Description of problem: Version-Release number of selected component (if applicable): Xenserver 6.5 sp1 Host all updates Write speed when mounting glusterfs as nfs in fstab in xenhost i get 100MB w/s fstab entry: 192.168.0.33:/sr-tier2 /mnt/glusterfs nfs rw,hard,intr,rsize=8096,wsize=8096 0 0 mtab entry (from created SR Storage of xen that VM's using): 192.168.0.33:/sr-tier2/63aba108-3684-a37d-49ef-8b36018c05e7 /var/run/sr-mount/63aba108-3684-a37d-49ef-8b36018c05e7 nfs rw,tcp,hard,rsize=1048576,wsize=1048576,addr=192.168.0.33 0 0 2x Centos 7 Guest VM with GFS 3.7 and NFS volume replica with 2 nodes Speed test inside GFS VM with mounted glusterfs -> 250 MB/s (Glusterfs itself seems to be good but nfs server cuts performance) Centos 6.8 test vm which is running from the gfs shared nfs Write speed inside the test vm is at 30mb/s <--- PROBLEM Read speed inside the test vm is at 130mb/s How reproducible: use gfs as shared nfs storage for xen. Steps to Reproduce: 1. Create gluster volume with nfs 2. Mount NFS in XenCenter and also mount the nfs share inside xen with fstab mount 3. Copy Test VM on it 4. on both NFS native mount and SR make tests with hdparm and dd Actual results: VM writes very slow VM reads are good Expected results: VM writes equal to the speed of when xen mounting nfs nativly. Additional info: nfs async could perform better write speed in nfs. so i've tried the following paramaters inside gluster nfs.trusted-sync nfs.trusted-write with on and off because there it seems to treat packages as async, but i did not see any difference when adjusting them?
Also known problem in VMware: https://communities.vmware.com/message/761747
This bug is getting closed because GlusteFS-3.7 has reached its end-of-life. Note: This bug is being closed using a script. No verification has been performed to check if it still exists on newer releases of GlusterFS. If this bug still exists in newer GlusterFS releases, please reopen this bug against the newer release.