Can't access Task Queues: "Failed to load"

[Andrea De Togni]

Hi

for an unknown reason, when I click "Task Queues" on developer console I see "Failed to load" and nothing else.

It's been a while since I don't login to check (the app was and is working perfectly fine) but I need to check the queues now.

Anyone can help?

THanks

[Nicholas (Google Cloud Support)]

Good day,

Unfortunately, 'Failed to load' messages provided by the Developers Console are not the most descriptive. Could you test the Task Queue view in the Developers Console from an Incognito Mode Chrome window or Firefox Private Browsing window? If you still get the same error, could you attach a HAR file containing the network requests that returned the 'Failed to load' error? I guide to do so can be found at the Google Apps Toolbox HAR Analyzer page. Often times when the on-screen error message is somewhat generic, a more descriptive one may be found within the HTTP responses received by your browser. This may be helpful to get a better idea of what the issue might be.

Cheers,

[Andrea De Togni]

Hi Nicolas

 

yeah, I understand such a message may be difficult to debug.

First of all I’ve tried both with Firefox (I normally use Chrome) and with Chrome in Incognito. Same error.

 

Please find attached the har file.

 

Thanks

 

Andrea De Togni

Head of Data and Technology
 

Office: +39 0230573837 Mobile: +39 3666370687

 

--
You received this message because you are subscribed to a topic in the Google Groups "Google App Engine" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/google-appengine/U_FLOa0W_4w/unsubscribe.
To unsubscribe from this group and all its topics, send an email to google-appengine+unsubscribe@googlegroups.com.
To post to this group, send email to google-appengine@googlegroups.com.
Visit this group at https://groups.google.com/group/google-appengine.
To view this discussion on the web visit https://groups.google.com/d/msgid/google-appengine/bd34a688-d446-48c9-8383-fb02144fa570%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Privileged/Confidential Information may be contained in this message. If you are
not the addressee indicated in this message (or responsible for delivery of the
message to such person), you may not copy or deliver this message to anyone. In
such case, you should destroy this message and kindly notify the sender by reply
email. Please advise immediately if you or your employer does not consent to email
for messages of this kind. Opinions, conclusions and other information in this
message that do not relate to the official business of Group M Worldwide LLC and/or
other members of the GroupM group of companies shall be understood as neither given
nor endorsed by it. GroupM is the global media investment management arm of WPP.
For more information on our business ethical standards and Corporate Responsibility
policies please refer to WPP's website at http://www.wpp.com/WPP/About/

[Nicholas (Google Cloud Support)]

Thank you for provided the logs I requested.

I have removed the HAR file from the thread as it contained project-specific information. Though, I did view it first to see the HTTP responses with errors. Unfortunately, it did not reveal much new information about the error message. I have not yet been able to reproduce this issue so any other details about this may be helpful.

1. Could you test if all owners, developers and viewers of you project get the same error?
   
2. Do you have any other projects experiencing this issue?
   
3. Can you add a second owner to this project and view these task queues while signed in to this account?
   
4. Though the Task Queue REST API is in Alpha, can you test a call to 'Taskqueues: get' or 'Tasks: list' using the API Explorer on the linked pages? If so, feel free to post the result here using mock project IDs, task queue names and URIs.

[Andrea De Togni]

Hi Nicholas

thank you first of all

1. There's just me...

2. Unfortunately this is the only project right now on App Engine

3. I've tested and added another owner. Same issue.

4. Not sure how to do. The API Explorer requests a "taskqueue" name, but I don't know what should I put there. However, here is the cron.yaml content (nothing "secret" here, just a list of tasks):

cron:

- description: DBM update cities

  url: /tasks/update_cities.php

  schedule: every day 6:20

  timezone: Europe/Rome

- description: DBM import logs (330345)

  url: /tasks/import_dbm.php?partnerid=330345

  schedule: every day 6:45

  timezone: Europe/Rome

- description: DBM import logs (300232)

  url: /tasks/import_dbm.php?partnerid=300232

  schedule: every day 6:46

  timezone: Europe/Rome

- description: DBM import logs (306134)

  url: /tasks/import_dbm.php?partnerid=306134

  schedule: every day 6:47

  timezone: Europe/Rome

- description: DBM import logs (310742)

