[pve-devel] ZFS plugin: Patch to solve issue with nexenta

[Alexandre DERUMIER]

Hi,

the patch seem to generate now similar guid than nexenta

600144f0a06f0700000055428993c0c1  34359738368          /dev/zvol/rdsk/MainPool/vm-296-disk-9 -> generate with last patch
600144f0a06f07000000558811e9da14  34359738368          /dev/zvol/rdsk/MainPool/vm-296-disk-8
600144f0a06f070000005c6f449515d5  34359738368          /dev/zvol/rdsk/MainPool/vm-296-disk-7
600144f0a26587000000519e535b0001  34359738368          /dev/zvol/rdsk/MainPool/vm-277-disk-1
600144f0a265870000005242d93c0001  34359738368          /dev/zvol/rdsk/MainPool/vm-291-disk-1-> generate with nexenta
600144f0a265870000005242dafb0003  34359738368          /dev/zvol/rdsk/MainPool/base-292-disk-1 -> generate with nexenta
600144f0a26587000000527922280001  34359738368          /dev/zvol/rdsk/MainPool/vm-296-disk-5
600144f675f1c980c38752a0647ba5e6  34359738368          /dev/zvol/rdsk/MainPool/vm-296-disk-6 -> generate with previous patch
600144ff713ed897226e5d64732418ff  34359738368          /dev/zvol/rdsk/MainPool/vm-292-disk-3 -> generate with previous patch

But It doesn't help for the gui error

My understanding is that nexenta generate a cache somewhere, and something it's not filled

some log:

Nov 06 17:03:09 (1:1.5) EXCEPTION: ObjectNotFound: ScsiDiskContainer::list_lun_mapping_entries(MainPool/vm-296-disk-7): Unable to locate the specified zvol 'MainPool/vm-296-disk-7' 

2013-11-06 17:10:08,255 nza ERROR com.nexenta.nms::/Root/Stmf/ScsiDisk::list_lun_mapping_entries(dbus.String(u'MainPool/vm-296-disk-9')): com.nexenta.nms.ObjectNotFound: Unable to locate the specified zvol 'MainPool/vm-296-disk-9' 



after waiting some minutes, or rlauching the nms service (webgui service) (#svcadm restart nms)

It's working again

(The module is in python, but closed sources, compiled .pyc, I can't see what it is doing)


restarting nms stop the web gui for some minutes,

but in the logs it's showing:
Nov 06 17:28:15 Starting...
Nov 06 17:28:15 Syncing devices...
Nov 06 17:28:22 Syncing LUNs...

So I think it's rebuild the luns cache or something similar


I'll try to find an easy way to reload this cache. (through ssh command)


----- Mail original -----

De: "Michael Rasmussen" <mir at datanom.net> 
À: "Alexandre DERUMIER" <aderumier at odiso.com>, pve-devel at pve.proxmox.com 
Envoyé: Mardi 5 Novembre 2013 19:28:52 
Objet: Re: [pve-devel] ZFS plugin: Patch to solve issue with nexenta 

On Tue, 05 Nov 2013 18:45:34 +0100 (CET) 
Alexandre DERUMIER <aderumier at odiso.com> wrote: 

> I'll try tomorrow ! 
> 
If the solution works it will cause no problems to only use 11 digits 
when naming a lun since max luns per target is limited to 2^14 or 16384 
and using 11 hex digits is an astronomic large number 11^16 

-- 
Hilsen/Regards 
Michael Rasmussen 

Get my public GnuPG keys: 
michael <at> rasmussen <dot> cc 
http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E 
mir <at> datanom <dot> net 
http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C 
mir <at> miras <dot> org 
http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 
-------------------------------------------------------------- 
There is an order of things in this universe. 
-- Apollo, "Who Mourns for Adonais?" stardate 3468.1