Thread: WHat do people think about using in production?

We still are experiencing issues with SQLASCII and 7.4 ODBC, any feedback on the readiness of the libpq stuff?

I have not grabbed a current copy and built it since last Thursday, so not sure if any other changes are planned I should wait for, or even if you feel its ready or not for prime time yet.

I appreciate any feedback as I want to try the conversion again (the conversion went ok, but once in Unicode my app was having issues on the 8.0 pre libpq driver, and at that moment the libpq driver had memory leaks which I have verified seem to be fixed.)

Joel Fradkin

My personal opinion is that atm the libpq driver is already as good as or better than the pre-libpq driver.  There may be a few as of yet undiscovered reliability issues or busted features, so there is some risk with the switchover.  Since Dave fixed the excel bug, today I’m going to switch over my application stack to the new driver and give a more qualitative analysis.

Since the new driver is a much better design, (and supports the level 3 protocol!) it will only get better.  I think that you can also give some really good feedback about utf-8 capabilities of the driver (expect some problems here).  I hate to suggest new technology to bail you out of a nasty situation but ISTM this is the only situation with a high chance of long term success.  The only other option I see is to build a stronger test suite and hang back for a bit and closely monitor progress.

How heavily have you load tested the new driver?

Merlin

I couldn’t re-produce my problem in testing either, just in production.

I was unable to really hammer the new driver as my current test equipment is a 650meg XP box with our mail and monitoring software running as well against a 650 meg fedora core 3 database box (think this could of handled more, problem was in the front end IIS).

I was able to get 4 concurrent users testing without a loss of memory or any errors reported.

I am going to give it a try (let me know when the latest and greatest are in CVS so I can do a test build of what I will plan to go live with).

Worst case is I convert and it blows up bad like last time, best case is it stays afloat. I am currently having my IIS boxes crash any where from a few hours to a few days (just can not seem to nail down the cause, I could try turning on logs since Dave figured out I did not have permissions correct to write to c drive and see if it crashes, but I am fearful of the amount of log files and slow down).

-----Original Message-----
From: pgsql-odbc-owner@postgresql.org [mailto:pgsql-odbc-owner@postgresql.org] On Behalf Of Dave Page
Sent: Tuesday, July 26, 2005 8:47 AM
To: Merlin Moncure; Joel Fradkin
Cc: pgsql-odbc@postgresql.org
Subject: Re: [ODBC] WHat do people think about using in production?

 

---

From: pgsql-odbc-owner@postgresql.org [mailto:pgsql-odbc-owner@postgresql.org] On Behalf Of Merlin Moncure
Sent: 26 July 2005 13:24
To: Joel Fradkin
Cc: pgsql-odbc@postgresql.org
Subject: Re: [ODBC] WHat do people think about using in production?

My personal opinion is that atm the libpq driver is already as good as or better than the pre-libpq driver.  There may be a few as of yet undiscovered reliability issues or busted features, so there is some risk with the switchover.  Since Dave fixed the excel bug, today I’m going to switch over my application stack to the new driver and give a more qualitative analysis. 

 

Somewhat unintentionally I should add - it started working after what I believed were unrelated fixes :-) 

 

Since the new driver is a much better design, (and supports the level 3 protocol!) it will only get better.  I think that you can also give some really good feedback about utf-8 capabilities of the driver (expect some problems here).  I hate to suggest new technology to bail you out of a nasty situation but ISTM this is the only situation with a high chance of long term success.  The only other option I see is to build a stronger test suite and hang back for a bit and closely monitor progress.

 

I agree - the new driver does seem to be at least as good. There are some more improvements in CVS that haven't made the snapshots yet that you might want to wait for - specifically, backend error messages are reported properly now, rather than a default message being issued, a few leaks and other minor issues have been solved, and SSL support has been added (not that I expect Joel will want to use that).

 

How heavily have you load tested the new driver?

 

I've hammered it with ODBCBench, and a simple IIS app thrashed with ApacheBench. I've not made it crash yet :-) (but then I couldn't reproduce Joel's problems with the older driver).

Regards, Dave.