readthedocs / readthedocs.org

The source code that powers readthedocs.org
https://readthedocs.org/
MIT License
8.04k stars 3.58k forks source link

build "Unexpected Error Occurred"; status information? #2852

Closed jantman closed 7 years ago

jantman commented 7 years ago

Greetings,

I tried running two builds this morning ( https://readthedocs.org/projects/biweeklybudget/builds/5438984/ and https://readthedocs.org/projects/biweeklybudget/builds/5439003/ ) as I experienced some failures in my TravisCI jobs that test doc builds, and wanted to see if they were happening on RTD as well. Both builds failed after 0 seconds with "Error: An unexpected error occurred".

Not sure if this is an intermittent error or something specific to my project. I figure I'll just try again in a while.

However, that brought up a question for me... is there any sort of RTD system status page/info that I could use to try and determine if this is an issue that's specific to me, or some systemic problem? If not, I'd be happy to discuss helping with such an effort offline...

Thanks, Jason

molpopgen commented 7 years ago

This has been happening for several days now. See issue #2849, for example. Still no word as to why.

robbmcleod commented 7 years ago

Same issue for myself with NumExpr:

https://readthedocs.org/projects/numexpr/builds/5446157/

rolfedh commented 7 years ago

Same issue for my project: Consistent "Failed: Unexpected Error Occurred", with an occasional "Passed". This sounds a lot like github.com/rtfd/readthedocs.org/issues/2591

When reporting a bug, please include as much information as possible that will help us solve this issue. This includes:

Project name: supernet-documentation URL: https://readthedocs.org/projects/supernet-documentation/builds/ Action taken: Reviewed documentation and stackexchange topics. Tossed and recreated my project (plain vanilla, default settings), and my conf.py. Modified my conf.py. Built it locally without any errors or warnings. Joined the readthedocs "Gold" program. Expected result: Expected it to build on readthedocs. Actual result: "Unexpected Error Occurred"

This build failure also happens with your readthedocs demo project: https://readthedocs.org/projects/rolfedh-demo/builds/

This makes me believe this is a software bug, not a project misconfiguration.

rolfedh commented 7 years ago

It seems like those "Unexpected errors" have gone away and everything is building normally now. How about for you, @jantman @molpopgen @robbmcleod ?

robbmcleod commented 7 years ago

Yes it looks like I have documentation now, thanks for the heads up.

molpopgen commented 7 years ago

My builds are still failing. I'm guessing there's a possibly undocumented change to RTD. I have made no changes to my settings, and am getting a failure from docutils:

https://readthedocs.org/projects/fwdpy11/builds/5456959/

On Mon, May 22, 2017 at 2:13 PM Robert McLeod notifications@github.com wrote:

Yes it looks like I have documentation now, thanks for the heads up.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/rtfd/readthedocs.org/issues/2852#issuecomment-303219419, or mute the thread https://github.com/notifications/unsubscribe-auth/AGHnH7gi8nAWbKmAJHVCeUd-D8VKCEknks5r8fgfgaJpZM4NfByQ .

-- Kevin Thornton Associate Professor Ecology and Evolutionary Biology UC Irvine http://www.molpopgen.org http://github.com/ThorntonLab http://guthub.com/molpopgen

rolfedh commented 7 years ago

@molpopgen, You are getting a long list of warnings/error messages. After researching/troubleshooting those messages, you should probably create a new separate issue. This issue is different from yours. It's about users getting an abrupt "Error: An unexpected error occurred". If this issue has been resolved, yours may get closed along with it.

molpopgen commented 7 years ago

Yes, those warnings are known and (I believe) are not causing the error. They're the issues that sphinx/breathe has with parsing doxygen's XML output for C++11. Everything "just works" on several of my own systems with the same or newer versions of the packages that RTD uses.

I'll try suppressing the creation of doxygen output and then file a new issue.

On Tue, May 23, 2017 at 9:06 AM Rolfe Dlugy-Hegwer notifications@github.com wrote:

@molpopgen https://github.com/molpopgen, You are getting a long list of warnings/error messages. After researching/troubleshooting those messages, you should probably create a new separate issue. This issue is different from yours. It's about users getting an abrupt "Error: An unexpected error occurred". If this issue has been resolved, yours may get closed along with it.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/rtfd/readthedocs.org/issues/2852#issuecomment-303446806, or mute the thread https://github.com/notifications/unsubscribe-auth/AGHnH6wj82TkYMCpnchg78wLPQw9Y2Pqks5r8wK3gaJpZM4NfByQ .

-- Kevin Thornton Associate Professor Ecology and Evolutionary Biology UC Irvine http://www.molpopgen.org http://github.com/ThorntonLab http://guthub.com/molpopgen

agjohnson commented 7 years ago

The primary issue in question, mentioned by OP, looked related to a server issue that I cleared up today. Projects receiving a blank failure of "An unexpected error occurred" were likely hitting this issue.

@molpopgen You are indeed hitting a separate issue. The error is coming from docutils ultimately, and as you are building with Sphinx 1.6.1, you might be hitting incompatibility with the versions of Sphinx and doxygen. We install the version of doxygen installed with Ubuntu 16.04, if you can test this version locally with your version of Sphinx (currently 1.6.1), we can determine if there is a fix on our side. Feel free to open up a separate issue for this.

@jantman It looks like your builds are working as expected now, so I'll close this issue. To answer your second question, no we don't have a status page. As much as we'd like to have something like this, we don't have the capacity to justify spending time on this currently. Perhaps in the future we will, however.

I'm going to lock this issue, lest it attract all the other generic failures. Feel free to open a new issue if you are noticing something similar.