Jake Moshenko
21cb9f1aa1
Handle null executor cancellations separately from other exceptions
2017-05-15 13:45:44 -04:00
Jake Moshenko
ba07270bb2
Turn off in-app sentry logging, only log 500s at the WSGI layer
2017-04-18 16:38:22 -04:00
Jake Moshenko
22f5934f34
Add error logging to Marketo calls
2017-04-17 10:19:52 -04:00
Joseph Schorr
c9bb132339
Increase cloudwatch send timeout to reduce how often we hit the API
2017-02-01 13:09:00 -05:00
EvB
a7122db250
fix(cloudwatch): randomize sleep interval
2017-01-05 11:41:12 -05:00
Joseph Schorr
1a61ef4e04
Report the user's name and company to Marketo
...
Also fixes the API to report the other changes (username and email) as well
2016-11-14 17:34:50 -05:00
Jake Moshenko
f04b018805
Write our users to Marketo as leads.
2016-10-14 16:29:11 -04:00
Jake Moshenko
013e27f7d5
Clean up mixpanel analytics a bit.
2016-10-13 15:03:04 -04:00
Joseph Schorr
a1009af61c
Move aggregator into its own repo and add it to the image
2016-07-05 15:39:51 -04:00
Joseph Schorr
713ba3abaf
Further updates to the Prometheus client code
2016-07-01 14:16:51 -04:00
Jake Moshenko
668a8edc50
Refactor prometheus integration
...
Move prometheus to SaaS and make it a plugin
Move static callers to use metrics_queue plugin
Change local-docker to support different quay clone dirnames
Change prom_aggregator to use logrus
2016-07-01 14:16:50 -04:00
Matt Jibson
3d9acf2fff
Use prometheus as a metric backend
...
This entails writing a metric aggregation program since each worker has its
own memory, and thus own metrics because of python gunicorn. The python
client is a simple wrapper that makes web requests to it.
2016-07-01 14:16:50 -04:00
Joseph Schorr
63a8b197e4
Break out 5XX errors into their own metric
...
First part of #983
2015-12-16 13:56:07 -05:00
Matt Jibson
4aa5ab88dd
Use real cloudwatch limit
...
Although cloudwatch allows 40KB of data, it may be from no more than 20
different metrics. Until we can do that, limit the total points to 20.
2015-08-26 16:48:48 -04:00
Joseph Schorr
4625ecf273
Fix tests in response to breakage in #351
2015-08-17 16:26:20 -04:00
Matt Jibson
9a7e5bb35e
Batch cloudwatch puts
2015-08-17 12:03:49 -04:00
Matt Jibson
7c3b555ee9
Code review
2015-08-12 16:31:01 -04:00
Matt Jibson
f043bc1379
Don't enable the metric queue if there's no Cloudwatch
2015-08-12 15:14:09 -04:00
Matt Jibson
b483209862
Wrap API and registry requests with common metric timings
...
Record response times, codes, and rollup non-2XX responses.
2015-08-12 12:16:00 -04:00
Matt Jibson
b04c190ca0
Prevent the metric queue from growing unbounded
2015-08-12 12:16:00 -04:00
Matt Jibson
cfb6e884f2
Refactor metric collection
...
This change adds a generic queue onto which metrics can be pushed. A
separate module removes metrics from the queue and adds them to Cloudwatch.
Since these are now separate ideas, we can easily change the consumer from
Cloudwatch to anything else.
This change maintains near feature parity (the only change is there is now
just one queue instead of two - not a big deal).
2015-08-12 12:15:52 -04:00
Jake Moshenko
18100be481
Refactor the util directory to use subpackages.
2015-08-03 16:04:19 -04:00
