<div dir="ltr"><div class="gmail_extra"><br><div class="gmail_quote">On Tue, Mar 18, 2014 at 8:59 AM, Vick Khera <span dir="ltr"><<a href="mailto:vivek@khera.org" target="_blank">vivek@khera.org</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="gmail_extra">after locking the set and restarting the slon daemons, I waited about 4 minutes yet the cleanup never happened.</div>
<div class="gmail_extra"><br></div><div class="gmail_extra">
do i need to let the slons run with short cleanup set for a while before I try the upgrade? ie, clear out any 10 minute timeouts somewhere?</div></blockquote></div><br>It appears if I wait until there are 0 backlogged events, and I manually run cleanupevent('10 seconds'::interval) four times, both tables get truncated immediately. i will try this on my next maintenance interval on production.</div>
<div class="gmail_extra"><br></div><div class="gmail_extra">is something else preventing cleanupevent() from completing? on my produciton it looks as if the lock; cleanupevent(); commit sequence sits idle (not waiting on a lock) for a few minutes then executes and start over again idle for a while again, even with the interval of 10 seconds.</div>
<div class="gmail_extra"><br></div></div>