So ... what exactly have you got and what exactly are you trying to do? You seem to be saying you're doing md raid5 on 3 drives, direct on the drives themselves, and are then partitioning that md device (which is a bit odd, but, whatever), however you also show data which seems to suggest you have the drives themselves partitioned - you can't really do both, as those may quite be stepping on each other's data, and likely won't work and/or may corrupt data. Also, if you take your md device, say it's /dev/md0 (or md0 for short), and you partition it, the partitions would be md0p1 md0p2 etc., those would be pretty non-standard and atypical names, is that what you actually did? Or what did you do? If you did partitioning, e.g. MBR or GPT direct on the drive, after creating md raid5 direct on the drives, you likely clobbered at least some of your md device data.

So, which exactly is it and what are you trying to do?

Also, if you partition md device, you likely have to rescan the md device after it's started to be able to see/use the partitions, e.g. partx -a /dev/md0

But if you've got partitions on the drive, and are doing it that way, then you'd do your md devices on the drives' partitions - that would be more typical way - though can do driect on drives, but partitioning md device would be quite atypical. Typically one would put filesystem or swap or LVM PV or LUKS on md device, or use btrfs or zfs directly on it, but generally wouldn't do a partition table on it.

So, how exactly do you have your storage stack on those drives, from drive itself on up to filesystem or whatever you're doing for data on it? What are all the layers and what's the order you have them stacked?

# mdadm --examine /dev/sdc /dev/sdb /dev/sdd
/dev/sdc:
MBR Magic : aa55
Partition[0] : 3907029167 sectors at 1 (type ee)
/dev/sdb:
MBR Magic : aa55
Partition[0] : 3907029167 sectors at 1 (type ee)
/dev/sdd:
MBR Magic : aa55
Partition[0] : 3907029167 sectors at 1 (type ee)

That shows each drive MBR partitioned, with single partition of type
ee GPT, so, you have GPT partitioned drives, not md direct on drives.

So, if you put md on partitions, should look for it there:
# mdadm -E /dev/sd[bcd][1-9]*

I created an ext4 partition on md0 before moving the data
created the array first and then partitioned md0

So, which is it? What devices did you create md0 on, and what
device did you create the ext4 filesystem on?

# fdisk -l
Disk /dev/sdc   
Disklabel type: gpt
Disk /dev/sdd
Disklabel type: gpt
Disk /dev/sdb
Disklabel type: gpt

So, you've got an empty GPT partition table on each.

Yeah, you can't have md device direct on drives and also have partition
table direct on on same device (e.g. /dev/sdb). You get one, or the other,
not both on same device.

Maybe i should recreate the array ?

# mdadm --create /dev/md0 --level=5 --raid-devices=3 /dev/sdb /dev/sdc /dev/sdd --uuid=a10098f5:18c26b31:81853c01:f83520ff --assume-clean

Not like that, that may well corrupt your data on the target - but you
may have already messed that up anyway.

recreated the array and it mounts and all files are there

Might appear to, but no guarantees you haven't corrupted your data - and
that may not be readily apparent. Without knowing exactly what steps
were used to to create the filesystem, and layers beneath it, and other
things you may have done with those drives, no easy way to know whether
or not you've corrupted your data.

Also, what have you got in your mdadm.conf(5) file? That may provide information on how you created the md device, and on what ... but if you've been recreating it, that may have clobbered the earlier info. What's the mtime on the file, and does it correlate to when you first made the array, or when you subsequently recreated it?

webmin, huh? Well, also check logs around time you first created the md device, it may possibly show exactly how it was created and on what devices.