Greetings butter-heads, I would like to implement a redundant (raid1) disk array on heterogeneous disks using btrfs. A more detailed description of what I want to do can be found here:
http://superuser.com/questions/387851/a-zfs-or-lvm-or-md-redundant-heterogeneous-storage-proposal/388536 In a nutshell: organize your heterogenous disks into two "halves", the sum of which are of roughly equal size, and create a raid1 array across those two halves. For various reasons I decided to go with btrfs over zfs. What I have done is to create two lvm Logical Volumes, one using a single large disk, and another as a linear concatenation of several smaller disks. It works, so far, and I could automate it with some scripts. In the long term, I would like this to be something that btrfs could do by itself, without LVM. Having absolutely no knowledge of the btrfs code, this seems easy, I'm sure you'll tell me otherwise. ;) But one needs: 1) The ability to "group" a heterogeneous set of disks into the "halves" of a raid1. I don't understand what btrfs is doing if you give it more than 2 devices and ask for raid1. 2) Intellegently rebalance when a new device is added or removed (e.g. rearrange the halves, and rebalance as necessary) While btrfs seems to support multi-disk devices, in trying this, I encountered the following deadly error: creating a raid1 btrfs with more than 2 devices cannot be mounted in degraded mode if one or more are missing. (In the above plan, a filesystem should be mountable as long as one "half" is intact) With 1 of 4 devices missing in such a circumstance, I get: device fsid 2ea954c6-d9ee-47c4-9f90-79a1342c71df devid 1 transid 31 /dev/loop0 btrfs: allowing degraded mounts btrfs: failed to read chunk root on loop0 btrfs: open_ctree failed btrfs fi show: Label: none uuid: 2ea954c6-d9ee-47c4-9f90-79a1342c71df Total devices 4 FS bytes used 1.78GB devid 1 size 1.00GB used 1.00GB path /dev/loop0 devid 2 size 1.00GB used 1023.00MB path /dev/loop1 devid 3 size 1.00GB used 1023.00MB path /dev/loop2 *** Some devices missing Also I discovered that writing to a degraded 2-disk raid1 btrfs array quickly fills up the disk. It does not behave as a single disk. Both these errors were encountered with Ubuntu 11.10 (linux 3.0.9). I tried with 3.0.22 and I got "failed to read chunk tree" instead of the above "failed to read chunk root" and furthermore after mounting it degraded, I could not mount it non-degraded, even after a balance and a fsck. So, any comments on the general difficulty of implementing this proposal? Can someone explain the above errors? What is btrfs doing with >2 disks and raid1? Any comments on what parts of this should be inside btrfs, and which parts are better in external scripts? I think this feature would be extremely popular: it turns btrfs into a Drobo. P.S. why doesn't df work with btrfs raid1? Why is 'btrfs fi df' necessary? -- Cheers, Bob McElrath "The individual has always had to struggle to keep from being overwhelmed by the tribe. If you try it, you will be lonely often, and sometimes frightened. But no price is too high to pay for the privilege of owning yourself." -- Friedrich Nietzsche -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
