ZFS pool - HEALTH fault

[Herbert J. Skuhra]

On Wed, 21 Nov 2018 22:28:57 +0100, "James B. Byrne via freebsd-questions" wrote:
> 
> On Wed, November 21, 2018 12:45, Kevin P. Neal wrote:
> > On Wed, Nov 21, 2018 at 09:33:32AM -0500, James B. Byrne via
> > freebsd-questions wrote:
> >> Immediately following an upgrade from FreeBSD-11.1 to 11.2p4 we
> >> received the following warning.  I can see the suggested remedy
> >> but I need to know why this has happened first.  We have
> >> previously upgraded numerous hosts from FreeBSd-11.1 to 11.2,
> >> all of which are using zfs, and no other system has reported
> >> such a fault,
> >>
> >>  ZFS pool - HEALTH fault
> >
> > This message that says "HEALTH fault" comes from what? What command
> > prints this message?
> 
> It comes from a regularly scheduled health check script whose relevant
> portion is:
> 
> condition=$(/sbin/zpool status | egrep -i
> '(DEGRADED|FAULTED|OFFLINE|UNAVAIL|REMOVED|FAIL|DESTROYED|corrupt|cannot|unrecover)')
> if [ "${condition}" ]; then
>         emailSubject="`hostname` - ZFS pool - HEALTH fault"
>         problems=1
> fi
> 
> Given that the status is none of the above I infer that one or more of
> the words |corrupt|cannot|unrecover| appeared in whatever message
> zpool status emitted.  However, running this command from the command
> line produces nothing containing any of the trigger words.

UNAVAIL in "still be used, but some features are unavailable".

--
Herbert