Hi, while running the CF 207 smoke and acceptance tests repeatedly, we noticed sporadic connection resets during 'cf login' (see log snippet from login log below). The connection reset is happening

+1 for password expiry; that'd be really handy to have. -- Regards, Daniel Jones EngineerBetter.com

Hello, thank you for the hint! I'd prefer to do something at application level, like a shell script to run in parallel to the application that every X seconds prints the output of netstat, as the

i may miss standup but I'll be there for IPM

fyi, we use UAA for our social login , take a look www.merv.id F -- Frans Thamura (曽志胜) Java Champion Shadow Master and Lead Investor Meruvian. Integrated Hypermedia Java Solution

1. For users created in UAA database, are there any policies we could apply regarding password expiry/strength of the password/lockout on repeated retry failures etc..? Currently there is a

Looking into using UAA and have couple of questions: 1. For users created in UAA database, are there any policies we could apply regarding password expiry/strength of the password/lockout on

Kim and James, thanks for responding and asking clarifying questions. Regarding latest, the end manifest ended up with lucid stemcell references. I changed them to trusty and then the deploy failed

Great writeup Dieu, thanks!

Hi Tom, The docs team was also not aware that this document isn't working. It looks like there have been various tweaks to this doc over time, including pretty recently. Do you have any more details

i was not aware that the documentation was not working for the latest cf-release versions. did you find that the documentation was out of date? -- Thank you, James Bayer

Yes, we recently ran some performance tests with the gorouter. You can find the results in this doc [1] and raw results and additional graphs [2] Related stories in tracker [3][4] [image: Inline

Is there any perf test data on the gorouter? Number of parallel connections possible given a specific VM size would help us in our sizing efforts. Thanks

Hi Dieu, Thanks for the response. Yep, that answers my question, and addresses some of the assumptions I'd been making. Good to hear the changes will be coming in v3. Have a good weekend! --

Thank you Dieu. This is great to know. John

You could set up a security group that logs all outbound connections. These are logged on the DEAs. You would then need to correlate the warden handle with the application. I'm working with the docs

Yes, we'd recommend using only the role methods on the organization resource as a general rule. Those roles are org specific so yes the association only exists to facillitate granting of org roles.

In cf-release v210 [1] we upgraded etcd to v2.0.1. We're currently running it with 3 nodes in our CI and Prod environments. We'll be looking at upgrading to v2.0.11 in the next couple of

I remember in the older version of CF release, CF was using an older version of etcd (v.0,4.x), which has issue running etcd in cluster mode from time to time, and there was a discussion that we would

For similar further diagnostics, I've updated docs with a working way to be able to invoke steno-prettify to get cc_ng logs as timestamps,