Originally posted by: Kunjuttan
Originally posted by: Kunjuttan
Hi Cris,
Thanks for the update.Can you give me the full syntax to luxadm or mpathadm command.
Originally posted by: stephen2615
Hi Cris,Thanks for the update.Can you give me the full syntax to luxadm or mpathadm command.
Assuming it is a Solaris system, cfgadm is your friend and the sysadm guys will know all about it. Generally the luns presented to the host will be in order. So if they have unconfigured a lun on the Solaris host, they should be able to tell you which one (as in the order number) it is. You should be able to count it back in Storage Navigator.
I cant stress strongly enough that you have to chat with the sysadm guys about anything you intend on doing. If HDLM is on the Solaris system, it will help you find which lun needs to be removed.
To find out what luxadm or mpathadm does, just do a google search on them by using man luxadm or if you have an account on the host, do the same thing.
It is a daunting thing when you don't really understand Solaris. I had a system once with 192 LUNs on it but they were divided across the FEP on the USP. At least that breaks it down a bit. It is basically a matter of determining the FEP and then the LUN number on the port and cfgadm will give you that.
If you want to be really confused and then admire what the sysadm does, have a look at:
http://blogs.warwick.ac.uk/peggleton/entry/disks_cfgadm_fcinfo/
IF it is not Solaris, then it is a completely different thing.
Stephen
Originally posted by: cris
Yes cfgadm will also work; so many ways to skin a cat on Solaris. And I couldn't agree more with Stephen, get your sys guys involved. They should be telling you which LUN they want removed and diagnosing all the outputs.
Originally posted by: stephen2615
Yes cfgadm will also work; so many ways to skin a chat on Solaris. And I couldn't agree more with Stephen, get your sys guys involved. They should be telling you which LUN they want removed and diagnosing all the outputs.
I just discovered something that will help you immensely. It's a EMC document it will give you exactly what you need to find out.
https://community.emc.com/docs/DOC-12191
I have a cat here that needs to be skinned...
Originally posted by: cris
Oh no, we are now taking content from other vendors communities!!!!
Good document....nice find.
I just discovered something that will help you immensely. It's a EMC document it will give you exactly what you need to find out.
[url="https://community.emc.com/docs/DOC-12191"]https://community.em.../docs/DOC-12191[/url]
I have a cat here that needs to be skinned...
Originally posted by: Kunjuttan
Hi Cris/Stephen,
Thanks for your valuable inputs.The document shared by stephen is very useful.Is there any same kind of document available for Hitachi too.
As you said, I am asking the system admins to give the exact LUN ids, but they are providing only the one which I shared.From that we couldnt find the exact LUN.
Originally posted by: cris
The document was fairly generic and had OS native commands, it can easily be used in a hitachi context.
As for your problem, you'll need to get the OS team involved. We can TRY to help if you can provide some useful outputs, but not with what you've currently provided.
Originally posted by: stephen2615
Hi Cris/Stephen,Thanks for your valuable inputs.The document shared by stephen is very useful.Is there any same kind of document available for Hitachi too.
As you said, I am asking the system admins to give the exact LUN ids, but they are providing only the one which I shared.From that we couldnt find the exact LUN.
That document from the EMC site is not storage related in anyway. It is a procedure for Solaris.
You have been given c2t201200A0B84829E2d9 which looks very close to the output from the Solaris command format. The issue I see here is that the World Wide Name (WWN) is not what I would expect to see. I know Solaris can do some strange things especially from the output of cfgadm -al. It has been five years since I last did this. 2012 (the first four numbers at c2t) looks very much like the WWnN of a HBA or perhaps a Cisco switch but not the WWN of a storage array.
What you should be seeing is something that starts with 50060e8005277210 which is the WWN for a HDS array. Without being onsite to see the output from all those commands in that EMC file, I can't give you any advice. The output from cfgadm -al -o show_FCP_dev and luxadm probe would be very useful indeed. If the sysadm has unconfigured the LUN but it is still attached, it will be easily identifiable.
Whatever this is, I would take a stab and say it is LUN 9 (d9). Alas I don't know if the lun number starts at zero or one.
IF the sysadm goes through that EMC document, they have to be able to let you know what lun it is.
Cheers
Stephen
Originally posted by: Kunjuttan
Thanks Cris/Stephen,
I am working with OS team in this issue with the help of the document provided.
Originally posted by: Newbie
Thank You Stephen for sharing the document. It is very helpful.
Originally posted by: cris
That looks like a Solaris MPxIO device. You'll need to find out what the underlying device that is maps to is first. Use luxadm or mpathadm to figure this out and then use the LUN UUID or the H-LUN ID to figure it out. Alternative, you can use HDLM, HCS, or some other native host tool. hrmmm, maybe even a getconfig?