"We probably still have users who will expect to be able to configure a
system to halt & prompt prior to booting a system with a degraded
array."
I've dropped BOOT_DEGRADED, precise because we don't have such users.
Instead we have multiple cases where boot-degraded check produced false
positives (e.g. both drives _are_ present, yet the check claimed they
are degraded) and i have been through a lot of bug reports claiming that
raid was setup precisely to boot unattended / degraded. Halting the boot
is not useful, as remote-hands in the data center are often needed to
resolve the situation. I could be persuaded to e.g. boot in read-only
mode, but even such semantics are causing our server users grief (e.g.
see mountall bugs where machines are booted RO if last mount time is in
the future).
** Changed in: mdadm (Ubuntu Trusty)
Status: New => Incomplete
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1302195
Title:
System with degraded array boots regardless of BOOT_DEGRADED flag
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/mdadm/+bug/1302195/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
