| Publisher throughput | unlimited | 10,000,000 kB per 100 seconds (100 MB/s) |
| Subscriber throughput | unlimited | 20,000,000 kB per 100 seconds (200 MB/s) |
| Administrative operations | 10,000 per 100 seconds (100 ops/s) | 1,000 per 100 seconds (10 ops/s) |
Sharing the solution with the broader discussion list:
My theory is that you are using the gcloud.pubsub client library with default application credentials.
The implementation there makes OAuth super simple at the expense of using shared quota. It's actually using a project id that's different from yours to authenticate & get quota.
If I'm right, you will fix the spikes by creating and downloading service account credentials and forcing the client library to use those for the connection.
In python, this looks like:
c = pubsub.Client(o.project).from_service_account_json('/Users/kir/Downloads/kir-learns-cloud-931199eea5d9.json')
topic = c.topic(o.topic)
print "Using client_id %s"%c.connection.credentials.client_id
In node, you can probably do something like:
var gcloud = require('gcloud')({
projectId : config.get('$YOUR_PROJECT_ID'),
keyFilename : config.get('/Users/kir/Downloads/kir-learns-cloud-931199eea5d9.json')
});
var pubsub = gcloud.pubsub();
var topic = pubsub.topic(topicName);
Show quoted text
--
Kir Titievsky | Product Manager | Google Cloud Pub/Sub
--
You received this message because you are subscribed to the Google Groups "Google Cloud Pub/Sub Discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cloud-pubsub-discuss+unsub...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cloud-pubsub-discuss/b7d63374-9fdf-46d3-b273-e2b0f522a4dd%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.