Re: Allowing REINDEX to have an optional name
От | Simon Riggs |
---|---|
Тема | Re: Allowing REINDEX to have an optional name |
Дата | |
Msg-id | CANbhV-E5N6x9p0pgxG7bZziycOBP5pKyk0GqvRu5mbWpLLcuDg@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: Allowing REINDEX to have an optional name (Tom Lane <tgl@sss.pgh.pa.us>) |
Список | pgsql-hackers |
On Fri, 15 Jul 2022 at 15:03, Tom Lane <tgl@sss.pgh.pa.us> wrote: > > Alvaro Herrera <alvherre@alvh.no-ip.org> writes: > > Ah, one theory is that the libpq_pipeline program is getting linked to > > an installed version of libpq that doesn't contain the fixes. Maybe you > > can do `ldd /path/to/libpq_pipeline` and see which copy of libpq.so it > > is picking up? > > That's pronounced "otool -L" on macOS. But in any case, it's going > to point at the installation directory. One of the moving parts here > is that "make check" will try to override the rpath that otool tells > you about to make test programs use the libpq.dylib from the build tree. > I say "try" because if you've got SIP enabled (see what "csrutil status" > tells you), it will fail to do so and the installed libpq will be used. > Maybe that's old. > > Standard recommendation on macOS with SIP on is to always do "make > install" before "make check". Thanks, will investigate. -- Simon Riggs http://www.EnterpriseDB.com/
В списке pgsql-hackers по дате отправления: