On Tue, Oct 30, 2012 at 10:45 AM, <Eric_YH_Chen@xxxxxxxxxx> wrote: > Hi all: > > I have some question about the limitation of CephFS. > Would you please help to answer these questions? Thanks! > > 1. Max file size It's currently set to an (arbitrary) 1TB. It can be set wherever you like but is limited by the "mds max file size" at MDSMap create time in order to bound the time the MDS can spend checking one file under failure scenarios. > 2. Max number of files Well, you'll have problems past 2^64 due to wraparound. There might be some more earlier than that but our inodes are always 64-bit so the number would be large. There are some more practical limitations related to scale. Currently we don't enable directory fragmentation by default (stability concerns), so your directories should be a fractional size of your MDS cache size. Your working set should be smaller than your MDS cache size. Etc. > 3. Max filename length POSIX/Linux limits. 1024 characters, I think? > 4. filename character set, ex: any byte, except null, "/" POSIX. I believe you're correct. > 5. max pathname length POSIX/Linux limits. 4096 characters, I think? > And one question about RBD > 1. max RBD size Hmm, I'm actually not sure on on this. Again, any limit here is either arbitrary or imposed by somebody else. -Greg -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
