Random (?) slow write speed to OMV under Xubuntu, Win7 is ok

    • OMV 3.x (stable)

    This site uses cookies. By continuing to browse this site, you are agreeing to our Cookie Policy.

    • Random (?) slow write speed to OMV under Xubuntu, Win7 is ok

      I understand it's likely NOT an OMV issue, but I will try seeking help here nevertheless.


      I am experiencing some very slow write speeds (about 12-15 MB/s, sometimes even less than that - over a gigabit network) under Xubuntu 16.04 lately.

      At first I thought it was due to the recent 3.x OMV upgrade but since it runs off a USB pendrive I immediately reverted to the old 2.x distribution and the issue is still there.

      Some details about the tests I made:

      - client PC is a Core i3 6100, 16GB RAM, main drive is a Samsung SSD, test data being copied to the NAS is on a separate WD Green 500GB hard disk
      - network is gigabit (ethtool confirms gigabit full duplex)
      - client PC has dualboot with Windows7 64bit, and NAS is working flawlessly there (100+ MB/s), as it did under Xubuntu until recently
      - iperf benchmarks on both Xubuntu *and* the NAS shows about 950 Mb/s
      - the network share is accessed via cifs/smb
      - sometimes write speed (data used for the test is usually huge files ranging from 500 MB to 2-3 GB) goes back to normal, then horribly crawls at 1 MB/s or even less

      Considering the huge performance differences between Xubuntu/Linux and Windows7 on the very same hardware I have to guess the issue isn't on the server side (i.e. the NAS) but on the client instead.

      But I really have no idea what to look for, so any suggestion is surely appreciated.
    • First of all, thanks for the quick reply! :)

      Shares are currently mounted via static entry in fstab (left basically untouched since forever, I had a "proper" NAS before switching to OMV so I just used the same IP and share names/path), but I am not a huge Linux expert unfortunately... I *guess* the answer is "no, I don't use any virtual filesystem" but could you please confirm this is the case?
    • Source Code

