Not really sure how to fix this lint complaint:

Warning: function-result-limit: maximum number of return results per function exceeded; max 3 but got 4 (revive)
Warning: function-result-limit: maximum number of return results per function exceeded; max 3 but got 5 (revive)

There are other input plugins that also violate this limit (e.g. kernel_vmstat has a function with 4 return statements) so I'm struggling to see how this check is valid.

Sorry about all the commits, but make test doesn't seem to work correctly for me locally, so I'm leaning on circle to see my tests pass.

Thanks for the new plugin! As a user of Linux mdraid arrays, I'm keen to monitor them!

I ran it on a system here with four disks, one which is failed, and I don't see it detecting the failed disk? Perhaps the disk just dropped but hasn't failed?

Here's the /proc/mdstat output

Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10] 
md0 : active raid5 sdd1[3] sdb1[1] sda1[0]
      5860144128 blocks super 1.2 level 5, 64k chunk, algorithm 2 [4/3] [UUU_]
      bitmap: 8/15 pages [32KB], 65536KB chunk

unused devices: <none>

Here's the output:

2021-05-25T21:45:53Z I! Starting Telegraf 
2021-05-25T21:45:53Z I! Loaded inputs: mdstat
2021-05-25T21:45:53Z I! Loaded aggregators: 
2021-05-25T21:45:53Z I! Loaded processors: 
2021-05-25T21:45:53Z I! Loaded outputs: file
2021-05-25T21:45:53Z I! Tags enabled: host=shirka
2021-05-25T21:45:53Z I! [agent] Config: Interval:10s, Quiet:false, Hostname:"shirka", Flush Interval:10s
mdstat,ActivityState=active,Devices=sda1\,sdb1\,sdd1,Name=md0,host=shirka DisksActive=3i,DisksTotal=4i,BlocksSynced=5860144128i,BlocksSyncedSpeed=0,BlocksSyncedFinishTime=0,DisksFailed=0i,DisksSpare=0i,BlocksTotal=5860144128i,BlocksSyncedPct=0 1621979160000000000

Thanks for the new plugin! As a user of Linux mdraid arrays, I'm keen to monitor them!

I ran it on a system here with four disks, one which is failed, and I don't see it detecting the failed disk? Perhaps the disk just dropped but hasn't failed?

Here's the /proc/mdstat output

Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10] 
md0 : active raid5 sdd1[3] sdb1[1] sda1[0]
      5860144128 blocks super 1.2 level 5, 64k chunk, algorithm 2 [4/3] [UUU_]
      bitmap: 8/15 pages [32KB], 65536KB chunk

unused devices: <none>

Here's the output:

2021-05-25T21:45:53Z I! Starting Telegraf 
2021-05-25T21:45:53Z I! Loaded inputs: mdstat
2021-05-25T21:45:53Z I! Loaded aggregators: 
2021-05-25T21:45:53Z I! Loaded processors: 
2021-05-25T21:45:53Z I! Loaded outputs: file
2021-05-25T21:45:53Z I! Tags enabled: host=shirka
2021-05-25T21:45:53Z I! [agent] Config: Interval:10s, Quiet:false, Hostname:"shirka", Flush Interval:10s
mdstat,ActivityState=active,Devices=sda1\,sdb1\,sdd1,Name=md0,host=shirka DisksActive=3i,DisksTotal=4i,BlocksSynced=5860144128i,BlocksSyncedSpeed=0,BlocksSyncedFinishTime=0,DisksFailed=0i,DisksSpare=0i,BlocksTotal=5860144128i,BlocksSyncedPct=0 1621979160000000000

This should be resolved with the most recent commits. A disk marked as _ is actually (according to mdstat's docs) different than an actually failed disk, so we'll be counting those separately.

The same collector was approved and merged into Prometheus' node_exporter: prometheus/procfs#380

@ssoroka I believe I've addressed all your concerns. Any chance you could take another look at this?