<br><div class="gmail_quote">On Thu, Jan 24, 2013 at 2:06 PM, Wesley Craig <span dir="ltr"><<a href="mailto:wescraig@columbia.edu" target="_blank">wescraig@columbia.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
You might find this helpful:<br>
<br>
<a href="http://lwn.net/Articles/328363/" target="_blank">http://lwn.net/Articles/328363/</a><br>
<br>
regarding fsync (lmtpd) latency on Linux. Lowering the lmtpd process limit ought to fix your load problem, but you should be aware how cyrus master implements that limit: your postfix processes will still make connections, but they will appear to hang when you're at the limit. This will transform your spike to a hill, but you may end up with lots of waiting postfix processes.<br>
<br>
</blockquote><div><br>Responding to the list where I'd like to hear more feedback on this.<br><br>We are running ext3 but could easily convert it to ext4. Has anyone<br>experienced that upgrade on their cyrus mail partitions? Does it<br>
improve latency enough to make a significant difference when IO is<br>saturated?<br><br>I have reduced the lmtpd max to 5. Now the load maxes at about 10.<br>It does take longer to work through the queue, but it eventually<br>
settles down (e.g. 10 minutes instead of 3 or 4)<br><br><br></div></div>