[pgpool-hackers: 4519] Re: Introducing Global Connection Pooling Feature for Pgpool-II
Muhammad Usama
muhammad.usama at percona.com
Thu Aug 29 21:42:10 JST 2024
Hi Ishii-san,
Thank you for reviewing the global connection pool implementation and for
the patch to fix the rewrite timestamp test case. I have pushed your
changes to the GitHub repository.
The SSL context was created before spawning child processes, which was
necessary because the global connection pool needs to share backend sockets
across different processes, which requires a consistent SSL context in all
child processes. There is still a TODO item on my list to pass the SSL
session back to the global connection pool along with the socket when the
backend uses SSL. I am working on finding some time to implement this.
Additionally, I have fixed the connection idle time functionality, so we
should see one less failure in the regression tests. I am continuing to
work on the missing functionality to ensure the regression tests pass
cleanly.
Thanks
Best regards
Muhammad Usama
On Wed, Aug 28, 2024 at 11:27 AM Tatsuo Ishii <ishii at postgresql.org> wrote:
> > Hi Usama,
> >
> > I have looked into the GitHub repository you are working on for this
> > project:
> >
> >
> https://url.avanan.click/v2/___https://github.com/codeforall/pgpool2___.YXAzOnBlcmNvbmE6YTpnOmQ2NTFkYmFkZjVkNTM3ZDc5YWRiMzQ0YTAxZjEzYzc5OjY6MTljNjpmMDRhZGQ5YWZiNjU3MThhZTBhY2U1NzJjODQwYTE3N2RkNmRmOTBjMmE5ZjRlMmRlMDU5MjBjOGZmNzgzNmE2OnA6VDpO
> >
> > I noticed that you modified src/main/main.c:
> >
> > *** /home/t-ishii/work/Pgpool-II/current/pgpool2/src/main/main.c
> 2024-08-25 10:18:43.006798216 +0900
> > --- src/main/main.c 2024-08-27 16:00:51.388409353 +0900
> > ***************
> > *** 314,320 ****
> > --- 314,323 ----
> > * If ssl is enabled, initialize the SSL context
> > */
> > if (pool_config->ssl)
> > + {
> > SSL_ServerSide_init();
> > + SSL_ClientSide_init();
> > + }
> > #endif /* USE_SSL
> */
> >
> > /* check effective user id for watchdog */
> >
> > Can you explain why you need the modification? I thought SSL is not
> > related to the global connection feature...
>
> Probably I found the answer myself. It seems the commit message
> explains the reason of the change.
>
>
> https://url.avanan.click/v2/___https://github.com/pgpool/pgpool2/commit/9edbb207840704c35b4e405f067c1323ed13c400___.YXAzOnBlcmNvbmE6YTpnOmQ2NTFkYmFkZjVkNTM3ZDc5YWRiMzQ0YTAxZjEzYzc5OjY6YTgxNDo3NWI4ZWVhYjRmM2QwZDllZjlkNTE3NmE3M2RmMjNhYmIyMmQ0Njc1MmEyMGQxNmEzZGFjYmYzNTg4NGUxMDRiOnA6VDpO
>
> So I ran the regression test and got a few failures. Among them I have
> fixed 010.rewrite_timestamp. Patch attached.
>
> Best reagards,
> --
> Tatsuo Ishii
> SRA OSS K.K.
> English:
> https://url.avanan.click/v2/___http://www.sraoss.co.jp/index_en/___.YXAzOnBlcmNvbmE6YTpnOmQ2NTFkYmFkZjVkNTM3ZDc5YWRiMzQ0YTAxZjEzYzc5OjY6ZjA0NTo5ZmYyNWFjZjA5M2RjMjBkZWU2MDczNTBjMzM3YzE1MTY1NzBiZDNiOGFmM2JiNTYzNThhNjBmZjcxOTBlZDhhOnA6VDpO
> Japanese:
> https://url.avanan.click/v2/___http://www.sraoss.co.jp___.YXAzOnBlcmNvbmE6YTpnOmQ2NTFkYmFkZjVkNTM3ZDc5YWRiMzQ0YTAxZjEzYzc5OjY6NGRiNjpjYzljODJlYTRjZjM1MmI1MzllYzllNWU3NzI3ZjhkNzlhYzdkODc2ODhlZTRjNTliMzE1YWZhM2U1NmZiN2UzOnA6VDpO
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.pgpool.net/pipermail/pgpool-hackers/attachments/20240829/4c5599ef/attachment.htm>
More information about the pgpool-hackers
mailing list