Bruce Krysiak

3.0.0.beta3 error is: NoMethodError: undefined method `add_remote_task' for #Paratrooper::Configuration:0x007fe202afa3c0

For what it's worth, my team is hitting the need for this capability as well. The scenario: - we have a known code defect - we don't have time to...

Again, you may not want to do this, just putting it out there since it was a bit of a pain when switching back and forth with Faraday.

In my initial tests, moving to Typhoeus (https://github.com/neo4jrb/neo4j-core/issues/272) seems to have solved my issues with `puma` and `preload_app!` as well.

Seeing the same on: RubyMine 2018.3 EAP Build #RM-183.4284.57, built on November 9, 2018 RubyMine EAP User Expiration date: December 9, 2018 JRE: 1.8.0_152-release-1343-b15 x86_64 JVM: OpenJDK 64-Bit Server VM...

I'd like to weigh in here that this functionality would be very helpful for my project too in case @gitstart you need any moral support 🤞🙂

I see this issue very reliably when using [neo4j-core](https://github.com/neo4jrb/neo4j-core) 6.1.5, [neo4j](https://github.com/neo4jrb/neo4j) 7.2.3 and [resque-pool](https://github.com/nevans/resque-pool) 0.6.0, with a worker pool of 4. In my Rails 4.2 project, I have a `resque.rake`...

For completeness, errors I see look like this (a bunch removed - almost all are the "too many bad responses" errors): ``` /Users/bruce/.rvm/gems/ruby-2.3.1@havn/gems/neo4j-core-6.1.5/lib/neo4j-server/cypher_response.rb:188:in `[]': no implicit conversion of Symbol into...

Auggghh you're right again. It looks like the various `Resque` hooks actually make it harder to do `mark_queued` and `mark_unqueued` properly since failures can happen in many different places, all...

I added a PR using the resque hooks, which I think is what you were going for - in https://github.com/neighborland/resque_solo/pull/10