On Tue, Aug 2, 2022 at 6:55 PM Drouvot, Bertrand <bdrou...@amazon.com> wrote: > > Hi Bharath, > > On 7/14/22 11:43 AM, Bharath Rupireddy wrote: > > On Fri, Jul 8, 2022 at 5:54 PM Bharath Rupireddy > > <bharath.rupireddyforpostg...@gmail.com> wrote: > >> Looking at v2-0003 patch and emit_log_hook, how about we filter out > >> for those connectivity errors either based on error codes and if they > >> aren't unique, perhaps passing special flags to ereport API indicating > >> that it's a connectivity error and in the emit_log_hook we can look > >> for those connectivity error codes or flags to collect the stats about > >> the failure connections (with MyProcPort being present in > >> emit_log_hook)? This way, we don't need a new hook. Thoughts? > > Bertrand and Other Hackers, above comment may have been lost in the > > wild - any thoughts on it? > > Thanks for your feedback! > > I can see 2 issues with that approach: > > - We’ll not be able to track the “startup timeout case” (well, we may > not be able to track it anyway depending of what next to [1] will be) as > it does not emit any log messages. > > [1]: > https://www.postgresql.org/message-id/a1558d12-c1c4-0fe5-f8a5-2b6c2294e55f%40amazon.com
Yes, we wanted to be very quick in StartupPacketTimeoutHandler because it is a timeout signal handler after all. > - We’ll depend of the log_min_messages value (means > edata->output_to_server needs to be true for the emit_log_hook to be > triggered). Hm, we can just say that 'log_min_message setting will enable/disable the feature'. I agree with your first point of not having an error in StartupPacketTimeoutHandler hence I don't think using emit log hook for the connection failure stats helps. -- Bharath Rupireddy RDS Open Source Databases: https://aws.amazon.com/rds/postgresql/