  url: /tasks/import_dbm.php?partnerid=310742

  schedule: every day 6:48

  timezone: Europe/Rome

- description: DBM import logs (305636)

  url: /tasks/import_dbm.php?partnerid=305636

  schedule: every day 6:49

  timezone: Europe/Rome

- description: DBM import logs (300234)

  url: /tasks/import_dbm.php?partnerid=300234

  schedule: every day 6:50

  timezone: Europe/Rome

- description: DBM import logs (297350)

  url: /tasks/import_dbm.php?partnerid=297350

  schedule: every day 6:51

  timezone: Europe/Rome

- description: DBM import logs (300231)

  url: /tasks/import_dbm.php?partnerid=300231

  schedule: every day 6:52

  timezone: Europe/Rome

- description: DBM import logs (208845)

  url: /tasks/import_dbm.php?partnerid=208845

  schedule: every day 6:53

  timezone: Europe/Rome

- description: DBM import logs (297377)

  url: /tasks/import_dbm.php?partnerid=297377

  schedule: every day 6:54

  timezone: Europe/Rome

- description: DBM split by day

  url: /tasks/run_queries_splitbyday_dbm.php

  schedule: every day 7:10

  timezone: Europe/Rome

- description: Queries and sampling

  url: /tasks/run_queries.php

  schedule: every day 7:15 

  timezone: Europe/Rome

- description: Export results to storage

  url: /tasks/export_results.php

  schedule: every day 7:25

  timezone: Europe/Rome

- description: Aggregate results in storage 

  url: /tasks/aggregate_results.php

  schedule: every day 8:00

  timezone: Europe/Rome

- description: Set headers and public

  url: /tasks/add_headers_set_public.php?day=1

  schedule: every day 8:10

  timezone: Europe/Rome

- description: Set headers and public

  url: /tasks/add_headers_set_public.php?day=2

  schedule: every day 8:11

  timezone: Europe/Rome

- description: Set headers and public

  url: /tasks/add_headers_set_public.php?day=3

  schedule: every day 8:12

  timezone: Europe/Rome

- description: Set headers and public

  url: /tasks/add_headers_set_public.php?day=4

  schedule: every day 8:13

  timezone: Europe/Rome

- description: Set headers and public

  url: /tasks/add_headers_set_public.php?day=5

  schedule: every day 8:14

  timezone: Europe/Rome

- description: AN import files

  url: /tasks/apnx_getfiles.php

  schedule: every 45 minutes from 4:00 to 6:00

  timezone: Europe/Rome

- description: AN bigquery import

  url: /tasks/import_apnx.php

  schedule: every day 6:30

  timezone: Europe/Rome

- description: AN split by day

  url: /tasks/run_queries_splitbyday_apnx.php

  schedule: every day 7:00

  timezone: Europe/Rome

To unsubscribe from this group and all its topics, send an email to google-appeng...@googlegroups.com.
To post to this group, send email to google-...@googlegroups.com.

Visit this group at https://groups.google.com/group/google-appengine.
To view this discussion on the web visit https://groups.google.com/d/msgid/google-appengine/bd34a688-d446-48c9-8383-fb02144fa570%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Nicholas (Google Cloud Support)]

Thank you for providing the cron.yaml file containing 23 cron jobs. Upon inspecting your project (from the HAR file previously provided) at the time of this writing, I see there are no deployed App Engine modules. These are required for cron jobs to execute. 'Failed to load' is not very descriptive but may in fact be caused by the lack of deployed modules.

I'm uncertain at this point about the scope of these errors. I would like to assert whether the issue is regarding the viewing of cron jobs from the Developers Console (possibly because there are no deployed modules) or if it also regards the functioning of cron jobs entirely. If also the latter, please note the following important points found in Scheduling Tasks With Cron for PHP:

Free applications can have up to 20 scheduled tasks. Paid applications can have up to 100 scheduled tasks.

For a cron task to be considered successful it must return an HTTP status code between 200 and 299 (inclusive).

