[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#335230: More stub code



Alle 20:09, martedì 25 ottobre 2005, hai scritto:
> I've been able to setup an almost-working evms box (at least evms_query
> reports useful data).

You have to boot from a emvs volume to have a proper evms box, or to look at 
it from an evms enabled cd-rom, but I think you already saw that.

> I tried following Erik's suggestions and also implemented part of the
> init() method and reworked some of Marco's code.
> Some parts are still to be finished, but I'd like some feedback
> expecially on evms_query commands and the expected content of some
> EvmsDev properties.

I saw the code, really great!
I think the properties are all here, for devices and for evmstab generally, 
good work!

> The init() function will try to discover all the evms devices,
> starting from `evms_query volumes` it will build an array of EvmsTab-s
> reporting the devno string as reported by evms_query info
> ("$major:$minor"), the volume name and a list of disks.

Really nice.

> I'm currently interested in understanding which information we need to
> gather, the tryEvms can be fixed later (it'll probably need to loop on
> the EvmsDev->devices array as the tryRaid does).
> Am I on the right way?
>
> I'll add also the plugins support shortly.
>

Yes, we need only to scan for plugins if we like.

The list of all plugins needed is given by, in shell terms and with my system 
examples:

# evms_query volumes | xargs -i evms_query plugins  \{\} | uniq
Ext2/3
XFS
SWAPFS
# evms_query regions | xargs -i evms_query plugins  \{\} | uniq
LVM2
MDRaid1RegMgr
MDRaid5RegMgr
# evms_query disks| xargs -i evms_query plugins  \{\} | uniq
LocalDskMgr

So we need to do the equivalent of calling evms_query plugins for each volume, 
region and disk, the we need a link to lib name; I think that for the time 
being a simple 2 colum structure with (name of the plugin | name of the 
library) would be enough to have evms working, but also it could be not 
future proof.


-- 
ESC:wq



Reply to: