Been out all day, while keeping this issue in mind. I came across a "what if" What happends if I create a small log manually? Done that and the results were good. Tested on a full day log and the "issue" was back. Tested old DBs and seems that the amount time to save VTs is proportional to size of the log. I think we can exclude the network from the equation. Can this be resumed as an hardware issue? I'm running RD on a dual-core AMD A6-5400K/4gb. RIPCD is on the top 3, mostly in 1st place, of the cpu list while running RD. Always
On Sat, Feb 22, 2014 at 11:16 AM, Cowboy <[email protected]> wrote: > On Friday 21 February 2014 09:15:02 pm Lorne Tyndale wrote: > > Another thing with MySQL, ensure that you've got hostname lookups turned > > off in my.cnf > > Yeah, I thought about that, too. > ( actually a better way than dinking with other things, like > file system flags, and resolver settings that won't fix the problem, > but merely hide it ) > > Except that none of that has been changed between > problem and no problem. > > Pedro indicated that restoring an older database, no problem. > Then, restoring the new database, problem. > > It seems that ideally, we could step back through time, restoring > daily backups one at a time, until we find the point at which > something changed. > Somehow, I doubt we have that many backups, so the best we > can do is try to determine what changes exist between problem > and no problem, and try to isolate those changes. > > Especially in RDAdmin, what's different ? > > -- > Cowboy > > http://cowboy.cwf1.com > > "Just once, I wish we would encounter an alien menace that wasn't > immune to bullets" > -- The Brigader, "Dr. Who" > > _______________________________________________ > Rivendell-dev mailing list > [email protected] > http://caspian.paravelsystems.com/mailman/listinfo/rivendell-dev >
_______________________________________________ Rivendell-dev mailing list [email protected] http://caspian.paravelsystems.com/mailman/listinfo/rivendell-dev
