Lun Masking

Hello
I have an xserve raid (firmware 1.5.1/1.5.1.c) configured with 3 slices and Lun Masking which 3 xserves connect to. They are all attached to a qlogic fc switch.
Recently the company bought 2 more servers to support another area and they need them to mount 2 of the slices. I added the WWN to the corresponding zoneset in the switch, and added the WWN to the lun masking configuration (using raid admin 1.5) and i got the volumes mounted in the new servers.
The next day i recieve a phone call asking me to switch the volumes from the servers, slice 2 to server 4 and slice 1 to server 5. After making the appropiate changes in server admin and rebooting the servers the volumes in both servers would not mount. In Disk Utility i can see the volume and the corresponding raid slices, however Disk Utility tells me they are degraded and that i should run first aid on the disk. The first aid does nothing to get the volume mounted on the servers.
If i put the lun masking settings back (slice 1 to server 4) it mounts without a problem. Is there a cache on the lun masking settings? Is it the firmware version? a downgrade and upgrade perhaps?
If anyone has encountered something similar, any help would be appreciated.
Thanks!
Raul Basurto Rosenzweig

Hello William
I did reboot the hosts. At least the 2 i'm trying to get the volume to mount. The other 3 server that also use that raid, i can't reboot them, at least not during the daytime, since they are in production.
I did reboot the raid and the hosts.

Similar Messages

Trying to Disable LUN Masking under 1.5.1

Hi,
I have an Apple XServe Raid with 7x750Gb drives. The array is configured with three slices which would mean three LUNs. Even though Lun Masking is no longer an option in 1.5.1 it still shows up as enabled on the XServe Raid. Non of the luns are visible to the connected host. There is no switch involved.
How can I disable Lun Masking under 1.5.1?
Thanks,
Jason

Hello nosajesahc, and welcome to the Appleboards,
Do you have a copy of the the Admin Tool version 1.5? LUN Masking was present in that version of the tool and think you should be able to turn it off and the resume using the current version.
HTH,
=Tod
http://www.apple.com/support/downloads/xserveraidadmintools15.html

LUN masking and zoning question.

Hi all,
If my storage array presents a number of LUNs/LDEVs down two physical fibre cables which are outputs from the primary and secondary controllers, and those two cables are ports on the MDS switches, do you need to know the LUN number as it was created on the array?
I read the Fabric Manager Configuration Guide and it talks about LUN masking at the array. It says to check the Check LUN box. I have not tried this so I was wondering if the LUNs on the array will appear for me to select. If there are say 8 LUNs presented (4 primary and 4 secondary), are they visible when using this option?
I have a fair sized number of new blades with HBA's with dual ports that need to connect to a large HP array. The setup will be one path goes from a 9216 to a 9509 via port channels (and another path through a similiar setup) to the HP array which will be presenting LUNs through LUN masking to the individual WWN's of the blade HBAs ports. There will be literally hundreds of LUNS coming out of the array so if they are presented in Fabric Manager, it would make my life easier especially when the array is looked after by HP and they might not be around when I start configuring the SAN.
Thats the joy of outsourcing.
Any help would be very welcome.
Regards
Stephen

Finding out what LUNs are available is done via SCSI commands INQUIRY and REPORT_LUNS. In general, having the MDS discover the luns is not done since the MDS does not need to know. The end devices that are zoned and configured in the array's LUN security will need to know and they will use the two above frames to find this info out. The check luns checkbox that you reference is when doing LUN zoning (which is part of Enterprise license). In fact, the manner in which we are able to zone based upon a LUN is by trapping and manipulating the REPORT_LUNS frame. A generic implementation is to simply zone an initiator and a target via their PWWN's. Then, in the array, there is a security mechanism that allows the PWWN of a device to access a particular LUN. That LUN is then associated with a particular volume of disk space. However, we offer the capability to zone with the granularity of the LUN/PWWN combination if you so desire it.

