Backup SD Card by copying out files?

[TMNTturtlguy]

@pjft Thanks for all the help as usual!

Yes, confirmed I am running off of the USB on the original card. Here is the info for the original card running off of the USB:
Working card with working USB - all commands shown

rpi-clone card non working usb - all commands shown

[pjft]

@TMNTturtlguy Is your USB drive a Kingston drive?

In the "bad" card, you mention the USB drive blinks, then stops. Can you unplug it, run those commands (dmesg | tail, lsusband df), and then replug it and run the commands again?

Also, does cat /var/log/syslogoutput anything for you? If so, what does it state for both cards?

I'm somewhat stumped, but we'll make it work.

Alternatively, if this thread doesn't help (just noticed I forgot to paste the link earlier), I'd love to read up on both boot logs (dmesg) - for the good and bad cards.

But a few things first to try. As I said, I have this working though I am not using the automated mount from RetroPie, but rather the old method via fstab. I am confused as to why you would have such differences between cards, though.

[TMNTturtlguy]

@pjft The usb drive is a computer stores brand - they offer lifetime warranties on them, so if they go bad, they just give you a new one. So i can't be positive, but I assume it is the one that shows up.

No USB inserted

USB inserted

Cat /var/log/syslog for non working Card

cat /var/log/syslog for working card

[pjft]

@TMNTturtlguy Thanks.

A few questions, if possible - and thanks for putting up with all of this.

So, the USB drive seems to be detected - as you can tell by the Kingston device showing up in both versions. However, it's not being mounted for some reason.

1. What does ls /dev/sd*show in your bad card?
2. What does cat /etc/usbmount/mount.dshow in both?

I'm also wondering if that's your entire syslog for the working card? Seems awfully short.

Also, on the one that doesn't work, there seems to be an awfully long (4-min?) gap between a device being apparently disconnected, and a new one connecting? See device disconnect 4, and new number 5. Does that happen at all boots?

I'm happy to help you set it up as mounted via the old process, though I'm more keen on exploring why the difference in behavior here.

[Sano]

I think a "sudo fdisk -l /dev/sda" could provide some useful information too.

[pjft]

@Sano If you have the expertise here, would love your inputs as I'm certainly out of my league here :)

By the way, love your avatar!

[TMNTturtlguy]

@pjft The syslog for both cards is very very long, i just cut out the parts i saw about the usb drive. is there a way i can output the entire log? I went into the folder, but the log is not a typical file type that i could open and read.

1. ls /dev/sd* shows: /dev/sda /dev/sda1
2. cat /etc/usbmount/mount.d in the bad card = cat: /etc/usbmount/mount.d: Is a directory

Sano: sudo fdisk -l /dev/sda on bad card

i will do the same for good card and post back momentarily
Thanks!

[pjft]

@TMNTturtlguy So, the drive is there, and it's in the right mount point (sda1) - for whatever reason, it fails to be mounted.

Can you do ls /etc/usbmount/mount.dand then cat /etc/usbmount/mount.d/<filenames>one by one, for both cards?

[TMNTturtlguy]

@pjft not sure i am doing the cat /etc/ correct, i am guessing i am supposed to be putting something else in instead of <filenames>? but here is what i get

good card

Bad card

[Sano]

@pjft Thanks :) Had it for many years now.

@TMNTturtlguy could you please do a ls -l /media and post the result ?

[pjft]

@TMNTturtlguy Definitely - replace <filenames> with 10_retropie_mount

:)

And follow @Sano 's advice, as he definitely seems more knowledgeable about this :)

[TMNTturtlguy]

@Sano ls -l/media on bad card gives me this: total 0

Do you want that for the good card?

[Sano]

Nope, just do a "sudo mkdir /media/usb0" and retry :)

Here is what /media should be like :
pi@retropie:~ $ ls -l /media
total 68
lrwxrwxrwx 1 root root 4 nov. 5 22:35 usb -> usb0
drwxr-xr-x 2 root root 3488 nov. 5 22:35 usb0
drwxr-xr-x 2 root root 3488 nov. 5 22:35 usb1
drwxr-xr-x 2 root root 3488 nov. 5 22:35 usb2
drwxr-xr-x 2 root root 3488 nov. 5 22:35 usb3
drwxr-xr-x 2 root root 3488 nov. 5 22:35 usb4
drwxr-xr-x 2 root root 3488 nov. 5 22:35 usb5
drwxr-xr-x 2 root root 3488 nov. 5 22:35 usb6
drwxr-xr-x 2 root root 3488 nov. 5 22:35 usb7

[TMNTturtlguy]

@pjft bad card

[TMNTturtlguy]

@Sano

[Sano]

Ok, try to plug your USB thumbstick now. This should work.
It seems that the content of the folder /media was not copied.
It is used by usbmount as mountpoints.
Fortunately, you could just recreate the folders with mkdir, and eventually the symlink witn ln -s.
Alternatively you can reinstall the usbmount package.

[TMNTturtlguy]

@Sano and @pjft it worked!

Sorry for the delay, i was on a work call. This absolutely worked. I need to make sure that everything is there, but when i restarted, the USB was recognized and the roms appeared!

So I am 100% sure that my process to create the copy using rpi-clone was correct, and i performed the clone twice with identical results. Every other aspect of the clone looks to be 100% correct at this point. I am going to add the mkdir command to my personal instructions. When i do this for my other build I will see if i have to follow the same steps. If so, i might attempt a wiki on how to use the rpi-clone tool with the automated usb service.

Thanks to both of you for all the help. i can't believe it was something so simple after all that work.

[Sano]

Nice.
Glad I could help a bit. :)
And don't blame you, a lot of resolutions tend to be simple when the problem is identified.

[pjft]

@Sano Thanks for stepping in here!

Any idea why the usb0 folder would not have been present at the time of cloning the card? Would it be related to unmounting that particular drive before cloning, would that remove the folder?

[Sano]

@pjft My pleasure.

Regarding the usbx folders, they're created during the usbmount package intall and should be there.
I suppose there is an exclusion on /media in the clone script in order to avoid copying external disks content, but with a depth at 0, which would cause this issue...
This is only deduction, the problem could be elsewhere.