My problem is not being notified in any way by the system or the filesystem about a device that does not work properly.
Everybody here feels not good if anybody lost hardware, or data. Because everybody fears it, and nobody wants it to happen to anybody else. Not to lose data is the game's middle name.
FreeBSD is not the kind of OS that bombards its users with unasked and so most of the times unwanted, so useless information aka spam.
Since (almost) every information is somehow available somewhere, and the system cannot know what you want to know, what's important to you when, and what's not, and while giving
every information available were
way too much - that made the system unusable (you were drowning within information within seconds),
it's up to you to ask it yourself about what you want to know.
It's within the very nature of any unix[like] system not only to give in the contrary to Window's incomprehensible 'Star Trek' gibber gabber comprehandable informations, but also to provide a whole shebang of tools and ways for automation.
Never forget: With your computer you're sitting in front of a machine that's original fundamental core entity is being for to automate things. That most computer users are not aware of, or forgot that, cause other systems degrade the computer to an electrical typewriter with porn player function is partially the fault of according operating systems, on the other part of users neither thinking of, nor caring for it.
Just with shell scripts, and cron jobs alone you can do
a lot.
In this particular case I don't even have a script. The following most primitive solution works for me.
All my
/root/.cshrc end with:
Code:
echo
uname -v
freebsd-version -u
echo
zpool status
echo
zfs list
echo
zpool list
echo
uptime
Everytime I login as root I get a brief overview over the system, and see if a drive from a pool failed.
It's my solution. It works for me. I see it often enough. If you want to see it more often, or need the system to inform you automatically, you need to find your own way.
Of course there is no 100% guarantee for never ever lose all my data. There never is. But I rely on the zfs pools' raid redundancy, plus the redundancy of backups done to other machines also using zfs pools with even more redundancy, to minimize the risk at a reasonable price that everything is irrecoverably gone before I notice any problem at all.
But that's just my idea.
Feel free to adopt this to your needs, or do it completely otherwise. If you ask others here you will get a whole bunch of other, even better ideas. Maybe a shell script that's done by cron hourly, using diff (or something else) to see if something has changed since the last capture, produce some output the way you like to inform you...
However, it's up to you anyway to figure out what fit your needs best.
All help can be provided here can only give you ideas about individual solutions what others do, helping about details with yours, and may help as far as possible when something went wrong.
As I already said at the top:
Everybody here feels not good if anybody lost hardware, or data. And everybody here wants to help - as far as it's possible.
Sounds like truism but it's just pure, simple basics:
After all prevention is always the best solution anyway. It sounds cynical when the desaster stroke, so most won't say it loud then. But if things are lost irrecoverably, all you can do is to learn from it for the future.