So, what happened to LUN Masking

I noticed in the release notes for the new firmware update (1.5.1) for the Xserve RAID that LUN masking has been removed from the RAID Admin utility...
And low and behold... yup... it's not in the admin utility anymore... anywhere...
So, what's with the removing of a pretty important feature with a minor revision?
everything... Mac OS X (10.4.8)

I'm not sure what your IT background is, but LUN masking in a storage environment is NOT an advanced feature, and the concept of hardware RAID is common in any serious server platform.
"Advanced" need not equate to complexity, enter Apple.
Regardless, it is simply unacceptable to remove a core functionality you've provided since the inception of a product with no advance notification that said functionality is going to be removed, and no documentation other than "Removed LUN Masking from Advanced tab" when it does happen.
So Apple needs to charge $1.99 to enable previously shipped Core 2 Macs to use Wireless-N, in order to comply with Sarbanes-Oxley. Okay, fair enough. So how much $ is Apple going to give back to me per Xserve RAID that my company has purchased, since we paid for functionality that they're now taking away? Just some food for thought.
And yes, I have notified both my enterprise account rep and my field engineer of my disatisfaction with this issue. I'm an Apple fan-boy, and even I think this is ridiculous on Apple's part.

LUN masking on or off?

Hello, may be this is stupid question.
Since yesterday, I used only one disk array on the right side of XServe Raid. I created raid5 array on left side, but this array is not visible on my xserve. New array is online and I have last firmware on both controllers also. How can I arrange this? I tried it with LUN Masking enabled and disabled on both controllers, but was not successful. XServe RAID is connected to xserveG4 only - both fiberchannel cables are connected to one machine. Enclosure is first generation XServe RAID.
Thanks for info.
Leos

Have you tried shutting everything down and powering up your XServe G4 first, then once it's up powering up your RAID? Or vice versa? It tends to get moody if you don't do it in the right order.

LUN Mask changes without reboot?

Are there any controller firmware enhancements in process to allow LUN masking changes without a controller reboot? We have several XRAIDs in our SAN with masking enabled. With the current limitation we must schedule several servers to be down before making the changes. That's really a pain...
Other OS

Apple won't typically comment on any future products or enhancements.
However, your request is a common one. I suggest you contact Apple (via web site or via your sales rep, if you have one) and suggest the enhancement and let them know your pain.

Lun Masking limited to 8 WWWN's?

I have a couple 7TB xServe RAIDs, with each bank of drives configured for (1) RAID 5 array (6 drives, 1 hot swap) & sliced into 6 LUNs. Everything works as expected until I reach 8 entries in the LUN Masking table. The option to add a ninth host is greyed out. If I remove any the the working 8, the options is available again. Is there any work-around for this? The docs don't indicate a restriction on the number of hosts that can be specified. With the above config, the 8 LUN per controller limit is not passed, so I don't understand why the number of hosts is giving a problem.
thx,
Vince

Thanks for the reply. I am using VMware ESX server on the hosts that will be sharing LUNs. It handles the sharing natively, and also handles multi-pathing but the 8 host limit means that I can only have 4 hosts multi-pathed to the xServe.
I'd also like some LUNs to be masked exclusively for other non-ESX servers to avoid contention but the 8 host limit is making that impossible.
I have been very satisified with the xServe RAID so far except in the limitations (6 slices, 8 LUNs, and now 8 hosts for LUN masking). These are artificial limits that would make the xServe a better product if addressed.

LUN masking and mapping in Nexus 5548up OS 5.1(3)N2(1)

Hi,
I just like to know how we can do Boot LUN Masking for ESXi Host in nexus 5548 , that only use by specific host and other host cannot see it.

If you are looking for MDS like LUN zoning, Nexus 5000 does not support it. You will need to configure LUN masking on your storage array

RAID, clients, switch and LUN masking

Hello,
To cut to the chase of it, I'm not sure if my LUN masking is set correctly.
I have a RAID unit with 4 slices. In RAID admin I have set the LUN mapping and masking according to the instructions in the RAID admin guide.
I also have 4 xserves ( all 10.4.6 ) - each is assigned ( mapped to and masked from ) its respective slice of the RAID through an emulex 355. The emulex utilizes a configuration file specified by emulex corporation for this specific apple/raid configuration.
User home directories reside on the RAID slices.
PROBLEM: When I log in a user on a desktop client ( 10.4.6 ) I can see the shared homedirectories on the other slices of the RAID via private/Network/Servers. This makes me nervous.
Note that the xserves DO NOT mount the other RAID slices.
Am I worrying over nothing or did I miss something in my setup? I'll be glad to post further details if requested.
Thanks for you time.
Anthony
Xserves, RAID, Emulex 355 Mac OS X (10.4.6)

p.s.
we are NOT using xsan.
and, to clear up a statement in the orginal post, xserves are mapped TO thier respective RAID slice and masked FROM the other slices.
Anthony
Mac OS X (10.4.6)

Firmware 1.5.1 + LUN Masking

Is it possible to still use LUN masking on a controller with 1.5.1 firmware, as long as I'm using RAID Admin 1.5.0?

