Normally under that load the database becomes the problem: 100
concurrent database connections.
Are you using connection pools?

Massimo

On Mar 24, 11:21 am, AchipA <[email protected]> wrote:
> While I do pretty fancy stuff with web2py occasionally, in this
> particular case it's a vanilla application. The model did not change
> for quite a while now. I could pretty reliably replicate the load
> conditions (on thath machine at around 100 concurrent active
> connections per WSGI process and a load of 60, it starts giving the
> above error). Yes, I know, that's pretty extreme, but hey, you have
> take it to extremes to earn the name of an enterprise framework,
> right ? :)
>
> On Mar 24, 3:07 pm, mdipierro <[email protected]> wrote:
>
> > I will look into this error but you have an error into the _create
> > function. This is called ONLY when a new table is created. Heavy or
> > light load should not affect this since function is called only once.
>
> > My guess is that you are editing the models in real time while the
> > server is under heavy load.  In general this is not a good idea
> > because the migration can take long time, log the database until
> > commit, and create too many database connections.
>
> > Anyway. The error should not occur and I will fix asap.
>
> > Massimo
>
> > On Mar 23, 9:28 am, AchipA <[email protected]> wrote:
>
> > > I have an interesting situation, where while in a Apache/WSGI 1
> > > process environment, running under high load (which makes this a
> > > likely race condition/lock issue) web2py just explodes:
>
> > > [Mon Mar 23 14:19:44 2009] [error] Traceback (most recent call last):
> > > [Mon Mar 23 14:19:44 2009] [error]   File "/var/www/web2py/gluon/
> > > sql.py", line 859, in define_table
> > > [Mon Mar 23 14:19:44 2009] [error]     query = t._create(migrate=args
> > > ['migrate'])
> > > [Mon Mar 23 14:19:44 2009] [error]   File "/var/www/web2py/gluon/
> > > sql.py", line 1183, in _create
> > > [Mon Mar 23 14:19:44 2009] [error]     sql_fields_old = cPickle.load
> > > (tfile)
> > > [Mon Mar 23 14:19:44 2009] [error] EOFError
>
> > > If I moderate myself and lower the load on the particular instance,
> > > everything works nice. Anybody encounter anything similar ?
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"web2py Web Framework" 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/web2py?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to