How to form BCV device from the available unprotected devices

Devices 1DC3 and 1DC7 was mirrored devices. When you convert a mirrored device to an unprotected device, you are splitting the 2 mirrors. One of the unprotected devices will use the old symdev while the 2nd unprotected device will take a new symdev. That's why you have brand new unprotected devices at the very end of your symdev list.

As you said, converting from Unprotected to BCV is harmless

Hey Stefano

Just want to add something more here.

Server:> cat /tmp/conv_bcv.txt
convert dev 1DC7 to unprotected ;
convert dev 1DC7 to BCV ;
convert dev 1DC3 to unprotected ;
convert dev 1DC3 to BCV ;

After doing symconfigure

I have got 2 BCV disk of 34 GB & 8 unprotected disk of 8Gb each. .

Status earlier
< 1DC3 Not Visible ***:* 01B:DA 2-Way Mir N/Grp'd (M) RW 34526
< 1DC4 Not Visible ***:* 16C:DA 2-Way Mir N/Grp'd (m) RW -
< 1DC5 Not Visible ***:* 11D:DA 2-Way Mir N/Grp'd (m) RW -
< 1DC6 Not Visible ***:* 06A:DA 2-Way Mir N/Grp'd (m) RW -
< 1DC7 Not Visible ***:* 12C:DA 2-Way Mir N/Grp'd (M) RW 34526
< 1DC8 Not Visible ***:* 05B:DA 2-Way Mir N/Grp'd (m) RW -
< 1DC9 Not Visible ***:* 02A:DA 2-Way Mir N/Grp'd (m) RW -
< 1DCA Not Visible ***:* 15D:DA 2-Way Mir N/Grp'd (m) RW -
---Status After conversion

10538a10539,10546


So there is no harm in creating BCV devices out of unprotected devices???

Thanks & regards

BCV and unprotected are almost the same .... The only difference is that a BCV is a "smart" volume that allows you to move its slices around.

With a "regular" unprotected device, you can't use TimeFinder/Mirror to take snaps of your data. With a BCV (that is an unprotected volume with an added special flag, the BCV flag) you can use TF/Mirror and establish/split your STD-BCV pairs.

Please note that you can also have 2-way mir BCVs (protected BCVs) and even RAID5 BCVs. But hopefully you don't have to deal with them .. at least not now

Stefano

One more query

stripe_size=1920; is it some default stripe value

Because I found below info about one device in our setup

Meta Configuration : Striped
Meta Stripe Size : 960k (2 Cylinders)

Thnaks for your help

Generally speaking when you create a metavolume you have to choose between concat or striped metavolumes. I assume you can easily guess the difference between concat and striped.

However in good old symmetrix days, we were able to choose between different "stripe size" when creating striped metavolumes.
Now the only allowed stripe size is 1920 blocks = 960kb = 2 cylinders, so you can choose between either striping with 1920 as stripe size or concat.
Striping at metavolume level is ALWAYS done at CYLINDER level so you CAN specify 1920 but the code will translate it into "2 cyls".

Now let's mess up everything ..

Solution Enabler DO translate by default between "old" and "new" track size. Older DMX and symms use 32K tracks. Newer (DMX3 and 4) uses 64K tracks.
If your storage is a DMX3, Solution Enabler will show you a stripe size of 1920 blocks. But 1920 blocks in a DMX3 are 1 CYLINDER (since each track is exactly double the old track size). So the stripe size is 960Kb (and it's always true) but the number of cyls depends on the storage. 960Kb = 2 cyls on older storages. 960Kb = 1 cyl on DMX3/4.
However remember that Solution Enabler by default DO translate between old and new cyls.