On Thu, Jul 26, 2012 at 08:46:21PM -0700, Chad wrote: > > Thanks for the encouragement. I agree. It's sort of also hard to believe > that Ruote can generate so much SQL traffic that an Amazon RDS instance > shuts down the connection.
Hello, that's a new piece of information. So far the only concrete thing that came out of the thread was that some connections were lost because of the forking. How do you reach the conclusion that RDS shuts down the connection? Are we sure it's a corner case of the forking problem? > It seems, though, that Ruote is not happy after it encounters such an > error, since it accepts new jobs, but doesn't run them. :( I did some tests (with the mysql2 driver), they look like this: https://gist.github.com/3186470 sequel reconnects each time easily. > Tonight, we are going to work with DevOps, to see what's going on. > Meanwhile, the Redis Storage is not an option for us, on our platform, but > Dynamo DB is. Until we can figure out, with our DevOps teams, why RDS > would respond as it does, I'm probably going to write a proof of concept > Dynamo storage. Our DevOps team is sleeping, so not much I can do in the > meantime. So how does RDS respond? Anything in its log? If you want me to be useful you have to tell me all those details that matter (for example, you guys told me about Passenger and RDS in message 16/25 of this thread ;-) ) > By the way, did you get my email that I sent to your gmail account last > night? Hello, yes, thanks a lot. You guys'll owe me a dinner after one year of stability in production ;-) I'd be glad to work on a sequel storage that polls less, but that's not doable as a 3 hours hack. Best regards, -- John Mettraux - http://lambda.io/jmettraux -- you received this message because you are subscribed to the "ruote users" group. to post : send email to [email protected] to unsubscribe : send email to [email protected] more options : http://groups.google.com/group/openwferu-users?hl=en
