thumb service: RestClient error in downloading ... Maximum number of redirect reached:

[ndushay]

Error occurring in production:

RestClient error in downloading memento text ... Maximum number of redirect reached:

https://swap.stanford.edu/20150412231254/http://www.albany.edu/sourcebook/
https://swap.stanford.edu/20151012231302/http://www.albany.edu/sourcebook/

E, [2017-01-21T03:30:08.217341 #12093] ERROR -- : Error in inserting memento #<Was::ThumbnailService::Synchronization::MementoDatabaseHandler:0x00000003d5cfe8 @uri_id=2, @memento_uri="https://swap.stanford.edu/20150412231254/http://www.albany.edu/sourcebook/", @memento_datetime="20150412231254"> into database.
RestClient error in downloading memento text from https://swap.stanford.edu/20150412231254id_/http://www.albany.edu/sourcebook/.
Maximum number of redirect reached: 

[ndushay]

here is a comment in #26

Here's an example of an 3xx error coming back from : https://swap.stanford.edu/20150412231254id_/http://www.noaa.gov/deepwaterhorizon/

and

https://swap.stanford.edu/20151012231258id_/http
://www.ustr.gov/about-us/press-office/reports-and-publications/

GET https://swap.stanford.edu/20151012231258/http://www.ustr.gov/about-us/press-office/reports-and-publications/
Status  HTTP/1.1 302 Moved Temporarily
Request
Accept  text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8
Accept-Encoding gzip, deflate, sdch, br
Accept-Language en-US,en;q=0.8
Cookie  _ga=GA1.2.1345946349.1467053808;
has_js=1;
__utma=154396474.1345946349.1467053808.1467760443.1467760443.1;
__utmc=154396474;
__utmz=154396474.1467760443.1.1.utmcsr=(direct)|utmccn=(direct)|utmcmd=(none);
JSESSIONID=6CD2D01EE499033E95CEB75268B9B5ED
Upgrade-Insecure-Requests   1
User-Agent  Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/51.0.2704.106 Safari/537.36
Response
Connection  Keep-Alive
Content-Length  36536
Content-Type    text/html;charset=utf-8
Date    Wed, 06 Jul 2016 20:16:04 GMT
Keep-Alive  timeout=5, max=100
Link    ; rel="original", ; rel="timemap"; type="application/link-format", ; rel="timegate", ; rel="first memento"; datetime="Sat, 12 Oct 2013 23:12:58 GMT", ; rel="prev memento"; datetime="Sun, 12 Apr 2015 23:13:07 GMT", ; rel="memento"; datetime="Mon, 12 Oct 2015 23:12:58 GMT", ; rel="next last memento"; datetime="Mon, 12 Oct 2015 23:15:36 GMT"
Location    https://swap.stanford.edu/20151012231258/https://ustr.gov/about-us/press-office/reports-and-publications/
Memento-Datetime    Mon, 12 Oct 2015 23:12:58 GMT
Server  Apache-Coyote/1.1
X-Archive-Guessed-Charset   cp1252
X-Archive-Orig-Connection   close
X-Archive-Orig-Content-Length   248
X-Archive-Orig-Content-Type text/html; charset=iso-8859-1
X-Archive-Orig-Date Mon, 12 Oct 2015 23:12:58 GMT
X-Archive-Orig-Location https://ustr.gov/about-us/press-office/reports-and-publications/
X-Archive-Orig-Server   Apache
X-UA-Compatible IE=edge

[ndushay]

Might it be fixes needed here to cope better with clean redirects? https://github.com/sul-dlss/was-thumbnail-service/blob/f8361a1c3d4e28c192d34ba18f0995b34a208de6/scripts/rasterize.js#L32-L34

Error from honeybadger occurs:

• https://github.com/sul-dlss/was-thumbnail-service/blob/master/lib/was/thumbnail_service/capture/capture_thumbnail.rb#L16-L21 which leads to
• https://github.com/sul-dlss/was-thumbnail-service/blob/master/lib/was/thumbnail_service/capture/capture_thumbnail.rb#L35
• which leads to scripts/rasterize.js
  • https://github.com/sul-dlss/was-thumbnail-service/blob/f8361a1c3d4e28c192d34ba18f0995b34a208de6/scripts/rasterize.js#L38

[ndushay]

rasterize.js seems to be using this: http://phantomjs.org/api/webpage/

and the code seems to be using vars from http://phantomjs.org/api/webpage/handler/on-resource-received.html

this is trapping for redirects and just incrementing a counter: https://github.com/sul-dlss/was-thumbnail-service/blob/f8361a1c3d4e28c192d34ba18f0995b34a208de6/scripts/rasterize.js#L32-L3

@jmartin-sul says

so i’m not sure what’s going on in the JS. outside of the JS, it queues a job for everything that doesn’t have a flag indicating a captured thumbnail. so if something dies trying to get a thumbnail, it’ll keep getting queued every time.

unclear:

• how it is called so many times (by delayed job? within the thumbnail_service code called by the delayed job?)
• what it means to increment required_id or whatever it is
• what happens where if it hits a redirect?

[jmartin-sul]

the above linked phantom js docs say in part:

The response metadata object contains these properties:

• id : the number of the requested resource

but it's still unclear to me what exactly that means. does id denote that the response corresponds to the Nth request that was made? if so, it'd seem to me that the loop would either run for one iteration, or for infinite iterations in the case of redirects, since the value that's checked keeps getting incremented. but like i said, unclear to me.

here's the code that will queue unsuccessful thumbnail jobs in perpetuity:
https://github.com/sul-dlss/was-thumbnail-service/blob/master/lib/was/thumbnail_service/capture/capture_manager.rb#L12
https://github.com/sul-dlss/was-thumbnail-service/blob/master/lib/was/thumbnail_service/capture/capture_thumbnail.rb#L29
https://github.com/sul-dlss/was-thumbnail-service/blob/master/lib/was/thumbnail_service/capture/capture_thumbnail.rb#L43-L46

if process_thumbnail never marks is_thumbnail_captured true, the job keeps getting requeued.