all operations on redis are atomic, so it's quite easy to write such a system, e.g. adding and popping a redis list.
On Monday, June 17, 2013 8:20:55 PM UTC+2, ssuresh wrote: > > > > Since all my analytics will be done through the database, i require the > log data in the table(by end of day at least).. > So if I write all logs to redis and then can I run scheduler to flush it > to db? Do you think that would be the right approach? > > On Monday, June 17, 2013 9:10:31 PM UTC+5:30, Niphlod wrote: >> >> continous operations kinda don't play well with logs files.... issues are >> more prominent in a multiprocess environment. >> >> Redis can do it without hiccup, if you have it in your environment... >> >> >> >> Il giorno lunedì 17 giugno 2013 15:22:43 UTC+2, ssuresh ha scritto: >>> >>> hi, >>> I am planning to build some sort of user action analytics into my >>> system. For this, I need to log all user actions into the table . Since an >>> additional db write for every function call can be a performance issue, my >>> idea is to introduce an async batch write(say after every hundred lines of >>> logs, a job will bulk insert that into the table). >>> >>> Please advice on the best way of doing this using web2py. >>> >>> regds, >>> Suresh >>> >> -- --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to web2py+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
