1 |
Am 23.12.2014 um 21:40 schrieb Rich Freeman: |
2 |
> On Tue, Dec 23, 2014 at 3:27 PM, Stefan G. Weichinger <lists@×××××.at> wrote: |
3 |
>> |
4 |
>> got my first two demo nodes up and in-sync ... what a success ;-) |
5 |
> |
6 |
> I started to look into ceph, and my biggest issue is that they don't |
7 |
> protect against silent corruption. They do checksum data during |
8 |
> transit, but not at rest. That means that you could end up with 3 |
9 |
> different copies of a file and no way to know which one is the right |
10 |
> one. Simply storing the data on btrfs isn't enough - that will |
11 |
> protect against files changing on the disk itself, but you could STILL |
12 |
> end up with 3 different copies of a file on different nodes and no way |
13 |
> to know which one is right, if the error happens at a higher level |
14 |
> than the btrfs filesystem/disk. |
15 |
|
16 |
but ... oh my. *sigh* |
17 |
|
18 |
I assume the devs there have a clever answer to this as well? |
19 |
|
20 |
At least for the future ... now that btrfs is declared stable at least |
21 |
for the more trivial setups (read: not RAID5/6) by Chris Mason himself |
22 |
... btrfs should be usable for ceph-OSDs soon. |
23 |
|
24 |
In the other direction: what protects against these errors you mention? |
25 |
|
26 |
S |