EVA5000/6000 vdisk presentation/extention under Linux.

[Ferdinand Hiller]

Hello All,

We just got our EVAs this weekend. We are all new to the EVA and after attending a class and reading documents there are somethings which did not appear on any of the documentation. Does anyone have any experience doing the following or can point me to some documentation?

Hardware: Eva5000 & Eva6000, each on a separate department with 2 fabrics per department. Using Qlogic driver for fail over.
And business copy.

10 Windows 2000/2003 hosts and 20 Linux servers running ES3 and ES4.

1. Under Linux is it possible to increase the size of a vdisk while it is in use, and have the OS recognize the new size? We have been able to do this under windows, but cannot find any documentation on how to do it under linux.

2. What is the procedure for having a linux host mount a snapshot of a vdisk? Snapshots will never be presented to server which owns vdisk.

Any help or pointers are greatly appreciated.

Thanks for your time!

[Keith Reed]

You might want to check out GNU parted, which will let you extend the partition; I can't recall if it will let you do it on the fly or not. The reason I can't recall is that I usually use LVM on Linux, which naturally DOES let you do just that, and is the method I would recommend over using Parted anyway.

If you are using a file system that supports resizing (such as reiserfs), you can increase the size on the fly. You would first extend the size of the logical volume, then issue the reiserfs_resize (or similar command for your fs) to increase the size of the file system, all possible while in use. XFS also supports this, though ext2 does not.

It will take some extra work upfront to implment LVM2 and a fs like reiser, but believe me, you will reap the rewards down the line!

I'm not sure about the procedure for mounting a snapshot. Anyone else?

[Simon P.]

I have similar question regarding the growing the disk size on the fly. Initially I thought we did something last October when we grew the disk capacity on the fly and the Redhat picked up the change dynamically but were trying to find out what we did.