Bill Kayser's Profile

Recent activity

Subscribe to this feed

Bill Kayser replied on March 16, 2009 18:53 to the idea "Free Silver Account for Open Source Projects" in New Relic:

This is a great idea. If you know of any open source or non-profit projects that would be interested in a New Relic sponsorship have the directors contact us at info@newrelic.com.

Do you think this is one of the best points? Mark it as a good point
Bill Kayser replied on March 16, 2009 18:47 to the problem "StringIO Parameters Not Being Rewound" in New Relic:

Ian, sorry for the delayed response, however you will be happy to know that with the information you provided we were able to reproduce the problem and implement a fix. We took a more conservative approach, which is not to modify the StringIO object at all and omit the display of its contents in errors and transaction traces.

The fix goes into 2.8.6. It's available right now as a beta and we'd greatly appreciate it if you'd try it out.

http://newrelic.rubyforge.org/svn/bet...

Do you think this solves the problem? Mark it as solved
Bill Kayser replied on March 06, 2009 22:45 to the problem "NewRelic error when run under Glassfish 2.1" in New Relic:

I believe I have a fix for this. It's not released yet but if you want to try it out it is on a branch of our git repo:

http://github.com/newrelic/rpm/tree/m...

If you have a chance to try that, let me know if it works!

Do you think this solves the problem? Mark it as solved
Bill Kayser replied on March 06, 2009 19:09 to the idea "Support for Rails Metal" in New Relic:

It's an idea we're investigating. Can you describe how you would like it to appear in the UI? How would you identify the response times of requests handled with metal? How would you distinguish them?

Do you think this is one of the best points? Mark it as a good point
Bill Kayser replied on March 02, 2009 01:55 to the question "Can't start mongrels with newrelic 2.8.4" in New Relic:

Dev mode won't talk to the server. If it tried, you would see messages on startup. Also, any communication with the server is done on a background thread well after the plugin initialization completes.

If it happens again, change the log_level to debug and see if that helps. Also uncomment the puts in the newrelic init.rb. I'd also pepper the plugin with puts (in config.rb and agent.rb) to try to find out where it is hanging.

If you learn any more about this can you please follow up with me at support@newrelic.com?

Thanks,
Bill

Do you think this answers the question? Mark it as answered
Bill Kayser replied on March 02, 2009 00:20 to the question "Can't start mongrels with newrelic 2.8.4" in New Relic:

Do you know what version you upgraded from?

This is the first report we've had of anything like this. I'm unable to reproduce it locally with the configuration you specified. The only thing that could tie together newrelic initialization, developer mode and mongrel is the addition of the /newrelic route.

To help isolate the cause, can you try the following?

In the file lib/new_relic/config/rails.rb, disable the method install_developer_mode by putting a return statement at the top of the method.

Does this resolve the problem?

Do you think this answers the question? Mark it as answered
Bill Kayser replied on February 27, 2009 22:47 to the problem "Stats rubber banding from minute to minute" in New Relic:

Joseph I don't have an explanation why this is happening but I'm certain it is not related to RPM in any way. We're just the messenger. :-) If you have reason to suspect that the data is incorrect or misleading, let me know and I'll investigate further.

Do you think this solves the problem? Mark it as solved
Bill Kayser replied on February 27, 2009 19:44 to the problem "Stats rubber banding from minute to minute" in New Relic:

It looks like an issue with your load balancer. Look at the cluster breakout closely. The 460 and 462 hosts consistently get more load. 390 is consistently on the bottom.

At 8:35 your throughput was 266 rpm. With 20 mongrels, that's an average of 13 per mongrel. But in the cluster breakout, your 460 and 462 mongrels were up in the 20's. And the 5001 mongrel on 390 did not complete any requests during that minute.

Change the time period to 12 hours and look at the cluster breakout. I think that will tell you all you need to know.

Do you think this solves the problem? Mark it as solved
Bill Kayser set one of Bill Kayser's replies as an official response to "One account with two different subscription levels for different hosts... possible?" in New Relic
Bill Kayser replied on February 27, 2009 00:49 to the question "One account with two different subscription levels for different hosts... possible?" in New Relic:

Yes, you need to create a separate account.

http://newrelic.tenderapp.com/faqs/ge...

Do you think this answers the question? Mark it as answered
Bill Kayser replied on February 26, 2009 17:38 to the question "RPM not working with Rails 2.3 edge" in New Relic:

I'm afraid there's no progress to report. It's scheduled for the next major release of the agent 2.9.0, due in 4-6 weeks.

Do you think this answers the question? Mark it as answered
Bill Kayser replied on February 25, 2009 03:03 to the problem "How can we able to add services in profile page?" in chi.mp:

Yeah I have to admit I'm stumped too. Services Link????

Do you think this solves the problem? Mark it as solved
Bill Kayser replied on February 25, 2009 02:37 to the question "Named Scopes Support?" in New Relic:

I should also mention you get this detailed view in the developer mode (free) as well.

Do you think this answers the question? Mark it as answered
Bill Kayser replied on February 25, 2009 00:51 to the question "Named Scopes Support?" in New Relic:

I agree although it would pretty much explode the metric namespace for us. You do get finer granularity in the transaction traces (part of Silver) but we're unlikely to add this much detail to the metric space any time soon.

It's probably more likely we'll break out View metric into the partial names. Would likely be a much smaller metric space and we've had a lot of requests for that.

Do you think this answers the question? Mark it as answered
Bill Kayser replied on February 24, 2009 18:48 to the problem "newrelic rpm interferring with merb-mailer 1.0.9" in New Relic:

feydr, i'll get right on this. meanwhile can you send this message to support@newrelic.com so I can track it and keep you better informed?

Do you think this solves the problem? Mark it as solved
Bill Kayser set one of Bill Kayser's replies as an official response to "Named Scopes Support?" in New Relic
Bill Kayser replied on February 24, 2009 00:06 to the question "Named Scopes Support?" in New Relic:

Aaron, sorry for the delayed reply.

Named Scopes will surface as Model#find metrics.

It's a good suggestion to call them out separately. We'll definitely take that under consideration.

Do you think this answers the question? Mark it as answered
Bill Kayser set one of Bill Kayser's replies as an official response to "uninitialized constant CommandError" in New Relic
Bill Kayser replied on February 22, 2009 19:37 to the problem "uninitialized constant CommandError" in New Relic:

Ed I found the problem. What's probably happening is that there is an error creating the deployment, like it can't retrieve the git revision or some permissions aren't in place. But this triggers a subsequent error.

I'll turn around a fix in a couple days, but in the meantime, can you change "CommandError" in the recipes.rb to "StandardError" and see what the result is? It probably won't submit the deployment, but it shouldn't break capistrano and you should see an indication of the real reason it's failing.

Can you follow-up to support@newrelic.com so I can keep you better informed?

Do you think this solves the problem? Mark it as solved
Bill Kayser set one of Bill Kayser's replies as an official response to "Error loading data for longer timespans" in New Relic

Bill Kayser's dashboard

I'm an employee of:

About me

Products & companies I use

Recent activity