Akira Hayakawa

I don't know what you guys are trying to do. LVM device is just a DM device that's managed by tool called LVM. Since stacking DM devices is valid. There...

I am afaid after you fsck the filesystem and change the state, we can't track the root cause any more but I wish you used this tool to check whether...

Please check the hardware (memory, disk), if I remember correctly, there was a case that the cause is the broken memory. It can be possible that your system is running...

Such operations in the userland will not cause data corruption in kernel level. The error message tells us that the segment has a checksum of 0 on the data but...

Please give segment ID as the parameter, not the inode number. > magic = 4194303 (unformatted) This means you haven't formatted the caching device.

You have to understand how dm-writeboost segments the caching device before using it. My guess is that, as I mentioned earlier, your system is broken. Don't know the disk, memory...

Really? I didn't tell you should zero the first sector but okay if it works. Actually, the most of the dirty blocks have been done write-back ``` current id =...

If you can. Yes. I don't precisely remember what issue number it was but I remember a user report was finally resolved after the memory was found broken.

First of all, I am doing tests on https://github.com/akiradeveloper/device-mapper-test-suite Sudden power failure is very difficult to test so I haven't. But disk corruption can be emulated by dm-flakey I had...

Well, I misunderstand your question. The writeback stuffs are irrelevant for power failure. dm-writeboost's designed to be robust for power failure issues written in the paper below: https://www.usenix.org/conference/fast13/understanding-robustness-ssds-under-power-fault Typically, partial...