Before forming this question, as a disclaimer, I am no expert user, but I am also no novice. Some of the advice I am given may be easily understood, and some others (especially those who use excessive jargon or acronyms) I may need further explanation.
backstory: I have a SilverStone SST-DS380B SFF Mini-Tower Case, and there are 6 hard drives in them:
/dev/sde
Device Model: WDC WD30EFRX-68N32N0
Serial Number: WD-WCC7K6DLV8PD
/dev/sdc
Device Model: WDC WD30EFRX-68N32N0
Serial Number: WD-WCC7K5LKUY53
/dev/sda
Device Model: WDC WD30EFRX-68N32N0
Serial Number: WD-WCC7K0NK7H07
/dev/sdf
Device Model: WDC WD30EFRX-68N32N0
Serial Number: WD-WCC7K0STUVPS
/dev/sdd
Device Model: WDC WD60EFRX-68MYMN1
Serial Number: WD-WXK1H642JZNP
/dev/sdb
Device Model: WDC WD30EFRX-68N32N0
Serial Number: WD-WCC7K3TKHLSV
Over time, I plan on replacing each drive with a 10TB version so the array grows to a nice 60TB.
I was following the instructions from this page on how to change a disk in a RAID 5 configuration. I use webmin to monitor the server, and so I noticed that the HDD total space in the array was up to 89% of capacity. So I hopped on to Newegg.com and purchased a 10TB HDD. My plan is to replace each existing 3TB drive with a 10TB over time. However, it is important to note there is a 6TB WD in the array which will also be replaced by a 10TB drive, over time.
I got up to the part, "Determine which drive the 250GB is and note the /dev/sd that it is.
Next, you are going to fail the drive from the array and remove it. I am just going to use my /dev/sdf drive as an example:" and tried to fail the drive (sda1), but the system did not recognize sda1. upon failing it I tried sda without the -1, and it successfully failed. I then removed it with the remove command, took it out of the server tower, put the new 10 TB in its place (but not before I initialized it using minitool partition app on a different computer with an external drive bay)
Needless to say the next steps from that site did not work. Alternating between mdadm and webmin raid management was probably a bad idea, but one thing led to another, and upon rebooting I found my RAID5 became a RAID0.
I tried a couple of ways to get the 10TB to play nicely with the rest of the array, but adding it and growing the array proved to be too difficult for me. I was able to finally find a way to add the 10TB, and grow it, but I think that's where everything went awry. the command mdadm --grow --raid-devices=3 /dev/md1 or mdadm --grow --raid-devices=3 /dev/md127 did not work, because it asked me to correct my syntax to include the size of the array, which I manually typed it to 5860533168. I let it do its thing, but then doubted my actions when I returned to webmin and saw the usable space I had created was 5 TB! so I manually stopped the grow command by using sudo mdadm --stop /dev/md127.
Doing so caused the Raid to change to Raid0. I figured to backtrack and undo this mess, I could put the original sda1 (the first 3TB drive I intended to replace with the 10TB) back in to the server, remove the 10TB, and somehow rebuild the array back to the RAID5.
Can someone help me return this RAID to a RAID5? Can it be done? my server is a headless server and so all my actions are done through ssh and webmin.
If I can clarify anything else I may have missed in explaining, please guide me. I imagine the first order of business would be to see log files and a history of my actions from the shell.
Help!
Now as I said, I am no expert nor newbie at this, a RAID5 can be returned to its form from a RAID0 if it was originally a RAID5.
