Jes Sorensen <Jes.Sorensen@xxxxxxxxxx> writes:
> NeilBrown <neilb@xxxxxxx> writes:
>> On Mon, 02 Mar 2015 11:32:16 -0500 Jes Sorensen <Jes.Sorensen@xxxxxxxxxx>
>>> Neil,
>>>
>>> Finally had time to go back and look at this - I don't quite understand
>>> your request here.
>>>
>>> As far as I can see, the above code isn't run at all during device
>>> creation, so I don't get how modifying it as you suggest will make
>>> /dev/md/111 appear if I create /dev/md111?
>>>
>>> Cheers,
>>> Jes
>>
>> Wheels within wheels of deception and indirection....
>>
>> mdadm doesn't create the devices directly, udev does that.
>>
>> IMPORT{program}="BINDIR/mdadm --detail --export $devnode"
>> ENV{DEVTYPE}=="disk", ENV{MD_DEVNAME}=="?*", SYMLINK+="md/$env{MD_DEVNAME}"
>>
>>
>> So the symlink from /dev/md gets created iff "mdadm --detail --export"
>> reports an MD_DEVNAME.
>>
>> Hence the change suggested.
>>
>> mdadm *can* create the devices itself, but only if udev isn't running, or
>> mdamd has been explicitly told to ignore udev. Normally that doesn't happen.
>
> Ahhh, I didn't see anything happening on my system, but it may be the
> RHEL/Fedora udev rules file is missing something. I'll have to take a
> closer look there!
Actually I remember why I didn't get that far - I added some debug code
to the place you suggested I modify, and it never triggered during
Create.
Cheers,
Jes
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
