Thank you for your response. I did run the scheduler process as www-data
and it was able to open the file, however, new error on the database insert.
scheduler_run table shows status of TIMEOUT on the task, but the timeout is
set to zero, which should be unlimited?
This script takes about 5 minutes to execute due to the large number of
inserts (I have been playing around with different ways to insert to MySQL
but so far 5 minutes is the shortest I can manage).
Is there some timeout setting at the system level that is causing this
error?
*Traceback (most recent call last):*
* File "/home/www-data/web2py/gluon/scheduler.py", line 315, in executor*
* result = dumps(_function(*args, **vars))*
* File "applications/viewer/models/scheduler.py", line 139, in
wrapper_monitor*
* db3.IGN_wrapper.insert(log_ts = int(time.mktime(time.strptime(item[0],
'%Y/%m/%d %H:%M:%S'))), entry_type = item[1], entry = item[2])*
* File "/home/www-data/web2py/gluon/packages/dal/pydal/objects.py", line
726, in insert*
* ret = self._db._adapter.insert(self, self._listify(fields))*
* File "/home/www-data/web2py/gluon/packages/dal/pydal/adapters/base.py",
line 746, in insert*
* raise e Error: (<type 'exceptions.SystemExit'>, SystemExit(1,)) *
Once again, this part works just fine when executed from a controller.
On Thursday, May 26, 2016 at 6:24:13 AM UTC-7, Niphlod wrote:
>
> if your files are created by the www-data process and are not for anyone
> else to touch, the problem will rise.
> you can't expect an app running in the same permission env on www-data to
> behave (permission-wise) as one running on your user ("fivestar", I guess?)
>
> study your permission (start with 777, and acknowledge 777 is not safe) or
> start the scheduler process in the same security env (su - www-data -c ...)
>
> On Thursday, May 26, 2016 at 4:28:39 AM UTC+2, lamest.e...@gmail.com
> wrote:
>>
>> I have a simple application whose scheduler automatically backs up
>> rolling log files in a database.
>> There is an option in the controller to manually update whenever you want
>> (rather than waiting for the log to roll).
>> The scheduler task has to handle such cases by comparing two versions of
>> the log file to avoid duplication (the instance at the time of manual
>> update, and the instance at the time of roll-over)
>>
>> The source of the files looks like this
>>
>>
>> <https://lh3.googleusercontent.com/-pAVmfnjxJ9w/V0URGeqr1HI/AAAAAAAALlM/AbULjjyP9PY-G81IqXfFVyyCs0c5C3UOgCLcB/s1600/Untitled3.png>
>>
>> I'm seeing this error in the database logs for the failed scheduled task:
>>
>> *Traceback (most recent call last):*
>>
>> * File "/home/www-data/web2py/gluon/scheduler.py", line 315, in
>> executor*
>>
>> * result = dumps(_function(*args, **vars))*
>>
>> * File "applications/viewer/models/scheduler.py", line 127, in
>> wrapper_monitor*
>>
>> * old_file = open(os.path.join(request.folder, 'temp',
>> 'wrapper.log.old'), 'w+')*
>>
>> * IOError: [Errno 13] Permission denied:
>> 'applications/viewer/temp/wrapper.log.old'*
>>
>>
>>
>> I believe it is failing on line 121 because the log.new is not blank (
>> and the script fails again at 127 due to same permission error)
>>
>>
>>
>> <https://lh3.googleusercontent.com/-4HCvHNy96UA/V0UPwYOSeGI/AAAAAAAALlA/4K9DQaUiiMQg3G0EWHEhULIswMaOzi35QCLcB/s1600/Untitled.png>
>>
>> This is what is left over after the error (it would normally clean up
>> after itself, had it completed successfully)
>>
>>
>> <https://lh3.googleusercontent.com/-SC2n_c-mQk4/V0UQhO19KqI/AAAAAAAALlI/VvsinBYd_BAH4gPQtf5iA368Okyx5zdRACLcB/s1600/Untitled2.png>
>>
>>
>> So I restored the log files to a state that the scheduler is designed to
>> handle, copy/pasted the scheduler task logic into the controller, and
>> manually executed it (via url in browser), and everything works great.
>>
>> This is what I see afterwards (commented out the code that cleans up
>> after itself).
>>
>>
>> <https://lh3.googleusercontent.com/-Iv_doOSbVl8/V0UR0od4iLI/AAAAAAAALlc/2zxN8t18EqU6dh1m2JVncoY14gBtqPUoACLcB/s1600/Untitled4.png>
>>
>>
>> This seems like a user/permissions 101 issue to me but I am pretty new to
>> this whole Linux thing. I am running Linux Mint 17.3 and used the included
>> script for Ubuntu to deploy web2py.
>>
>>
>> To start the scheduler I am executing *nohup python web2py.py -K
>> <appname> &*
>>
>>
>> The resulting Python process shows up as my username. The apache service
>> shows up as www-data. The service that is generating the logs is also under
>> my username, if that matters.
>>
>>
>>
>> Any help appreciated in getting this thing to run in the model via
>> scheduler.
>>
>
--
Resources:
- http://web2py.com
- http://web2py.com/book (Documentation)
- http://github.com/web2py/web2py (Source code)
- https://code.google.com/p/web2py/issues/list (Report Issues)
---
You received this message because you are subscribed to the Google Groups
"web2py-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to web2py+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