I'm not sure what your IT background is, but LUN masking in a storage environment is NOT an advanced feature, and the concept of hardware RAID is common in any serious server platform.
"Advanced" need not equate to complexity, enter Apple.
Regardless, it is simply unacceptable to remove a core functionality you've provided since the inception of a product with no advance notification that said functionality is going to be removed, and no documentation other than "Removed LUN Masking from Advanced tab" when it does happen.
So Apple needs to charge $1.99 to enable previously shipped Core 2 Macs to use Wireless-N, in order to comply with Sarbanes-Oxley. Okay, fair enough. So how much $ is Apple going to give back to me per Xserve RAID that my company has purchased, since we paid for functionality that they're now taking away? Just some food for thought.
And yes, I have notified both my enterprise account rep and my field engineer of my disatisfaction with this issue. I'm an Apple fan-boy, and even I think this is ridiculous on Apple's part.

Replacement for LUN masking

Hello,
I currently have an Xserve RAID with 14x400GB drives running firmware 1.5/1.50f. Each side is a RAID5 array split into two LUNs (total four LUNs). I have a Qlogic SANbox 1400 fibre channel switch, and each LUN is attached to four computers (1 Leopard, 1 Ubuntu, 2 Windows). I use LUN masking to make sure each computer can only mount the appropriate LUN.
Anyways, I'm considering upgrading to firmware 1.5.1 for 750GB drive support, but I understand the RAID Admin tool no longer supports LUN masking (though the firmware on the XRAID seems to).
My question is, how do I use Zoning on the Qlogic switch as a replacement for LUN masking? I often see Zoning mentioned as a substitute for LUN masking, but I can't figure out how to zone down to the LUN level. I can create a zone that include the WWPNs of the HBA and the Xraid controller, but not down to the LUN level. The result is that the computers have access to LUNs they shouldn't.
What am I missing? Is the Qlogic 1400 incapable of this?
Thanks!

I have sliced the LUNs.
I've come across a small number of other posts suggesting the same thing - upgrade the firmware to 1.5.1 and use the old 1.5 tool for LUN masking administration. I'm OK with this if it does indeed work, I just don't want to experiment with production storage

How to automatically masking/unmasking LUN for ip failover?

Hi friends
I need to connect 2 Xserve quad xeon to an Xraid through a Qlogic SANbox 5600Q
All I need to do is to create an ip failover system between the two Xserve.
I don't need a xsan solution, so all I want to do is an "automatic" masking / unmasking of the Xraid when the master Xserve go offline.
Apple manual speaks about LUN masking, but doesn't explain how to made it.
Anybody have some idea?
ps. sorry f my bad english

Thanks William
The problem is maybe nearly solved, I've created the 2 zones that I need and I've made a little script for the zone switching.
Premised that I'm not a script expert;
#!/bin/bash
sleep 3
echo -e "Fake\r" #first fake password for jumping to login
sleep 3
echo -e "login\r" #switch user login
sleep 3
echo -e "password\r" #switch password
sleep 3
echo -e "admin start\r" #start the administration session
sleep 2
echo -e "zoneset activate DestinationZone\r" # start newzone
sleep 2
echo -e "admin end\r" #end the administration session
} | telnet xxx.xxx.xxx.xxx;
I don't like the temporary solution, but at the moment Qlogic support it has not still answered to my questions.
I've tried also with ssh but at the moment does not work, login problems (I dont know if I can create public and private keys on switch).
Any suggestion is welcome.

Is there a way to make LUNs for apple XRAID visible to XSAN 3?

I have an apple XRAID with two controllers and 14 750GB drives. This and the server is connected to a silkwork 200e. I can mount the raid in Disk Utility, etc. but I can't see the LUNs in XSAN 3.
Interestingly if I break the array in disk utility and then re-create it the LUNs will show up temporarily in XSAN.
I took a look at the array in RAID Admin, but the LUN masking is turned off as far as I can tell.

You can't use a software RAID set with Xsan. Break the array. Partition each of the two LUNs as free space. If necessary, use the "dd" command in TS1268 to wipe out existing filesystem metadata. You should be able to label the LUNs for Xsan then.

