<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=ISO-8859-1">
  </head>
  <body bgcolor="#ffffff" text="#000000">
    Hello XFS team,<br>
    <br>
    I have run into trouble with XFS, but excuse me if this question has
    been asked a dozens times.<br>
    <br>
    I'm am filling a very big file on a XFS filesystem on Linux that
    stands on a software RAID 0. Performance are very good until I get 2
    "holes" during which my write stalls for a few seconds.<br>
    Mkfs parameters:<br>
    mkxfs.xfs -b size 4096 -s size 4096 -d agcount=2 -i size=2048<br>
    The RAID0 is done a 2 SATA disks of 500 GB each.<br>
    <br>
    My test is just running "dd" with 8M blocks:<br>
    dd if=/dev/zero of=/DATA/big<br>
    (/DATA is the XFS file system)<br>
    <br>
    The system is basically a RHEL5 with a 2.6.18 kernel and XFS
    packages from CentOS.<br>
    <br>
    The problem happens 2 times: one time around 210 GB and the second
    time around 688 GB (hole in performance and response time is bigger
    the second time -- around 20 seconds)<br>
    <br>
    Do you have any clue ? Do my mkfs parameters make sense ? The goal
    here is really to have something that is able to store big files at
    a constant throughput -- the test is done on purpose.<br>
    <br>
    <div class="moz-signature">-- <br>
      <font color="blue"><b>Mathieu Avila</b><br>
        IT &amp; Integration Engineer<br>
      </font>
      <a class="moz-txt-link-abbreviated" href="mailto:mathieu.avila@opencubetech.com">mathieu.avila@opencubetech.com</a><br>
      <br>
      OpenCube Technologies <a href="http://www.opencubetech.com">http://www.opencubetech.com</a><br>
      Parc Technologique du Canal, 9 avenue de l'Europe<br>
      31520 Ramonville St Agne - FRANCE<br>
      Tel. : +33 (0) 561 285 606 - Fax : +33 (0) 561 285 635<br>
    </div>
  </body>
</html>