Still wrestling with this, anyone have any suggestions of somethings I
should maybe try?
Thanks.
On Feb 17, 4:17 am, "[EMAIL PROTECTED]" <[EMAIL PROTECTED]> wrote:
> Thanks for the reply. I tried what you suggested, still with the same
> results.
> I also tried some things from another discussion I found on here, but
> that didn't work either.
> I am using TG 1.0 and SQLObject
>
> # tried both with and without the extra URI setting
> from turbogears import database
> database.set_db_uri('postgres://URI')
>
> # No errors or exceptions from this code, but same result, TG/
> SQLObject hangs
> # on to the DB connection and doesn't let it go.
> from sqlobject.util.threadinglocal import local as threading_local
> hub.threadingLocal = threading_local()
> hub.begin()
> loads = Test.select()
> hub.commit()
> hub.end()
>
> The scheduler in production runs every 10 minutes and with 100 DB
> connections available, I have to restart TG every 16 hours at to make
> sure we don't get DB connect limited exceeded errors. I have debug and
> debugOutput turned on for SQLObject but I am not seeing anything that
> is really helping me figure out what I should do. I will look for more
> information about acquiring and releasing DB connections from a
> scheduler thread.
>
> On Feb 17, 2:29 am, "Richard Clark" <[EMAIL PROTECTED]> wrote:
>
> > Ok, keep in mind this stuff is for tg 1.0, it may not apply to later
> > versions.
>
> > The important thing to remember about the scheduler is that you're
> > running in a "user" created thread when your job gets called, you're
> > not in a tg created thread.
>
> > As such, the database needs managing in the same fashion it would if
> > you were running your own thread. You need to acquire a thread local
> > store, create a new database connection, etc.
>
> > It works basically like this, in your scheduled function:
>
> > def MyScheduledFunction:
> > # By getting the correct threading_local, the hub will presume it
> > isn't already connected (if this thread hasn't already been used) and
> > get you a connection
> > from sqlobject.util.threadinglocal import local as threading_local
> > hub.threadingLocal = threading_local()
> > hub.begin()
> > # now you can do your stuff
> > hub.commit()
>
> > This may or may not help with your particular problem, I have dozens
> > of solutions to things that slowly go out of fashion as turbogears
> > moves on :)
>
> > > Everytime the job runs (10 seconds) it opens a Postgres connection and
> > > leaves it in the idle in tranaction state. I can't figure out how to
> > > get the job to properly close the connection. I tried adding
> > > hub.begin() before and hub.end() after the select() call , but with
> > > the same results. Any ideas or nudges in the right direction? Thanks.
>
> > > I am using a file called jobs.py
>
> > > from model import Test, hub
> > > def test_schedule():
> > > test_result = Test.select()
> > > return
>
> > > def schedule():
> > > from turbogears import scheduler
> > > scheduler.add_interval_task(action=test_schedule,
> > > taskname='Test Schedule',
> > > initialdelay=0,
> > > interval=10)
> > > return
>
> > > In my start-test.py I have the entry
>
> > > from test import jobs
> > > jobs.schedule()
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"TurboGears" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/turbogears?hl=en
-~----------~----~----~----~------~----~------~--~---
