Sorry, I forgot --- this should have gone only to patches. ---------------------------------------------------------------------------
Bruce Momjian wrote: > Tom Lane wrote: > > The recent change to make log_min_messages SUSET provokes the following > > behavior: > > > > $ export PGOPTIONS="-d 5" > > $ psql > > psql: FATAL: 'log_min_messages': permission denied > > $ > > > > Considering that I *am* superuser, this is quite unacceptable. > > If you don't want to revert the change, propose another solution. > > Here is a proposed fix for the new SUSET of various variables. The > solution is to create a new GUC context called PGC_USERLIMIT, which > limits changes by non-super users. For example, non-super users can > turn on logging, but can't turn it off, and log_min_* logging can have > added output, but not less output. > > The first part of the patch prevent client PGOPTIONS from lowering the > debug level. The second part adds this new GUC context, then allows it > to be set properly. The tests are in two parts --- the first prevents > non-super users from changing the value inappropriately, and the second > allows postgresql.conf changes to apply to existing backends, i.e. if > postgresql.conf turns logging off via SET, turning it on via > postgresql.conf should propogate to the client, because the client can't > turn something off that the admin wants turned on --- that is the tricky > part that we have to be able to handle the settings in any order. > > Peter, how does this look? Is reset_val the proper value to test? -- Bruce Momjian | http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup. | Newtown Square, Pennsylvania 19073 ---------------------------(end of broadcast)--------------------------- TIP 4: Don't 'kill -9' the postmaster