Chris Hoogendyk wrote: > Robert Hajime Lanning wrote: > >> Eric Sorenson wrote: >> >>> On Jul 2, 2009, at 8:53 AM, Atom Powers wrote: >>> >>> >>>> I didn't see AFS mentioned yet. My, admittedly incomplete, >>>> understanding of afs is that it provides a single namespace (directory >>>> tree) to all clients but the files themselves may be stored on a local >>>> or remote server; a bit like Microsoft DFS. >>>> >>>> Would this not unify the storage while maintaining local access speeds >>>> for files created locally, and still allow any host/client the ability >>>> to access any file on any server? >>>> >>> You really don't want to make every server a fileserver in AFS though, >>> the way the OP requested. File servers are special in AFS in that you >>> have to carefully (and manually) manage backup copies of volumes so >>> there's always n+1; if a volume with no backup copies goes away the >>> whole cell comes to a screeching halt. >>> >> You don't have to. The advantage of AFS is that is uses a local disk as >> cache. And it can work in offline mode. Though, I don't know it's >> limits in doing that. It might be RO when offline. >> >> Moose! care to comment? :) >> > > > > I snagged the thread quoted above, but started a new thread so as not to > be a hijacker. ;-) > > Anyway, this is exactly the sort of knowledge I'm looking for with > regard to ZFS, and have had trouble finding. I want to know what the > failure modes are for a zpool, and what the implications of that are for > how you configure a system. > > For example, I have 4 extra internal drives that I have configured as > raidz in a zpool and defined several zfs file systems on. > > Now, I have an external J4200 with 12 drives. I want to use that in zfs > also. > > Suppose I combine it all in one single large zpool. Then I have an > internal raidz and an external .... (I'm thinking raidz2 with one hot > spare -- that gives me more redundancy than two raidz and more space > than two raidz2). Now suppose I'm just not paying attention, or I'm on > vacation, or an emergency order for a new drive ends up snafu, and I > lose 2 drives on the internal raidz. That would mean that the internal > raidz is simply lost. What happens to the zpool? I presume with Sun's > emphasis on data protection and error recovery that it would degrade and > I would not lose stuff that was on the external array. Perhaps it would > even be smart enough to start copying stuff out when the internal raidz > had lost one drive. But I don't really know, and haven't been able to > find answers. Therefore, I'm thinking maybe I should reduce risk at the > expense of system complexity and less flexibility and just have a second > zpool. Then I have to decide which zpool to put a file system in, think > about space usage, etc. > > Anybody know? > > if you have one zpool and you lose a raidz2 in it, your zpool is basically toast. It stripes across all the components inside it. Does it copy data off of a failing raidz2? no. You should have adequate spares and redundancy to recover from there. One nice thing you can do is replicate (if so inclined) very easily from one zfs server to another at regularly defined intervals, including snapshots. So if the first one fails, you can point all the clients at the second one. We typically have larger stripe sizes on our secondary and compression turned on for more storage space and lower overhead. We can backup 2 primaries to a secondary this way.
_______________________________________________ Tech mailing list Tech@lopsa.org http://lopsa.org/cgi-bin/mailman/listinfo/tech This list provided by the League of Professional System Administrators http://lopsa.org/
