UPDATE: My reasoning was incorrect below. It wasn’t the moving of some of the lock code to C that caused the issue. It was moving -D_POSIX_C_SOURCE up to the Makefile that caused the problem. The solution below is still the same though.
pgBackRest 2.07 was announced today. As usual, I immediately downloaded it and tried to get it up and running on my MacBook (currently running Mojave). It wasn’t as straightforward as one might hope, and the online instructions assume a Linux system, so I figured I’d write this up for anyone else attempting the same.
read moreAfter a lively discussion at work today about monitoring tools and use cases, I decided to see if I could use tail_n_mail, which I already use to monitor my PostgreSQL logs, to monitor my pgBackRest logs. It turns out that it can, and can do so fairly trivially.
read moreI’ve noticed several individuals inquiring lately about pgBouncer and how they can avoid putting all users and their passwords in it’s auth_file. After the most recent such inquiry (hi Richard!) I decided I’d write this"post"to
hopefully make it clearer how to use ‘pass-through auth’ and avoid maintaining your users and their passwords in an external file. So let’s see what this takes, shall we?
I recently helped a customer upgrade their PostgreSQL instance from 9.4.x on RHEL to 10.x on Ubuntu. While it initially sounded daunting, the use of pglogical and some planning actually made it rather straightforward. While there’s nothing new or original in this post, I still felt compelled to write it up both for posterity’s sake and for anyone else that might find the info useful as an example in their own endeavors.