No LUNs in Xsan admin

I'm using Xsan 1.1 on OSX 10.4.7 and I added an new Xserve RAID into the mix. But the trouble is that the LUNs are not showing up in Xsan Admin (v.1.3).
`cvlabel -l` shows the two LUNs, labeled "unknown". so does `diskutil list`.
All Xserve RAIDs have the same firmware, no LUN masking is on, and all the ports on the fibre switches are in the same zone.
Disk Utility.app can see the LUNs and will let me make a software RAID combo of them , but that's what I want. What I would like is to expand two of my data-only storage pools, if they'd show up in Xsan admin.
I've looked the manuals,man pages, discussions and so forth.
Any ideas?
Thx.

I reveiwed the following documentation for adding LUNs to storage pools:
http://docs.info.apple.com/article.html?artnum=303571
http://docs.info.apple.com/article.html?artnum=303570
http://docs.info.apple.com/article.html?artnum=301911
And it worked. Sort of.
I shutdown all the clients, except 2 or 3 (mdc, md backup and my admin Mac). Then I unmounted the volume from last remaining clients, and I stopped the volume. Then I proceeded to backup the cfg, check the volume, and add the luns to an existing storage pool. All went well, until it failed to update the volume with the new cfg. I reverted to the old cfg and added the LUNs to a new storage pool. That worked.
Thanks for everyone's thoughts on this one. I took more time to do it this time, then I did a month of so ago. Careful planning and backups (cfg and data) are helpful to peace of mind.
-x

Problems presenting iSCSI luns to Server2012r2

So I am trying to cluster 2012r2 OS and SQL and I ran into this problem.. First the back ground.... I have a super beefy computer 32 gigs (1TB SSD) running VM workstation 11. In there I have one 2012 running AD (my DC) and also SQL and one computer
running EMC Celerra appliance.., one computer running ESXi 5.5 and one computer running Vcenter.... So I am able to present using Celerra, iSCSI storage to my ESXi 5.5 no problem. Within ESXi 5.5/Vcenter, I have two server 2012r2 that are joined to my
domain... no problems there. I added the Initiators and I am able to see my Celerra IQN target, and on the Celerra side, I can see both of my nodes.... I even made sure I went in to (on the MS side) the advanced tab under discover
portal and took out the two default and used the MS iscsi initiator adapter and ip addy of the server there.... Unless I am missing something, I should be able to go to disk management and refresh and my disks should show up... Does anyone
else have any ideas of what to try?
2 things.... I hear you need scsi3 per connections to be able to do iscsi in 2012r2.... I thought Celerra was able to do this and I had no trouble presenting to VMware directly...so I am not sure that is a problem unless I need to enable something
on the celerra...
Second thing is I am only using 2 network adapters for each Node.... one for the heartbeat and am sharing the other 2 for data network traffic.... I didn't figure that would be a big deal since this is just a learning lab... I am going to split those apart
now and see if that helps.... Anybody have any ideas for me? TIA

Hello and thank you for your time to respond. Actually I figured out my problem late last night. I did as you said added separate adapters to use for the storage, but that wasn't my "real" problem. My real problem, and I don't know how this happened,
but in the lun masking section of the Celerra, I had the target (Celerra's) initiator instead of the host (Microsoft's). How that happened, I don't know, but it was one of those situations where you can look at it for a day and not see anything
wrong. I replaced it with Microsofts from both nodes and voila... The drives appeared... If I may indulge a couple (3) of more questions though.... So, when I made the host machines, I used the local storage for both the C (os) and D (binary) drives on
both separate nodes. Then I presented the common storage of the Q(quorum) M (MSDTC) and E (Data) drives from the Celerra. All fine and dandy so far...This was late last night, so I might not have paid attention too closely, but somehow the Cluster hijacked
the local drive (D)... I am not sure how this happened, since it is being presented right off the datastore on Vsphere... I want (need) that D drive to stay independent (local).... That is how we do it in production at work anyways.. Second question
is when I ran the suitable for cluster tests, it came clean except the MPIO, which gave an alert, which I ignored... I am assuming that is fine since this is all done with one connection.. and third.. was when we cluster SQL on 2008, we would always add clusterDtc
and then create a empty container, call it MSDTC, and put the clusterDtc in it and then delete the original clusterDtc that is not in the container. It would not let me move the clusterDtc in the empty container and I was wondering if this was
even necessary as I never understood this step as our work practices it. I also was reading in two different spots that with 2012r2 that you don't need to failover clusterDtc anymore, and then another one that says that you still do AND that you should make
a separate one for each SQL service... As a final question.... Can you cluster reporting services in SQL? TIA again for any help...

Lun Masking

Similar Messages

Maybe you are looking for