Назад ко всем обсуждениям

Socker refusal error?

Rapid account: Daled 12
daled12
7年前

Hi, we resolved the previous issue, it was our coding issue. Thanks.

But now we are experiencing a different issue. We are successfully making API calls, but about half the time they fail, but we still get charged for a call. We see this response:

[4/13/17, 11:55:21 PM] Manish Bhuva: Unirest\Response Object
(
[code] => 200
[raw_body] => {“status”: “unknown”, “error”: “lookup failed due to socket refusal (timed out)”}
[body] => stdClass Object
(
[status] => unknown
[error] => lookup failed due to socket refusal (timed out)
)

[headers] => Array
    (
        [0] => HTTP/1.1 200 OK
        [Content-Type] => application/json
        [Date] => Thu, 13 Apr 2017 16:45:38 GMT
        [Server] => Mashape/5.0.6
        [Via] => 1.1 008ae64ab7020a9aecc4c202669805d4.cloudfront.net (CloudFront)
        [X-Amz-Cf-Id] => 08NkIlqG8e3FrIMfJBE0vTGcS8F5xaDjHAYuL1Ig_dMT_qFRM9z6uw==
        [x-amzn-RequestId] => 9c9027d9-2068-11e7-ad1b-c920b44d75f6
        [X-Amzn-Trace-Id] => Root=1-58efab2d-afe04500e26c0f76daad080f
        [X-Cache] => Miss from cloudfront
        [X-RateLimit-Queries-Limit] => 50000
        [X-RateLimit-Queries-Remaining] => 44292
        [Content-Length] => 81
        [Connection] => keep-alive
    )

)
[4/13/17, 11:57:00 PM] Manish Bhuva: ==================
[status] => unknown
[error] => lookup failed due to socket refusal (timed out)

Is there something we can fix on our end? Thanks.

Rapid account: Nametoolkit
nametoolkit Commented 7年前

@daled12 The .ba TLD doesn’t seem to run a standard whois lookup service. They have a website which returns images with the information. We probably won’t support that any time soon unfortunately.

Rapid account: Daled 12
daled12 Commented 7年前

Ah I see, that explains most of our failures. Thanks. But what about this domain that failed? znanost.ba

Rapid account: Nametoolkit
nametoolkit Commented 7年前

@daled12 noted. As a quick fix for yourselves, if you omit the subdomain, it will work. whois only works for second level domains e.g. cnet.com as there is no way to separately own a third/fourth/fifth level domain.
I’d prefer to throw a validation error with an explanatory message than try to transform the input in future.

Rapid account: Daled 12
daled12 Commented 7年前

Thanks, but we are still having some issues. For example, we sent this domain: download.cnet.com
And we got null values. Here is the header response:

[headers] => Array
(
[0] => HTTP/1.1 200 OK
[Content-Type] => application/json
[Date] => Wed, 19 Apr 2017 06:05:01 GMT
[Server] => Mashape/5.0.6
[Via] => 1.1 aa9a6b87feabe1a30d21428a24c1a7d8.cloudfront.net (CloudFront)
[X-Amz-Cf-Id] => xiKBMwl-WcYHNw4WNj70jU229LLqPVm6onnCVfAtohVt9v0doBY9bA==
[x-amzn-RequestId] => 1f4a591d-24c6-11e7-8cde-19d12e7b1fc6
[X-Amzn-Trace-Id] => Root=1-58f6fe0c-08f1544819720831874a66ef
[X-Cache] => Miss from cloudfront
[X-RateLimit-Queries-Limit] => 50000
[X-RateLimit-Queries-Remaining] => 43692
[Content-Length] => 2905
[Connection] => keep-alive
)

Thanks.

Rapid account: Slick Net Aaron
SlickNetAaron Commented 7年前

Many thanks! We are back in business.

Rapid account: Nametoolkit
nametoolkit Commented 7年前

Hi guys,
Apologies, I believe the issue is resolved.

Regards,
Mark

Rapid account: Slick Net Aaron
SlickNetAaron Commented 7年前

We are also seeing similar issues… this time it says Error from cloudfront:

HTTP/1.1 502 Bad Gateway
Content-Type: application/json
Date: Mon, 17 Apr 2017 22:05:19 GMT
Server: Mashape/5.0.6
Via: 1.1 fd885dc16612d4e9d70f328fd0542052.cloudfront.net (CloudFront)
X-Amz-Cf-Id: w8NgT2Q_NlXvU9UPYNezIaT5NBMxJ-a-RraK_k2wCGHtH6b3Xi6QPw==
x-amzn-RequestId: f16c7c5e-23b9-11e7-b024-e501c127ff59
X-Cache: Error from cloudfront
X-RateLimit-Queries-Limit: 1000
X-RateLimit-Queries-Remaining: 1000
Content-Length: 36
Connection: keep-alive

Rapid account: Daled 12
daled12 Commented 7年前

Hi, do you mean what TLD in terms of the getting api info on that TLD? In that case, no, we call a couple thousand different tlds everyday, and they change everyday. If you mean, what is my website’s TLD? that is scanurl.net.

I’m not sure we really use the RAW data field, although it might help in debugging when things fail.

Thanks.

Rapid account: Nametoolkit
nametoolkit Commented 7年前

Hi Andy,
Do you know which TLD was giving you the most issues? That would help somewhat. Currently retrying would not necessarily help. I’ll see if I can get something working in the near future to reduce these errors. Thanks.

PS Do you get any value from the ‘raw’ field in the response? I am thinking of deprecating it as it would reduce the response sizes and I think
most people don’t use it.

Rapid account: Daled 12
daled12 Commented 7年前

Hi,

Ok thanks for the fix. Don’t worry about the bill, its not much money.

My concern is, I’d like to make sure we get the data from the API call. Can you suggest a way that we can get the data in case of a 500? Do most people just retry again a 1 second later?

Also, do you know if there is a request per IP policy for this provider? Because we are not making too many request, about 1 per minute.

Thanks, Andy

Присоединяйтесь к обсуждению – добавьте комментарий ниже:

Войдите / Зарегистрируйтесь, чтобы публиковать новые комментарии