Re: New Object Access Type hooks - Mailing list pgsql-hackers

From Mark Dilger
Subject Re: New Object Access Type hooks
Date
Msg-id B188F52A-69B8-46E3-A32B-2FAFD9447FEF@enterprisedb.com
Whole thread Raw
In response to Re: New Object Access Type hooks  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: New Object Access Type hooks  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers

> On Apr 4, 2022, at 12:05 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>
> I wrote:
>> The "terminating connection" warning absolutely should get through,
>
> ... oh, no, that's not guaranteed at all, since it's sent from quickdie().
> So scratch that.  Maybe we'd better add "could not send data to server"
> to the regex?

If it fails in pqsecure_raw_write(), you get either "server closed the connection unexpectedly" or "could not send data
toserver".  Do we need to support pgtls_write() or pg_GSS_write(), which have different error messages?  Can anybody
runthe tests with TLS or GSS enabled?  I assume the test framework prevents this, but I didn't check too closely.... 

Is it possible that pgFlush will call pqSendSome which calls pqReadData before trying to write anything, and get back a
"couldnot receive data from server" from pqsecure_raw_read()? 

It's a bit hard to prove to myself which paths might be followed through this code.  Thoughts?

—
Mark Dilger
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company






pgsql-hackers by date:

Previous
From: Robert Haas
Date:
Subject: Re: Pluggable toaster
Next
From: Andres Freund
Date:
Subject: Re: Mingw task for Cirrus CI