I have a storage network for client virtual machines on which aoe block
devices can be created and deleted dynamically by users. I'm using vblade as
the AoE target, and the stock linux AoE block driver from 2.6.25.11, i.e.
aoe.c v47. I've (trivially) patched the linux aoe.c to make aoe_deadsecs a
kernel parameter as well as a module parameter, as I compile the driver into
our kernels but still need to change aoe_deadsecs to a small (4s) value.

When a new AoE device appears on the network, every machine spots it quickly
and udev promptly creates a /dev/etherd/eM.N device. However, when a device
is removed (i.e. its vblade is terminated), the devices appear to stick in
/dev/etherd on the initiators.

I'm aware of /dev/etherd/flush and aoe-flush, but nothing I do with these
appears to cause the devices to be deleted by udev in the same way that,
say, a removed USB drive would be removed from /dev. (What's more, removing
them manually appears to be a bad idea because then they don't get recreated
by udev if a new target with the same shelf and slot number appears!)

I'm sure I'm missing something obvious, but what is the correct usage for
/dev/etherd/flush and/or /dev/etherd/revalidate to get the kernel to
properly remove all of the dead devices?

Cheers,

Chris.

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Aoetools-discuss mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/aoetools-discuss

Reply via email to