It could be. On Wednesday, 17 April 2013 10:23:45 UTC-5, Yarin wrote: > > 10 4 Massimo I will upgrade and update you guys on what I find. I wonder > also if that bug has something to do with the mysql locking issue I > describe here, which first started manifesting itself in March, a few days > after the first 'orphan records' began appearing... > https://groups.google.com/d/msg/web2py/iAlrv6yaHWk/m7Dr8ztcN9sJ > > On Wednesday, April 17, 2013 11:02:27 AM UTC-4, Massimo Di Pierro wrote: >> >> There was a buggy version that was doing this. Could have been 2.2.1. Not >> sure. Please upgrade to the latest. The reason is that is was saving >> sessions even when data was not stored/changed. Therefore is the cookie was >> not returned it would start making new session files/records. >> >> On Wednesday, 17 April 2013 08:26:30 UTC-5, Yarin wrote: >>> >>> 2.2.1 >>> >>> On Wednesday, April 17, 2013 9:16:41 AM UTC-4, Massimo Di Pierro wrote: >>>> >>>> Which web2py version? >>>> >>>> On Wednesday, 17 April 2013 08:08:48 UTC-5, Yarin wrote: >>>>> >>>>> (This question stems from an issue first brought up >>>>> here<https://groups.google.com/forum/?fromgroups=#!topic/web2py/iAlrv6yaHWk> >>>>> ) >>>>> >>>>> We store user sessions in the DB using the web2py_session_{app} table. >>>>> Inspecting the table, we've noticed a strange pattern happening over and >>>>> over again: >>>>> >>>>> Many of our app actions update the session. Generally, when a user >>>>> hits the app and causes a session update, a single session record >>>>> associated with that client gets updated in the DB. Additionally, >>>>> however, >>>>> we're noticing a large number of one-time 'orphan' session records which >>>>> will also be tied to the same IP, but that have created_datetime = >>>>> modified_datetime, meaning they were never updated. >>>>> >>>>> These one-time session records are usually spaced a few seconds or >>>>> minutes apart, as if they were being created once per request. Some days >>>>> there are dozens of these records, other days there are none. Each of >>>>> these >>>>> session records has its own unique key. >>>>> >>>>> It's happening in both remote production (LAMP) and local dev (MAMP) >>>>> environments. In both environments, however, no session records exist >>>>> that >>>>> are older than 3/10 (Today is 4/17), so something is clearing them out. >>>>> >>>>> I can't make sense of this. Ideas? >>>>> >>>>> >>>>> >>>>> >>>>>
-- --- 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 [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
