On Mon, Jul 25, 2022 at 2:40 PM Michael Paquier <mich...@paquier.xyz> wrote: > > On Mon, Jul 25, 2022 at 12:19:40PM +0530, Bharath Rupireddy wrote: > > Thanks a lot Cary for reviewing. It will be great if you can add > > yourself as a reviewer and set the status accordingly in the CF entry > > here - https://commitfest.postgresql.org/38/3666/. > > Hmm. This stands for the connection error, but there are other things > that could cause a failure down the road, like an incorrect system > ID or a TLI-related report, so that seems a bit limited to me?
Good point. The walreceiver can exit for any reason. We can either 1) store for all the error messages or 2) think of using sigsetjmp but that only catches the ERROR kinds, leaving FATAL and PANIC messages. The option (1) is simple but there are problems - we may miss storing future error messages, good commenting and reviewing may help here and all the error messages now need to be stored in string, which is complex. The option (2) seems reasonable but we will miss FATAL and PANIC messages (we have many ERRORs, 2 FATALs, 3 PANICs). Maybe a combination of option (1) for FATALs and PANICs, and option (2) for ERRORs helps. Thoughts? -- Bharath Rupireddy RDS Open Source Databases: https://aws.amazon.com/rds/postgresql/
