Can attaching an LCD display and (custom) i2c hardware disrupt the boot mount of mass storage?

We have a problem in that our device, based on BBB, has custom hardware connected to i2c. We use a 4D systems LCD display on the cape (with the power pin removed, don’t ask!) for UI. The software writes result data to /media/BEAGLEBONE which is mounted as mass storage by some wizardry in the boot process. However, this only seems to work when I don’t attach the LCD. I have worked happily with my HDMI out to the monitor, and files are written to the mounted folder and visible upon reboot. However, on a device with a screen attached, the entry in the mount table has vanished. HOWEVER this might be also dependent on having our i2c hardware connected. Does anyone have any wisdom to help shed light on this?

If anyone can explain exactly where and how the mass storage is mounted at boot time (it isn’t in /etc/fstab) that would be helpful?

It's exported by an init script that calls:


The current git master version of that file looks like:


Hi Robert

You’re just the guy I’d like to talk to. Do you have any thoughts on why the mounting of /dev/mmcblk0p1 to serve as a virtual mass storage partition might have been disrupted? Does adding a cape - specifically for an LCD screen - change the set of files that are generated or run during boot? For sure. on the board without LCD if I type “mount” I get

/dev/mmcblk0p1 on /media/BEAGLEBONE type vfat (rw,nosuid,nodev,relatime,uid=1000,gid=1000,fmask=0022,dmask=0077,codepage=437,iocharset=iso8859-1,shortname=mixed,showexec,utf

whereas on a board with the LCD I do not see this. Or rather, sometimes I do see it even with an LCD screen but we are not sure what conditions might make it vanish? Could we simply mount it manually by editing fstab?