts initially.
It just looks like some random damage to different pairs of pins. Two of the shorted GPIO pairs are adjacent pins on the board, but pins 8 and 9 are not next to one another. I soldered the header pins myself, but the pins are not shorted by by sloppy solder - the problem is somewhere on the board.

[Sprite]

What particular devboard do you have - an official Espressif one or a third-party one? Generally, the failure mode of silicon tends to be either A. pin is disconnected from the silicon (blown bond wire), B. pin is shorted to Vcc or GND, or C. chip is entirely dead. As such, I'd suspect the devboard to have bad soldering or something.

[lafar6502]

It's a board from Aliexpress marked WROOM-1 N16R8, so I suppose it's not Espressif official, but it looks very similar to DevKitC-1 (not exactly the same, for example button placement is different).
In my case it's D - board is functional except of the pins I listed, the pins are shorted in pairs but not to GND or VCC, just to the other pin on the board - and so, the pair can't be used together. If I use just one from each faulty pair, for example just the pin 8 and leave no9 disconnected - all fine, I2C will work on pin 8. So it might be some manufacturing problem after all. For now it's ok, i have still some GPIOs left and can wire around the problem. And have ordered another board to check if i'm lucky this time.
BTW GPIO MUX shines