Presently, the timeout for detecting my USB drive on restart is not long enough. It times out, looks for the sd card (which is not installed), then goes into a loop.
Powering off the Pi at this point and turning it back on gets me a complete boot, presumably because the drive is fully powered up at that point (?).
What’s the easiest way to increase the timeout value?
External USB drives can be slow to power on. The best solution for that, is to use an external drive that with it’s own power source, so it’s “always” turned on.
I’ve read in other forums that there are some boards with USB not working correctly (probably driver/kernel related) and require a command before reboot to stop the USB controller so it works after the reboot. Otherwise you need to power down and then power up for it to work. Can’t remember the command right now.
@johntdavis84 I have no clue why you choose not test the pieeprom-2021-01-11 with rpi-eeprom-git from the beta folder when you had the chance in another thread to see if it solved your issue. Since then it has been promoted yesterday to stable and I have pushed rpi-eeprom-git-20210114-1 to the unstable branch when the mirrors sync.
I am asking again along with @Rip2 to try the pieeprom-2021-01-11 image and see if it fixes your issue.
@Darksky , sorry for not replying clearly over on the other thread. I got distracted by the NVME-drive interface board I bought being a complete piece of garbage (including, apparently, having a dead 5V power input).
My, er, Pi Time was devoted to trying to straighten that out–including having to buy a new enclosure for the NVME drive and figure out how to return the thing I had already, and then I was sick enough of the whole thing not to wanna mess with it a while.
More than that, I’m confused by the output of rpi-eeprom-update I’m currently getting.
My reading of this is that it’s telling me I’m up-to-date on the stable branch, even though I’ve got a version of the EEPROM and boot loader from September 3.
Looks like the Jan 11image that got promoted to stable in git was not stable after all. This is the reason I tried to say in another thread why I did not like to have a git package unless it had a fix with some issue someone was having. They introduce or fix things and cause issues some where else they have to fix. I also noticed with the latest Jan 16 they put in stable it is tagged with BETA.
I just do not like putting out a package that could possibly break some one’s install.
Anyway the latest git is pushed to the unstable branch when the mirrors sync.
rpi-eeprom-git 20210116-1
[ray@pi4 ~]$ sudo rpi-eeprom-update -r
Removing temporary files from previous EEPROM update
[ray@pi4 ~]$
[ray@pi4 ~]$ sudo rpi-eeprom-update
BCM2711 detected
VL805 firmware in bootloader EEPROM
BOOTLOADER: up-to-date
CURRENT: Sat Jan 16 02:10:13 PM UTC 2021 (1610806213)
LATEST: Sat Jan 16 02:10:13 PM UTC 2021 (1610806213)
FW DIR: /lib/firmware/raspberrypi/bootloader/stable
VL805: up-to-date
CURRENT: 000138a1
LATEST: 000138a1
[ray@pi4 ~]$
Sorry I dropped off the edge of the world. I got distracted by just how unstable my drive enclosure turned out to be, and had to find something to replace it with before the power-drop induced drive crashes ate my data or ruined the hardware.
I ended up replacing the enclosure, and now I’m not having any trouble with the USB drive getting power from the Pi on boot, even without the beta EEPROM.