On 6/27/2022 8:57 PM, Samuel Sieb wrote:
On 6/27/22 17:37, Bill Cunningham wrote:
On 6/27/2022 8:31 PM, Samuel Sieb wrote:
Did you mean to send this off-list?
On 6/27/22 17:01, Bill Cunningham wrote:
<off point> ext2, yes version 2 I use, was mounted RO at a later point, SO I used mount -o remount,rw /dev/sda3 /. To remount to be able to change the fstab and save to a rw filesystem. <off point>
Why are you using ext2? That's probably the biggest cause of your problems. Otherwise, it would have most likely repaired itself.
It will run fsck, but missed a few things anyway. IDK that journals are needed so much anymore with memory management being what it is. If anything, btrfs I would think would be better. Maybe lvm. I'm only using 30 GB anyway. I usually don't have this problem.
The journal has nothing to do with memory management. It's about filesystem protection. btrfs would also be better because it is always "journaled" (not exactly, but fairly equivalent).
I thought it was involved in "swapping". Like a swap file my bad.
Samuel Sieb:
The journal has nothing to do with memory management. It's about filesystem protection. btrfs would also be better because it is always "journaled" (not exactly, but fairly equivalent).
Bill Cunningham:
I thought it was involved in "swapping". Like a swap file my bad.
File system journalling is about all writing to a drive being done as a sequence (this write, then that update, then the next update). During a crash, hopefully you only lost the last bit of writing, and what was done before that is still there.
On 6/29/2022 1:54 AM, Tim via users wrote:
Samuel Sieb:
The journal has nothing to do with memory management. It's about filesystem protection. btrfs would also be better because it is always "journaled" (not exactly, but fairly equivalent).
Bill Cunningham:
I thought it was involved in "swapping". Like a swap file my bad.
File system journalling is about all writing to a drive being done as a sequence (this write, then that update, then the next update). During a crash, hopefully you only lost the last bit of writing, and what was done before that is still there.
Oh I see. lol Well I added a journal file. In the root ".journal". Then dumpe2fs says clean now. I have about 30 GB of space with this filesystem, and I really truthfully don't get errors much; unless they are small ones. This time there was quite a few. I mounted ro and changed the fstab file from ext2 to ext3. Seems to be working. I never really looked into the ext3 or ext4 because ext2 has always worked. Huge_file support I don't need with a 30 GB space size. I used the tune2fs -j option too, so there may be some ext3 attribs that aren't around like dir_ option. But it seems to work. Then I remounted in rw and saved the fstab and all seems right. Thanks.