<div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote">2016-12-07 8:27 GMT+01:00 Peter Rathlev <span dir="ltr"><<a href="mailto:peter@rathlev.dk" target="_blank">peter@rathlev.dk</a>></span>:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">stores _everything_, including debug messages from "execute", you might<br>
want "Storage=volatile" there as well. You probably already have<br></blockquote><div><br></div><div>thanks, i missed this volatile thing<br> <span class="gmail-"></span><br><span class="gmail-"></span></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><span class="gmail-">
</span>What exactly does "slow down" mean here? Are you missing messages in<br>
the log files? Or are requests not answered in a timely fashion?<br></blockquote><div><br></div><div>"slow down" means an increment in the time consumed by bind to answer a query.<br></div><div>"heavy load" means about 20 millions query / day per machine, with peaks of about 400 query / second. <br>Using tcpdump i noticed an average increment of about 2% in the "Query time:" field of "dig", when using syslog instead of files, but I admit I only run a short and home-made benchmark<br></div><div>I also noticed different beaviour with different syslog daemons ( not using the last version ) : syslog-ng hangs named for 1-2 seconds, when using tcp and the remote syslog goes down ; rsyslog works more smoothly ; the old syslogd is unable to handle such a mass of messages<br></div></div><br></div><div class="gmail_extra">Regards<br><br></div><div class="gmail_extra">Ivan<br><br></div></div>