| From: | Tatsuo Ishii <ishii(at)postgresql(dot)org> |
|---|---|
| To: | pgpool-hackers(at)lists(dot)postgresql(dot)org |
| Subject: | Re: Close listening socokets before forking |
| Date: | 2026-03-09 09:20:31 |
| Message-ID: | 20260309.182031.585673783467995854.ishii@postgresql.org |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgpool-hackers |
> Currently when pgpool main process forks sub processes (child process
> - user session process, pcp_main process, health check process,
> streaming replication check process and life check process), they
> inherits pgpool and pcp listening sockets. However some of processes
> do not need those sockets:
>
> - child process - pcp sockets are unnecessary
> - pcp main process - pgpool sockets are unnecessary
> - health check, streaming replication check and life check process -
> pgpool and pcp sockets are unnecessary
>
> It could be potential problem when those process go down. Since they
> may keep the listening sockets for a while, which prevents next pgpool
> starting up from binding those ports.
>
> Attached patch closes those unnecessary sockets after forking. For
> this purpose new function close_listening_sockets() is introduced.
Pushed down to v4.4 (v4.3 was hard to port the patch).
Best regards,
--
Tatsuo Ishii
SRA OSS K.K.
English: http://www.sraoss.co.jp/index_en/
Japanese:http://www.sraoss.co.jp
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Nadav Shatz | 2026-03-09 09:22:15 | Re: Proposal: Recent mutated table tracking in memory |
| Previous Message | Tatsuo Ishii | 2026-03-09 05:18:39 | Re: Proposal: Recent mutated table tracking in memory |