Thanks for the quick response! You still using pythonanywhere?
On Monday, May 23, 2016 at 10:12:49 AM UTC-4, Ian Ryder wrote: > Hi Andre, yes, I did indeed write my own - there's a lot in the built in > one we just didn't need and we really needed to be able to see everything > that was happening. > > I don't see the issue too often now and we have processes that have run > for a month without issue...from memory I think the main thing is commit > the database before calling a long running process (db.commit()). > > On Monday, 23 May 2016 16:05:36 UTC+2, Andre Kozaczka wrote: >> >> Hi Ian - I'm running into the same issue as you. Do you know the root >> cause of the error? Did you end up creating your own scheduling handler? >> >> -Andre >> >> On Thursday, February 5, 2015 at 5:55:04 AM UTC-5, Ian Ryder wrote: >> >>> Think it is pretty similar to this - we have 3 databases open. Seems to >>> only happen when a big process / query runs. >>> >>> More worrying now is that I'm getting situations where a process starts >>> and just disappears into the ether - no timeout, no fail, no error message. >>> Just starts and is never seen again...not even sure where to start with >>> that. >>> >>> I think I'm going to roll my own scheduling handler, I don't need >>> anything too complicated right now - loop away, look for a new request, run >>> it in isolation and record everything that happens along the way. Might be >>> back if that doesn't work out :) >>> >>> On Tuesday, February 3, 2015 at 3:07:40 PM UTC+11, Kiran Subbaraman >>> wrote: >>>> >>>> Looks similar to this open-issue: >>>> https://github.com/web2py/web2py/issues/733 >>>> >>>> ________________________________________ >>>> Kiran Subbaramanhttp://subbaraman.wordpress.com/about/ >>>> >>>> On Tue, 03-02-2015 2:47 AM, Niphlod wrote: >>>> >>>> uhm. does pythonanywhere even support a running process outside the web >>>> one ? >>>> >>>> BTW: all "kinky" calls to the db are fail-safed in the scheduler (see >>>> "wrapped_*" functions), to alleviate locking (albeit primarely for >>>> sqlite). >>>> This has the added benefit of making the scheduler more resilient >>>> against blocking, but this "has gone away" sounds like some problems with >>>> the underlying connectionpool........ >>>> -- >>>> 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 [email protected]. >>>> For more options, visit https://groups.google.com/d/optout. >>>> >>>> >>>> -- 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 [email protected]. For more options, visit https://groups.google.com/d/optout.
