12c ASM blending 11g db and 12c db, error ora-600

A database of customers is a 12c cluster environment, with 11g DB and 12c db in the environment. When the customer is ready to pull up on a diskgroup database (this cluster is placed in multiple test libraries, it is centralized from dozens of production library disk through storage-level replication. There's a cluster in the production library. Only one database, only 2 diskgroup (data and Fra), but this test library centralized multiple production libraries, so there are more than 70 diskgroup), the error:

Ora-15001:diskgroup "DATA_MIDG" does not exist or isn't not mounted
Ora-00600:internal error code, arguments: [Kfnslavegroupuse], [Kfgn_max], [67], [3352857312], [], [], [], [], [], [], [], []

In fact, this DATA_MIDG is there, in ASM, the state is also mount.

This is actually because the bug 19404632 GRID 12.1 asm with DB 11.2 hits ORA600 [Kfnslavegroupuse] [Kfgn_max] ' When # of ASM Disk Groups > 63 (Doc ID 19404632.8)

Although, in the 12c, the number of DiskGroup can already reach 511, but this requires the database version of the coordination, the database must be 12c, can be used to DiskGroup 64 after the DiskGroup. If the database is 11g, it can be used before the DiskGroup number is 63 (inclusive), and DiskGroup number is 64 and will encounter the above ora-600 [Kfnslavegroupuse],[kfgn_max]. 67] The error. Here's 67, which means number 67th DiskGroup, which is the +data_midg of my controlfile.

The internal documentation for the bug also says that the Kfnsgorupuse function rejects a disk that has a database version less than 12c and that has more than 63 of DiskGroup number using a diskgroup numeber greater than 63rd. This is hardcode in code and cannot be adjusted through an implied parameter:

At present, although there are patch 19404632, but this patch will only ora-600 error information, converted into meaningful message, the actual need for the database version of 12c.

So the thing we can do is to plan ahead, for this 12c ASM Mix 11g db and 12c db, try to get 11g db to use the DiskGroup before number 63rd, and 12c db with DiskGroup after number 64th. This may require us to establish several diskgroup in advance to carry out "pit".