NFS + Infiniband problem
Rodney W. Grimes
freebsd-rwg at pdx.rh.CN85.dnsmgr.net
Mon Oct 29 15:06:38 UTC 2018
> Hello everyone,
> I have a several FreeBSD machines connected via Infiniband netwok ( FDR
> switch Mellanox SW3036 + ConnectX-3 VPI cards ).
> One of them is a NAS-server with multiply ZFS pools.
> All kernels (11.2-RELEASE on clients and 12.0-BETA1 (11.2 also tried) on
> server) are with infiniband connected mode (option IPOIB_CM, option SDM)
> and world with OFED stack support. (WITH_OFED='yes').
> File transfers via FTP or SSH between server and clients works almost
> flawless ( ~ 12 Gbit/s ).
> But when I try to copy in/out some significant data via NFS share mounted
> on clients, NFS i/o hangs at all or got extremely slow (couple kB/s)
> transfer speed after uncertain amount of copied data. For example, on the
> one node I can copy 1GB file, and after NFS hang on file with size 30 kb.
> Some details:
> [root at node4 ~]# mount_nfs -o wsize=30000 -o proto=tcp 10.0.2.1:/zdata2 /mnt
I am not sure what the interaction between page sizes, TSO needs,
buffer needs and all that are but I always use a power of 2 wsize
and rsize. You might try that. And as Rick suggested, turn of
TSO, if you can. Is infiniband using RDMA to do this, if so then
the page size stuff is probably very important, use multiples of
> [root at node4 ~]# dd if=/dev/zero of=/mnt/N1 bs=1m count=1024
> Ctrl-T for "hang" dd
> load: 0.01 cmd: dd 1061 [bo_wwait] 70.95r 0.00u 0.00s 0% 2112k
> load: 0.01 cmd: dd 1061 [bo_wwait] 72.89r 0.00u 0.00s 0% 2112k
> for "slow" dd
> load: 0.00 cmd: dd 2254 [nfsaio] 224.18r 0.00u 0.13s 0% 3132k
> load: 0.00 cmd: dd 2254 [nfsaio] 225.94r 0.00u 0.13s 0% 3132k
> I've tried mount with different wsize option with same result.
> Any help would be greatly appreciated.
> freebsd-infiniband at freebsd.org mailing list
> To unsubscribe, send any mail to "freebsd-infiniband-unsubscribe at freebsd.org"
Rod Grimes rgrimes at freebsd.org
More information about the freebsd-infiniband