On Mon, Jan 30, 2017 at 11:20 AM, Jonathan Vanasco <[email protected]> wrote: > > > On Sunday, January 29, 2017 at 12:44:57 PM UTC-5, Mike Orr wrote: >> >> I sometimes use a separate database connection with autocommit for >> logging, but I do it in a request finalizer. I also do it in a >> separate database because the database becomes large and other >> applications use it too. > > > This has to be done in real time to work correctly. my logging tables for > this type of stuff are on a separate backup strategy, and quickly cleared > out - there's no need for them after a reconciliation is done.
Your logs must be for a different purpose than mine. My logs are for future queries: how much is a section used, what are the most popular records and search terms, what are the most common failed searches (misspellings we may want to do something about), etc. You don't know what future needs will be. We have to strike a balance between the usefulness of the dataset and the volume of records. So "access log" records we keep for 60-90 days and make monthly counts from. Searches are kept until the next version is released, and then the previous one is archived as "top 100 version X.X search terms". As for the "orphan file/missing file" problem I mentioned with a non-transactional filesystem, we handle that with a routine that looks for such inconsistencies and makes a report. -- You received this message because you are subscribed to the Google Groups "pylons-discuss" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/pylons-discuss/CAH9f%3Duq%2BGwz39tHEgnQSa9frtwctXP3sw-ausq%2B%3D4CfZUzfRAg%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
