I have done the following steps 1. Created a raid 1 with two PCIe SSD's say /dev/ssd1 /dev/ssd2 in mdadm with 1.2 2. Created ext4 3. Mounted and created a file 4. Poweroff the system 5. Disconnect one PCIe SSD 6. Boot the system and mdadm --detail shows the status of array as FAILED 7. Stopped the array 8. Did mdadm --assemble --force /dev/md127 /dev/ssd1 9. Mounted the array and the file I created was intact 10. Issued mdadm --detail and status is reported as degraded For the second case created raid 1 with ddf Till step 7 everything was same as in with superblock 1.2 But when issued mdadm --assemble --force /dev/md126 /dev/ssd1 it shows that /dev/ssd1 is busy skipping. Checked in dmesg nothing is captured there Is there any mistake in the procedure ? Regards, Arka On Mon, Nov 28, 2016 at 2:50 PM, Arka Sharma <arka.sw1988@xxxxxxxxx> wrote: > Hello, > > I want test a redundancy scenario with RAID 1. I have crated an array > with mdadm and formatted ext4 and after mounting I have written a text > file. Now I want to switch off one device, and with another remaining > device I expect to view the text file. When I issue mdadm --detail it > shows the State as active,FAILED,Not Started. But in dmesg I can't > find any error message from md related to one of the physical disk > missing. I was searching about this and I found some references of > creating degraded RAID 1 with missing parameter in mdadm, but what I > want to simulate is that let's say one of the disk has gone bad and > not yet being detected under /dev and in this case we want to verify > that if data is intact. > > Regards, > Arka -- To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