      1. sysfs /sys sysfs rw,nosuid,nodev,noexec,relatime 0 0
      2. proc /proc proc rw,nosuid,nodev,noexec,relatime 0 0
      3. udev /dev devtmpfs rw,nosuid,relatime,size=8130136k,nr_inodes=2032534,mode=755 0 0
      4. devpts /dev/pts devpts rw,nosuid,noexec,relatime,gid=5,mode=620,ptmxmode=000 0 0
      5. tmpfs /run tmpfs rw,nosuid,noexec,relatime,size=1630124k,mode=755 0 0
      6. /dev/sda2 / ext4 rw,noatime,errors=remount-ro,data=ordered 0 0
      7. securityfs /sys/kernel/security securityfs rw,nosuid,nodev,noexec,relatime 0 0
      8. tmpfs /dev/shm tmpfs rw,nosuid,nodev 0 0
      9. tmpfs /run/lock tmpfs rw,nosuid,nodev,noexec,relatime,size=5120k 0 0
      10. tmpfs /sys/fs/cgroup tmpfs ro,nosuid,nodev,noexec,mode=755 0 0
      11. cgroup /sys/fs/cgroup/systemd cgroup rw,nosuid,nodev,noexec,relatime,xattr,release_agent=/lib/systemd/systemd-cgroups-agent,name=systemd 0 0
      12. pstore /sys/fs/pstore pstore rw,nosuid,nodev,noexec,relatime 0 0
      13. cgroup /sys/fs/cgroup/cpu,cpuacct cgroup rw,nosuid,nodev,noexec,relatime,cpu,cpuacct 0 0
      14. cgroup /sys/fs/cgroup/devices cgroup rw,nosuid,nodev,noexec,relatime,devices 0 0
      15. cgroup /sys/fs/cgroup/net_cls,net_prio cgroup rw,nosuid,nodev,noexec,relatime,net_cls,net_prio 0 0
      16. cgroup /sys/fs/cgroup/pids cgroup rw,nosuid,nodev,noexec,relatime,pids 0 0
      17. cgroup /sys/fs/cgroup/freezer cgroup rw,nosuid,nodev,noexec,relatime,freezer 0 0
      18. cgroup /sys/fs/cgroup/blkio cgroup rw,nosuid,nodev,noexec,relatime,blkio 0 0
      19. cgroup /sys/fs/cgroup/perf_event cgroup rw,nosuid,nodev,noexec,relatime,perf_event 0 0
      20. cgroup /sys/fs/cgroup/hugetlb cgroup rw,nosuid,nodev,noexec,relatime,hugetlb 0 0
      21. cgroup /sys/fs/cgroup/cpuset cgroup rw,nosuid,nodev,noexec,relatime,cpuset 0 0
      22. cgroup /sys/fs/cgroup/memory cgroup rw,nosuid,nodev,noexec,relatime,memory 0 0
      23. systemd-1 /proc/sys/fs/binfmt_misc autofs rw,relatime,fd=24,pgrp=1,timeout=0,minproto=5,maxproto=5,direct 0 0
      24. hugetlbfs /dev/hugepages hugetlbfs rw,relatime 0 0
      25. mqueue /dev/mqueue mqueue rw,relatime 0 0
      26. debugfs /sys/kernel/debug debugfs rw,relatime 0 0
      27. fusectl /sys/fs/fuse/connections fusectl rw,relatime 0 0
      28. tmpfs /tmp tmpfs rw,noatime 0 0
      29. binfmt_misc /proc/sys/fs/binfmt_misc binfmt_misc rw,relatime 0 0
      30. // /mnt/NAS/mugghiera cifs rw,relatime,vers=1.0,cache=strict,username=bennyuser,domain=OPENMEDIAVAULT,uid=0,noforceuid,gid=0,noforcegid,addr=,unix,posixpaths,serverino,mapposix,acl,rsize=1048576,wsize=65536,echo_interval=60,actimeo=1 0 0
      31. // /mnt/NAS/bennyuser cifs rw,relatime,vers=1.0,cache=strict,username=bennyuser,domain=OPENMEDIAVAULT,uid=0,noforceuid,gid=0,noforcegid,addr=,unix,posixpaths,serverino,mapposix,acl,rsize=1048576,wsize=65536,echo_interval=60,actimeo=1 0 0
      32. tmpfs /run/user/108 tmpfs rw,nosuid,nodev,relatime,size=1630124k,mode=700,uid=108,gid=114 0 0
      33. tmpfs /run/user/1000 tmpfs rw,nosuid,nodev,relatime,size=1630124k,mode=700,uid=1000,gid=1000 0 0
      34. gvfsd-fuse /run/user/1000/gvfs fuse.gvfsd-fuse rw,nosuid,nodev,relatime,user_id=1000,group_id=1000 0 0
      Display All
      I'm back, here's the full output of cat /etc/mtab as requested.

      Needless to say, NAS shares are shown at line 30 and 31

      Anything strange?
    • I did some extra testing during the weekend: tried booting from a live Xubuntu 16 USB (64 bit) and manually mounting the NAS share, and also installing an empty 2.5" HD with a dedicated share, just to be sure there were no issues with the previously installed drives. I also replaced the whole gigabit chain by inserting a gigabit switch before the modem/router with a different pair of ethernet cables. Still getting the same results so I reverted back to to original configuration.


      It seems to me there is a correlation (again, I may be missing something obvious) between stuff that is actually being copied and the resulting speed.

      When I try to trasfer mixed sized files (a mixture of 3-4 MB jpeg and several hundreds MB sized mp4 and/or tar.gz archives, for a grand total of about 2-3 GBs) average write speed is considerably higher, while transfering a single mp4 of the same size (i.e. 2-3 GBs) goes back to 15-20 MB/s.

      Is this expected behaviour?
      If so, I honestly hadn't noticed before. :-\

      PS: I also get some serious issues (basically, video constantly hangs and reproduction turns out being impossible) while trying to play the aforementioned 2-3 GB mp4 HD videos from the NAS (via VLC), if that rings a bell...