Search

3.2. RAID

download PDF

3.2.1. Upgrades

Performing an upgrade from a dmraid set to an mdraid set is not supported. A warning will be displayed when an upgrade of this type is attempted. Upgrades from existing mdraid sets and creation of new mdraid sets are possible.
The new default superblock can cause problems when upgrading sets. This new superblock format (used on all devices except when creating a RAID1 /boot partition) is now at the beginning of the array, and any file system or LVM data is offset from the beginning of the partition. When the array is not running, LVM and file system mount commands might not detect the device as having a valid volume or file system data. This is intentional, and means that if you want to mount a single disk in a RAID1 array, you need to start the array having only that single disk in it, then mount the array. You cannot mount the bare disk directly. This change has been made as mounting a bare disk directly can silently corrupt the array if a resync is not forced.
On subsequent reboots, the RAID system can then consider the disk that was not included in the array as being incompatible, and will disconnect that device from the array. This is also normal. When you are ready to re-add the other disk back into the array, use the mdadm command to hot add the disk into the array, at which point a resync of the changed parts of the disk (if you have write intent bitmaps) or the whole disk (if you have no bitmap) will be performed, and the array will once again be synchronized. From this point, devices will not be disconnected from the array, as the array is considered to be properly assembled.
The new superblock supports the concept of named mdraid arrays. Dependency on the old method of array enumeration (for instance, /dev/md0 then /dev/md1, and so on.) for distinguishing between arrays has been dropped. You can now choose an arbitrary name for the array (such as home, data, or opt). Create the array with your chosen name using the --name=opt option. Whatever name is given to the array, that name will be created in /dev/md/ (unless a full path is given as a name, in which case that path will be created; or unless you specify a single number, such as 0, and mdadm will start the array using the old /dev/mdx scheme). The Anaconda installer does not currently allow for the selection of array names, and instead uses the simple number scheme as a way to emulate how arrays were created in the past.
The new mdraid arrays support the use of write intent bitmaps. These help the system identify problematic parts of an array, so that in the event of an unclean shutdown, only the problematic parts need to be resynchronized, and not the entire disk. This drastically reduces the time required to resynchronize. Newly created arrays will automatically have a write intent bitmap added when suitable. For instance, arrays used for swap and very small arrays (such as /boot arrays) do not benefit from having write intent bitmaps. It is possible to add a write intent bitmap to your previously existing arrays after the upgrade is complete using the mdadm --grow command on the device, however write intent bitmaps do incur a modest performance hit (about 3-5% at a bitmap chunk size of 65536, but can increase to 10% or more at small bitmap chunk sizes such as 8192). This means that if a write intent bitmap is added to an array, it is best to keep the chunk size reasonably large. The recommended size is 65536.
Red Hat logoGithubRedditYoutubeTwitter

Learn

Try, buy, & sell

Communities

About Red Hat Documentation

We help Red Hat users innovate and achieve their goals with our products and services with content they can trust.

Making open source more inclusive

Red Hat is committed to replacing problematic language in our code, documentation, and web properties. For more details, see the Red Hat Blog.

About Red Hat

We deliver hardened solutions that make it easier for enterprises to work across platforms and environments, from the core datacenter to the network edge.

© 2024 Red Hat, Inc.