0

I posted previously about a corrupted SD card that was becoming read only, but could still be accessed.

After setting up a new card - starting with a fresh Raspbian image - I've hit the same problem even though I am now using a powered hub and powered drives: as soon as I plug in more than 2 drives, the corruption (or at least the read only state) comes back again.

This seems to be connected, but does anyone have any insight or experience of this or similar please?

0

I see 2 methods to solve this issue :

1 ) Find the root origin of the corruption

The only case of SD corruption I met was linked to unwanted reboots, always power-related. Did your system shutdown when you plug the last HDD ? Are you sure that your USB Hub has sufficient electrical specifications to handle both HDD ?

2) Use a read-only filesystem:

If the cause of corruption can't be identified, or can't be solved, you may want to use a read-only file-system.

The aim is to use your / and /boot partitions in read-only mode. Putting your main system in frozen-state protect it from corruption. The whole SD card don't have to be read-only, you still can have a read-write partition made for storing datas (which still can be corrupted !).

When you need to edit your system file (configuration of some service, i.e.) remount the target partition in read-write mode, do your work, and put the partition in read-only mode once done.

If the protection of the stored data is important too, you may want to use a tmpfs partition (RAM stored memory), and sync it from time to time to a read-only partition made for storing datas (putting it in read-write mode for a very small amount of time, reducing the risk of corruption)

Switching your system to read-only is a bit tedious, as you will see in the tutorial, you should make a backup of your SD-card once done.

https://hallard.me/raspberry-pi-read-only/

  • Sorry for slow reply, didn't get a chance to look at this until now. To answer your q's 1) I thought it was power related, which is why I got the new hub. There was no shutdown at any point, but suddenly it was ro. I am sure the hub has enough power to handle the HDDs. I am trying a totally new smaller SD today to rule-out that being the cause. (Frankly, I have abused SDs on other Pi's by pulling the power and never suffered corruption. Perhaps I've just been lucky, but it might be the Pi itself. I'll have to try a different one if I can't narrow things down.). I'll give 2) a go. Thanks. – Lorccan Jun 26 '17 at 11:27
  • About 1), did you take a look at dmesg, to get hints on what is happening? In any case, let's hope 2) will solve your problem ! – Technico.top Jun 26 '17 at 11:53
  • Stupid q, but where do I find dmesg? I don't quite get from scanning 2) which physical device the rw requiring elements are using. – Lorccan Jun 26 '17 at 12:21
  • dmesg is a common linux command, just type it in a console and you will see the main log of your system. I didn't undestand the second part (not native english speaker) sorry :) – Technico.top Jun 26 '17 at 12:38
  • 1
    Thanks. All clear now. I'll report back if I get anywhere with diagnosis. – Lorccan Jun 26 '17 at 13:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.