Unable to get up and running from the quick start about crawly HOT 13 CLOSED

Hey @Ziinc apparently your code works for me. Maybe you have some other details to share? Maybe I should try with different Elixir/Erlang versions?

from crawly.

Could you also try to do Crawly.fetch("https://www.erlang-solutions.com/blog.html") just to check if it's possible to make requests to the blog from your machine?

from crawly.

@Ziinc in any case tell me what you're trying to achieve so I could try to help.

from crawly.

I'm trying to integrate Crawly with an existing phoenix project.

My dependencies are as follows (though I doubt that there would be dependency conflicts:

defp deps do
    [
      {:phoenix, "~> 1.4.0"},
      {:phoenix_pubsub, "~> 1.1"},
      {:phoenix_ecto, "~> 4.0"},
      {:ecto_sql, "~> 3.0"},
      {:postgrex, ">= 0.0.0"},
      {:phoenix_html, "~> 2.11"},
      {:phoenix_live_reload, "~> 1.2", only: :dev},
      {:gettext, "~> 0.11"},
      {:jason, "~> 1.0"},
      {:plug_cowboy, "~> 2.0"},
      {:mix_test_watch, "~> 0.8", only: :dev, runtime: false},
      {:comeonin, "~> 4.1"},
      {:bcrypt_elixir, "~> 1.1"},
      {:distillery, "~> 2.0", runtime: false},
      {:httpoison, "~> 1.4"},
      {:ex_aws, "~> 2.0"},
      {:ex_aws_s3, "~> 2.0"},
      {:sweet_xml, "~> 0.6"},
      {:bureaucrat, "~> 0.2.5"},
      {:hound, "~> 1.0", only: [:dev, :test], runtime: false},
      {:mogrify, "~> 0.7.3"},
      {:honeydew, "~> 1.4.4"},
      {:crawly, "~> 0.5.0"}
    ]
  end

I was actually initially on 1.8.2 when I first encountered the issue, and I had updated to see if it would help.

This is what happens with the Crawly.fetch()

Interactive Elixir (1.9.1) - press Ctrl+C to exit (type h() ENTER for help)
iex(1)> Crawly.fetch("https://www.erlang-solutions.com/blog.html")
{:error,
 %HTTPoison.Error{id: nil, reason: {:option, :server_only, :honor_cipher_order}}}

I will try the quickstart on a fresh project and report back

from crawly.

@oltarasenko I've been able to get the quickstart to work in a new project, but it still does not work when i try add it into the existing project. I think it's an issue with dependency conflicts, notably httpoison.

from crawly.

@oltarasenko Seems like it is an issue with hackney:

https://elixirforum.com/t/hackney-error-option-server-only-honor-cipher-order/25541

I'll try recompiling and updating the deps and try again

from crawly.

Updating the httpoison dependency did the trick, it upgraded hackney to the latest version, where the ssl issue was fixed. Thanks!

from crawly.

I think instead of letting the httpoison errors get swallowed up, it would be good to let them surface as debug logs.

from crawly.

Ok, I am re-opening it, as the thing you have mentioned (https://github.com/oltarasenko/crawly/blob/master/lib/crawly/worker.ex#L43) requires a fix. Will process an error here & log the error message. Good catch!

from crawly.

@Ziinc Could you please have a glance at #15. It is pretty trivial for now. however, I plan to extend the worker quite soon (as I am currently working on a support of different user agents (aka webdriver support))

from crawly.

looks good. A possible extension (to add to backlog) for error behaviour could be a configurable fallback module to call when the backoff retries fail, thereby allowing the engine to exec some wrap up function, possibly to alert for errors.

from crawly.

A possible extension (to add to backlog) for error behaviour could be a configurable fallback module to call when the backoff retries fail, thereby allowing the engine to exec some wrap up function, possibly to alert for errors.

Possibly specified at either spider level or config level

from crawly.

This is now fixed in 0.6.0

from crawly.