Hi,
Simliar to https://www.pythonanywhere.com/forums/topic/2046/, I too am seeing multiple IOError's in the log.
The previous forum answer suggested a user was interrupting the page load causing the exception, however I do not see any activity in the web access log to suggest the site was being accessed at this time (see logs below).
I have a (probably) unrelated issue where some records are not being saved in the sqlite database on an infrequent basis - these are the only errors I cannot explain in the log file, hence my interest.
Am unfamiliar with exactly how user processes get moved between machines, but wondering if when a process gets moved open file handles (to logs and sqlite) could become stale raising the IOError below.
If anyone has suggestions on how to improve the logging to be able to see exactly what operation is causing the IOError to be raised that would be helpful.
Thanks,
Simon.
<-- error.log ---> 2016-06-12 21:57:25,283 :IOError: write error 2016-06-12 21:57:25,293 :Error running WSGI application Traceback (most recent call last): File "/bin/user_wsgi_wrapper.py", line 156, in call yield response GeneratorExit
<-- access.log over same time period -->
31.131.247.74 - - [12/Jun/2016:18:26:31 +0000] "GET /logout HTTP/1.1" 302 5 "http://sashdown.pythonanywhere.com/commitment" "Mozilla/5.0 (Linux; Android 6.0.1; SM-G935F Build/MMB29K) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/51.0.2704.81 Mobile Safari/537.36" "31.131.247.74" 90.194.96.218 - - [13/Jun/2016:07:34:41 +0000]54.227.2.20 - - [13/Jun/2016:11:37:32 +0000] "GET / HTTP/1.1" 404 100 "-" "python-requests/2.7.0 CPython/2.7.6 Linux/3.13.0-85-generic" "54.227.2.20"
sashdown
|
2
posts
|
