Sending crash id's to RabbitMQ

[grow...@gmail.com]

Hello,

Following the docs, I've got Socorro, ES and Kibana running and crashes being collected and processed.

It looks like I should also be able to have the crash id's sent to RabbitMQ?

While I've read the docs:

http://socorro-collector.readthedocs.io/en/latest/crashstorage_rabbitmq.html#crash-storage


I think I must be missing something, I'm not totally sure of the required process.

I've added: 'socorro.external.rabbitmq.crashstorage.RabbitMQCrashStorage' as a storage destination in my socorro-config/processor.conf but am getting no queue being generated, or any evidence of anything being sent/trying to be sent to it.

I've sent a test post to RabbitMQ via some Javascript to confirm that's setup Ok, and it all looks good - just nothing from Socorro.

Any help would be greatly appreciated,

Thanks,

mikey

[William Kahn-Greene]

Hi!

The docs link is for a separate project that never got finished. It's
for a stand-alone Socorro collector
(https://github.com/mozilla/socorro-collector) that was extracted from
the Socorro project (https://github.com/mozilla/socorro).

Are you really using the socorro-collector? Or are you using the
collector that comes with Socorro?

/will

> _______________________________________________
> tools-socorro mailing list
> tools-...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/tools-socorro

[grow...@gmail.com]

Aha, I did not realise that, thanks for the info, looking at the url, it's obvious now.

It's just the regular collector that comes with Socorro that I'm currently using.

Am I still correct in thinking I should be able to send the crash id's to RabbitMQ? The page (http://socorro.readthedocs.io/en/latest/configuring-socorro.html#distributed-socorro) seems to imply so, but I can't work out how to set this up.

Thanks for the fast response,

mikey

[William Kahn-Greene]

Yes--the collector's crashmover can send crash ids to a RabbitMQ queue.

You said you added
"socorro.external.rabbitmq.crashstorage.RabbitMQCrashStorage" as a
storage destination to the processor.conf file. Does that mean you're
trying to add a crash id to a queue post-processing?

The RabbitMQCrashStorage class (and it's configuration) is here:

https://github.com/mozilla/socorro/blob/master/socorro/external/rabbitmq/crashstorage.py#L27

You'll need to set up a connection context for it, too. It defaults to
using ConnectionContextPooled which is here:

https://github.com/mozilla/socorro/blob/master/socorro/external/rabbitmq/connection_context.py#L259

That subclasses ConnectionContext which has a bunch of configuration here:

https://github.com/mozilla/socorro/blob/master/socorro/external/rabbitmq/connection_context.py#L74


Does any of that help? Does it produce anything in the logs like an
error or some other indicator that it successfully or unsuccessfully
added a crash id to a queue?

/will