New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
SQL persistance problems #881
Comments
Okay, the relevant piece of logs is:
|
So -- several remarks -- first, the unit test cases for postgres pass for me, which tells me that you probably have postgress/odbc misconfigured. Please make sure that the unit test cases are passing. You will need to very carefully study the README file to do this. -- next, at this time, there is nothing at all in the chatbot that requires the database, although this will be interesting in the future, to save/restore everything that the chatbot might learn, conversationally. But this will require a big chunk of architectural work, so unless you are willing to go there, I'm not entirely sure why you are working on this. -- next, the people on #poostgres are giving you terrible advice-- they are clueless and don't know what they are talking about. Ignore them. -- next, there are two drivers in the atomspace: one that uses ODBC, and one that attempts to bypass it, and go directly (natively) to postgres. The ODBC driver works, and is stable. The native-postgres driver was never completed, and I have no clue if it actually works. So, in conclusion:
|
Let me try to answer all of this:
So, at this point I should be off to fiddling with unit tests :-) Thanks for the pointers! |
OK, again, there is a very detail HOWTO here https://github.com/opencog/atomspace/blob/master/opencog/persist/sql/README.md I suspect that a blanket-save of everything to the database will jsut be confusing, and that the chatbot and other systems will need to be altered to be selective for what they save, and for what they fetch. |
I've been following both the HOWTO and the test README before, but decided to double-check, hoping to find the error.
In my case, both tests basically caught the same error:
I still think I'm missing something basic. I also think that I need some way to examine that call to |
The database name is taken from the Mine looks like this:
|
Ok, this file does exist on my machine, its name starts with a
|
Comments on my semi-random poking around trying to figure it out:
Now, when I run a cogserver and
My guess is that there are 6 threads started, each of them (in due time; I haven't really thought trough all the workflow) tries to open a database by calling Update: some more code reading makes me think that
is from
. |
On Tue, Oct 4, 2016 at 1:22 PM, StrangeTcy notifications@github.com wrote:
The only way that the database would try to connect to it is if it appeared
For you, opening with opencog_test opencog_tester cheese should work, Yes, the error messages should be tightened up to indicate the actual --linas |
Yes; it's actually The
So, from this I take that both databases exist (were created) and are populated (as they should be after following the respective steps in the HOWTO), but somehow I can't connect to them. Update: I think I've figured out the dependencies and error propagation:
because: defined in
where
Great. But doesn't help me understand why I get the error, yet. |
I just added a postgres driver, so that odbc can be completely avoided. That should sharply simplify configuration mistakes. -- please try it! |
I've followed the instructions and did some editing on my own, and managed to get
guile -l run-chatbot.scm
from not working with sql at all toI tried to figure out what's wrong, but haven't yet.
I've probably made some simple mistake and some of my ODBC-related files are wrong. The Postgres database exists and I can connect to it with
psql
without any issue.All the necessary dependencies seem to be satisfied, but I'll use any double-checks offered.
The text was updated successfully, but these errors were encountered: