catching those kind of errors is going to be a real PITA... every backend (i.e. database engine) logs the error as it wishes, so we don't have a "method" for pointing what is going wrong. No logs you reported included "error cleaning up": if you told me earlier, I would have pointed you in the right direction.
PS: every web2py part is kinda "required" to have read and write permission everywhere.... nobody really had any problem with it ^_^ On Tuesday, September 17, 2013 1:47:34 PM UTC+2, ajith c t wrote: > > Hi Niphlod, > Got my scheduler working correctly. > I updated my web2py version to 2.5.1 from 2.2.1 as you mentioned > earlier. After that started two upstart jobs, one for my app and another > for the scheduler. Still it returned an error "error cleaning up.". > I couldn't find an appropriate solution for it net so checked the > code of scheduler.py file in gluon folder. Found that an exception raises > this error at line number 797. The exception doesn't say anything about the > error, just prints the message. After calling the Exception class and > printing the exception, found out that it was a permission problem. I had > not given the DELETE permission to the user. Once that was corrected, > everything worked fine as I needed, > > It would be helpful if you could also log the exception that displays the > message from the scheduler.py file. I had to spend lot of time to find this > error, maybe my mistake as I am a web2py beginner, but it would really help > if the cause of the exception is also logged, > > On Thursday, 12 September 2013 12:36:33 UTC+5:30, ajith c t wrote: >> >> Thank you for the response, will try everything you said and post it if >> every thing goes correct >> >> On Thursday, 12 September 2013 00:34:31 UTC+5:30, Niphlod wrote: >>> >>> >>> >>> On Wednesday, September 11, 2013 5:44:00 PM UTC+2, ajith c t wrote: >>>> >>>> sorry I am not executing the code in the load balancer. Let me say once >>>> more, so it will be clear. >>>> >>>> I dont have any web2py code in the load balancer. >>>> >>>> What I meant is I have my code in another server(say X) , to which the >>>> user is redirected by the load balancer(say Y). >>>> >>>> when I execute the webpy application from server X , the index function >>>> , where my scheduler.queue_task call comes,in the default.py controller >>>> will not be called as I cant access it from X. That is expected. So what I >>>> asked was , for breaking down purpose shouldn't I move that call into the >>>> scheduler.py file in the models folder as the model file will be called. >>>> >>>> >>> I really don't get this part. Every controller in web2py gets executed >>> as long as the request comes to the server. If you can reach a page and in >>> your controller for that page you use queue_task(), you should definitely >>> be able to enqueue a task. queue_task() is NOT meant to be used in models, >>> because you'd queue a task for EACH and EVERY request coming in. >>> >>> >>>> >>>> And I will use the "seperation of duties" approach. but can you just >>>> confirm that the commands are correct. >>>> >>>> >>> web part: >>> python web2py.py -a yourpassword -p 8000 -i 0.0.0.0 >>> >>> scheduler >>> python web2py.py -K appname >>> >>> let me stress it out once more: serving a web app in production with the >>> included webserver is not going to provide stellar performances, although >>> it definitely works. >>> >> -- 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/groups/opt_out.
