RE: libpq debug log - Mailing list pgsql-hackers

From Jamison, Kirk
Subject RE: libpq debug log
Date
Msg-id D09B13F772D2274BB348A310EE3027C643B24D@g01jpexmbkw24
Whole thread Raw
In response to Re: libpq debug log  (Andres Freund <andres@anarazel.de>)
Responses Re: libpq debug log  (Andres Freund <andres@anarazel.de>)
List pgsql-hackers
On February 14, 2019 6:16 PM +0000, Andres Freund wrote:

> Hi,

> On 2018-11-28 23:20:03 +0100, Peter Eisentraut wrote:
> > This does not excite me.  It seems mostly redundant with using tcpdump.

> I think the one counter-argument to this is that using tcpdump in real-world
> scenarios has become quite hard, due to encryption. Even with access to the
> private key you cannot decrypt the stream.  Wonder if the solution to that
> would be an option to write out the decrypted data into a .pcap or such.

I agree that network debug trace logs would be useful for users not knowledgeable
of postgres internals, so I understand the value of the feature,
as long as only necessary/digestible information is outputted.
I'll also check the patch later.

For Andres, I haven't looked into tcpdump yet, but I'd like to ask whether
or not the decrypted output to .pcap (if implemented) may be useful to
application users. What could be the limitations?
Could you explain a bit further on the idea?

Regards,
Kirk Jamison



pgsql-hackers by date:

Previous
From: "Tsunakawa, Takayuki"
Date:
Subject: RE: 2019-03 CF Summary / Review - Tranche #2
Next
From: Donald Dong
Date:
Subject: Re: Actual Cost