On Thu, 27 Aug 2020 12:44:11 -0400, Tom Lane <tgl@xxxxxxxxxxxxx> wrote: > It looks like send() itself is returning EACCES, which seems just > weird. The send(2) man page does cite some possible causes of > EACCES, but none of them seem relevant here. Oh, send(2) on OpenBSD [1] says that the firewall may cause such an error. I guess no one ever hits* this since on Linux this is "very unlikely" to happen. I think the problem is my cron fiddles with the vm's firewall at the same time. I'll look into that possibility. Thanks for the hint! * according to my web search :p [1]: https://man.openbsd.org/send.2#EACCES Daniel
