{upstream} inconsistency with the --no-degraded option

[ceg]

Binary package hint: mdadm

There is inconsistency in the --no-degraded and --incremental option.

They will start degraded arrays, if they have been --run degraded before.

Its clearer to report problems and and do interventions when staring degraded raids stays the resposibility of startup scripts.
(https://wiki.ubuntu.com/BootDegradedRaid)

[ceg (ceg)]

> clearer when staring degraded raids stays the resposibility of startup scripts.

Or the user manually starting an array. (possibly by right clicking)

Then, when two array disks are pluged in and it doesn't come up it is always clear they the array is not complete.

[ceg (ceg)]

Example: 3 member mirroring array

A)
One disk fails during uptime, raid keeps running with 2 disks and on a reboot the array will get started through the --no-degraded or --incremental command with 2 disks.

B)
One disk fails during downtime, the array won't be started with --no-degraded or --incremental. A specific --run is necessary to start it degraded.

[ceg (ceg)]

Current state of ubuntu systems with md raid: https://wiki.ubuntu.com/ReliableRaid

[Thomas Hotz (thotz-deactivatedaccount)]

Marking to confirmed, hopefully a developer can fix this soon. Thank you very much for reporting this bug!