Given the cron.yaml you've post above, was this file in fact included when deploying your application with appcfg.py? If unsure, you may use update_cron like so to specifically upload said cron.yaml (--noisy flag included for debugging purposes):

appcfg.py --noisy update_cron your/app/directory

To confirm that this file has been uploaded and parsed, one can request the parsed version back using cron_info like so:

appcfg.py --noisy cron_info you/app/directory

Once confirmed present in your deployment, you can confirm if these cron jobs are executing by dedicated a given module to their URLs (not using the module for anything else) or by inspecting the HTTP request reaching a given module looking for the X-Appengine-Cron: true header as shown in Securing URLs for Cron. This requires a deployed module receiving requests to URLs as described by your cron.yaml.

If the above is entirely successful, then this issue should be isolated to your viewing of task queues from the Developers Console. Otherwise, please post the output of the above recommendations replacing sensitive information where necessary, your app.yaml stripped of application ID and dispatch.yaml if there is one.

[Andrea De Togni]

Hi Nicholas

it's probably worth mentioning that:

- I've developed this app around an year ago

- it's a paid app

- those cron job have always been are ARE CURRENTLY working perfectly fine. I know it because a) I receive execution logs via email each time they run - daily and b) the expected calculation and resulting outputs are working fine and visualized at the address www.mappsrealtime.com

And

- I have no idea what is an "App Engine Module". I was initially deploying the app (as a set of yaml, php and js files) through GIT (when app engine was auto-deploying through GIT) and lately I've been uploading through the command "appcfg.py update ./"

So I'm 100% sure that crons are working fine (in fact the entire app is working fine). It must be some kind of error when visualizing the jobs...

Thank you 

[Alex Martelli]

On Fri, Jan 15, 2016 at 2:01 PM, Andrea De Togni <andrea....@groupm.com> wrote:

   ...

- I have no idea what is an "App Engine Module".

See https://cloud.google.com/appengine/docs/php/modules/ .

Alex

[Andrea De Togni]

Hi Alex

thank you. As far as I remember, when I was developing this app, Modules were optional. After all, again, this has IS working without any module configured.

[Andrea De Togni]

Hey, I just got this email:

Dear App Engine Developer,

 

You are receiving this notice because you are running the Task Queues service with a custom domain on the following application(s):

 

s~appnexus-logfiles-groupm

 

Currently, we have a bug in the system where tasks enqueued from a custom domain (e.g. your_domain.com) are always routed to the default module even if a non-default module is specified in the app's dispatch.yaml/xml file. This bug will be fixed by February 22, 2016.

 

No action is required on your part if your application expects documented behavior (Java documented behavior).

 

However, if your application depends on the existing behavior of tasks being routed to the default module, then you must add an entry to your dispatch.yaml/xml before February 22, 2016 that routes your Task Queues endpoint to the default module. You can read more about dispatch.yaml/xml here. Example entry:

 

- url: "*/my/task/queue/url/*"

module: default

 

This is the expected behavior after the fix:

1.    Have an application with a domain (e.g. your_domain.com) mapped to it

2.    Configure two modules, and a dispatch rule to route all */api/* traffic to the non-default module.

3.    Add a task to the queue from the custom domain and the default module, with a URL of /api/taskqueue (e.g. your_domain.com/api/taskqueue)

4.    The task should be run on the non-default module, where */api/* is routed.

 

Please fill out this support form if you have questions.

 

Has this anything to do with my issue?

Thanks

[Nicholas (Google Cloud Support)]

This is not likely related. Nonetheless, please take the time to review your dispatch.yaml to ensure the proper modules will be received task queue requests as you intend once this change is made on February 22.

Regarding your current issue, our engineers have identified an issue with the system that fetches data regarding task queues. Please file a bug using our public issue tracker and star the issue the receive updates. If you then provide a link to the issue here, I will link it to the bug so that anyone that stars the issue will receive updates on the bug's progress.

[Josh Moore]

This should be resolved now.  Can you try accessing the Task Queues page again to confirm whether this is resolved for you?

--
You received this message because you are subscribed to the Google Groups "Google App Engine" group.
To unsubscribe from this group and stop receiving emails from it, send an email to google-appengi...@googlegroups.com.
To post to this group, send email to google-a...@googlegroups.com.

Visit this group at https://groups.google.com/group/google-appengine.

To view this discussion on the web visit https://groups.google.com/d/msgid/google-appengine/7bfc87ac-299e-4b7b-9b57-9409810d9493%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

Josh Moore |

Cloud Solutions Engineer |

josh...@google.com |

[Andrea De Togni]

Hi, YES it's now solved and working fine.

Thank you!