Hello -
Recently our live dbxml has started segfaulting when creating a query context, but its not necessarily reproducible because it randomly happens (sometimes every 5 seconds, sometimes every 3 hours). I didn't think it would segfault in this case, but does this sound like something that could be a result of not enough locks/lockers/etc (although I usually think the proper error is thrown in that case)? Does this sound like something the 2.4.16 upgrade addresses?
Using 2.4.13 with python wrappers. I have single writer multiple readers set up. If it helps at all, below is the query that is
usually compiled before the segfault (I would say that 1 in 6 of these queries segfaults). I see it most often when load is high on the system.
bq. <address>{color:#003366}querystr = r'for $o in collection("rx.dbxml")/rx order by $o/@dos descending return <item id="{ replace(dbxml:metadata("dbxml:name", $o),""", "_") }"> {$o} </item>' % janusId \\ qc = mgr.createQueryContext() \\ {color}</address>
Thanks in advance!
eleddy