<br><br><div class="gmail_quote">On Sat, Feb 5, 2011 at 4:45 AM, ionral <span dir="ltr"><<a href="mailto:daniele.nuzzo@email.it">daniele.nuzzo@email.it</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
if I check the status of drbd the following response is:<br>
<br>
0: cs: Connected ro: Primary / Primary ds: UpToDate / UpToDate C r ----<br>
ns: 640864680 nr: 135234784 dw: 776099464 dr: 1520599328 al: 836478 bm:<br>
2185 lo: 0 pe: 0 ua: 0 ap: 0 ep: 1 wo: b OOS: 0<br>
<br>
I do not know if it is positive that the method is to write after is Barrier<br>
.... reading the manuals of drbd I noticed that in these cases (BBU)<br>
performance would be better for a wo:n<br>
<br>
What do you think?<br></blockquote><div><br></div><div>Modify your drbd.conf to look more like:</div><div><br></div><div> disk {</div><div> on-io-error detach;</div><div> # It is good to have backed up write caches</div>
<div> no-disk-barrier;</div><div> no-disk-flushes;</div><div> no-disk-drain;</div><div> no-md-flushes;</div><div> use-bmbv;</div><div> }</div><div> </div><div>This will completely turn off write ordering which. I have measured performance gains doing this with backed-up write caches.</div>
<div><br></div><div>-JR</div></div>