, Richard Galka wrote:
I had a setup for a while with freebsd, web2py (and corresponding
dependencies). There was very little difference on setup between amazone
ec2 and the bare metal machines I previously set everything up on. Just
remember, by default no guaranteed static IP...
I had a setup for a while with freebsd, web2py (and corresponding
dependencies). There was very little difference on setup between amazone
ec2 and the bare metal machines I previously set everything up on. Just
remember, by default no guaranteed static IP...
On Monday, April 30, 2012 4:59:21
I would also like to comment that I suspect (unconfirmed) that reloading of
modules while references are still around appears to potentially cause
memory leaks.. during dev while modules are reloaded, our app's memory
usage grows significantly. With this disabled, we find a consistent
For encrypting and decrypting files using a validator works quite well.
In our model we have something like:
Field('data',
'upload',
requires=[IS_NOT_EMPTY(),
IS_LENGTH(26214400, 0, error_message=File
too
For encrypting/decrypting DB columns, we also use a validator:
Ex:
db.define_table('table',
Field('name', 'text'),
Field('host', 'text', requires =
SecureValidator(symmetrickey)),
Field('secret', 'text', requires =
SecureValidator(symmetrickey))
)
If this is supposed to be a scalable application, then I'd really suggest
using some middleware layer. This way your front end(s) (web2py instances)
can make a request which could then be distributed to multiple services.
Example: First thought on a solution that comes to mind is web2py using
A couple minor points with the admin pages and the new debugging section:
In the Admin - Debug - breakpoints - Add breakpoints section, if the
current application has long directory file names then the table's hints
run off the page.
I think it may be pertinent to identify that in the
In general we use schemas to segregate data and provide access control
within web2py. (Authentication and access control is at the DB level and
not the web app level).
As mentioned by Manuele we also use executesql to set the search path as
required.
In our environment each schema may
A quick and dirty way to check what is going on is to add debug statements
in the web2py dal.py code.
I had a similiar issue, and the quickest way I found to check why compute
failed was to remove the except KeyError in the def _lisitfy of dal.py.
Then an error is thrown whenever the compute
9 matches
Mail list logo
