Charlie runs out of memory a lot, may explain flakiness

[mogul]

Now that we’ve got logs.cloud.gov up, we can see that Charlie logs errors 3-4x more than any other app on cloud.gov. It looks like it’s due to running out of memory a lot, and I guess we’ve seen that it’s flaky in Slack sometimes… @dlapiduz said

it is redis
we need to move to use our redis service
rather than IBMs
(yes that is happening)

I'm not sure who looks after that, but in any case I'm capturing that there's a problem to be solved here per conversation in #bots.

[mgwalker]

Not sure if this is still an issue. Charlie seems to have been pretty stable for a while now. I want to add a startup response where it posts into #bots so we can see if it's still restarting frequently and we just don't see it anymore. If we get a few weeks without anything, maybe we can close this one?

[afeld]

Sounds good. We can reopen if needed. cloudfoundry/nodejs-buildpack#82 may help, too.