| From: | Jeff Davis <pgsql(at)j-davis(dot)com> |
|---|---|
| To: | Amit Kapila <amit(dot)kapila16(at)gmail(dot)com>, Shlok Kyal <shlok(dot)kyal(dot)oss(at)gmail(dot)com> |
| Cc: | Noah Misch <noah(at)leadboat(dot)com>, pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: CREATE SUBSCRIPTION ... SERVER vs. pg_dump, etc. |
| Date: | 2026-08-06 02:18:02 |
| Message-ID: | 95a4a0d1f9b1a2e9d2e339b41b532131571150af.camel@j-davis.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Wed, 2026-08-05 at 16:47 +0530, Amit Kapila wrote:
> BTW, I had also looked at the overall patch series, the idea and
> high-level code looks good to me. Though I haven't done a detailed
> testing or review of the same but as Shlok and Kuroda-San seem to
> have
> reviewed/tested these patches, I think we can go-ahead with these
> fixes.
Thank you all, pushed.
AI-assisted summary of reported issues:
Finding 1: CREATE (connect=false) throws error during restore
fix: 5c39be9e6d, 254beb174e
Finding 2: OWNER TO could throw error during restore
fix: ac230cd3ce, 7f4fb05807
Finding 3: SCRAM passthrough
fix: d63f214d68
note: GSS delegation and other session state could still
cause differences between DDL and worker
Finding 4: REASSIGN OWNED fails across databases
fix: 03bcb6a777
Finding 5: error building conninfo before enabled check
fix: 8b73ceb78f
note: parallel worker fall-through left as edge case
Finding 6: concurrent DROP SERVER dangling subserver
note: withdrawn by Claude
Finding 7: ALTER that connects skipped foreign-server USAGE
fix: dcfb02deed, e90e198e3f
Finding 8: \dew omits fdwconnection
note: open
Finding 9: DROP SUBSCRIPTION hard-fails if mapping missing
with non-READY rels
note: discussed here:
https://postgr.es/m/CAA4eK1KzHQ_0rhN8uP+OB8dnB+uLBgod6tTx8j3GHxazz_=FiQ@m
ail.gmail.com
Finding 10: DROP SERVER CASCADE vs dependent subscription
fix: b66ea17a3a, b48caa7dfd
Finding 11: CREATE FDW docs omit CONNECTION return type
note: open
Finding 12: local ALTER forms fail without usable conninfo
fix: e90e198e3f, 6d04cd2b84
Finding 13: CREATE checks mapping before FDW capability
fix: 5c39be9e6d, 254beb174e
Finding 14: worker DEBUG1 logs subscription conninfo
fix: 448c4b3bfd
Finding 15: DROP SERVER errors before locking dependent
fix: b66ea17a3a
Finding 16: fdwhandler.sgml silent on connection functions
note: open
Finding 17: ALTER SUBSCRIPTION docs for SERVER / OWNER TO
note: partial
Finding 18: server application_name overrides subscription name
note: open; not clear whether this is correct or a bug
Finding 19: worker startup invalidation window for conninfo
note: open; similar issue is pre-existing
A few other problems were found and fixed along the way.
I believe the restore problem and other serious issues have been
addressed, so I will close the open item.
Regards,
Jeff Davis
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Peter Geoghegan | 2026-08-06 02:25:14 | Re: Whole row var issue |
| Previous Message | Xuneng Zhou | 2026-08-06 02:10:01 | Re: BUG: ReadStream look-ahead exhausts local buffers when effective_io_concurrency>=64 |