On Fri, Dec 18, 2020 at 08:41:01AM +0900, Michael Paquier wrote:
> Knowing that we are in a period of vacations for a lot of people, and
> that this is a sensitive area of the code that involves
> authentication, I think that it is better to let this thread brew
> longer and get more
nothing gets forgotten or cleaned up on the way until the
resowner that did the cryptohash allocation is handled.
Attached is a small extension I have played with by doing some error
injections, and a patch. If there are no objections, I would like to
commit this fix.
Thanks,
--
Michael
diff --git
now about the resowner dependency
only in the OpenSSL part of cryptohashes, without having to include
this knowledge in neither cryptohash.h nor in the fallback
implementation that can just use palloc() in the backend.
--
Michael
signature.asc
Description: PGP signature
On Fri, Dec 18, 2020 at 11:51:55AM +0200, Heikki Linnakangas wrote:
> Something like this. Passes regression tests, but otherwise untested.
... And I wanted to keep track of the type of cryptohash directly in
the shared structure. ;)
--
Michael
signature.asc
Description: PGP signature
eed to rush for this stuff.
I would rather wait more and make sure that we agree on an interface
people are happy enough with.
--
Michael
signature.asc
Description: PGP signature
r codes for the init, update and final but it is
possible to grab for errors then react on that). And we have even in
our tree code paths a-la-pgcrypto that have callbacks for each phase
with some processing in-between. HMAC also gets a bit cleaner by
keeping this flexibility IMO.
--
Michael
signature.asc
Description: PGP signature
On Sat, Dec 19, 2020 at 10:03:47AM +0530, Amit Kapila wrote:
> Glad to hear. I am confident that you can do justice to this role.
I also think you will do just fine. Thanks for taking care of this.
--
Michael
signature.asc
Description: PGP signature
ment from upthread to
return a failure if the caller passes NULL for the context, and
adjusted some comments. What do you think of the attached?
--
Michael
diff --git a/src/include/common/cryptohash.h b/src/include/common/cryptohash.h
index 6ead1cb8e5..817e07c9a2 100644
--- a/src/include/common/cr
On Fri, Dec 18, 2020 at 11:30:16AM -0500, Stephen Frost wrote:
> * Heikki Linnakangas (hlinn...@iki.fi) wrote:
>> +1
>
> Yeah, in a quick look, this looks like a good improvement.
Thanks. This has been applied as of 93e8ff8.
--
Michael
signature.asc
Description: PGP signature
extension allows its contents to be
relocated. The named schema must already exist.
While on it.. The docs are incorrect here. As you say,
CreateExtensionInternal() may internally create a schema.
--
Michael
signature.asc
Description: PGP signature
On Wed, Dec 16, 2020 at 10:01:11AM +0900, Michael Paquier wrote:
> On Tue, Dec 15, 2020 at 09:45:17PM -0300, Alvaro Herrera wrote:
> > I don't like this idea too much, because adding an option causes an ABI
> > break. I don't think we commonly add options in backbranches
hat CTAS or CREATE
MATVIEW will never ERROR if the relation already exists, with or
without EXPLAIN, EXECUTE or WITH NO DATA, so that gets us a consistent
behavior across all the patterns.
Note: I'd like to think that we could choose a better name for
CheckRelExistenceInCTAS().
--
Michael
From 6
ntexts. Forgot
that in the previous version actually.
--
Michael
diff --git a/src/include/catalog/index.h b/src/include/catalog/index.h
index c041628049..89394b648e 100644
--- a/src/include/catalog/index.h
+++ b/src/include/catalog/index.h
@@ -30,13 +30,16 @@ typedef enum
} IndexStateFlags
l-ish cleanup that removes
those references, as per the attached?
Thanks,
--
Michael
diff --git a/contrib/adminpack/Makefile b/contrib/adminpack/Makefile
index 630fea7726..851504f4ae 100644
--- a/contrib/adminpack/Makefile
+++ b/contrib/adminpack/Makefile
@@ -4,7 +4,6 @@ MODULE_big = adminpack
On Tue, Dec 22, 2020 at 11:24:40PM +1300, David Rowley wrote:
> On Wed, 11 Nov 2020 at 13:44, Michael Paquier wrote:
>> It seems to me that your patch is doing the right thing for adminpack
>> and that its Makefile has no need to include a reference to libpq
>> source path, n
eindex() will set options.tablespaceOid, not a pointer. Like
> this.
OK. Good to know, I have not looked at this part of the patch yet.
--
Michael
diff --git a/src/include/catalog/index.h b/src/include/catalog/index.h
index c041628049..89394b648e 100644
--- a/src/include/catalog/index.h
+++ b/src/include/catal
On Tue, Dec 22, 2020 at 03:12:15PM +0530, Bharath Rupireddy wrote:
> On Tue, Dec 22, 2020 at 2:07 PM Michael Paquier wrote:
>> Note: I'd like to think that we could choose a better name for
>> CheckRelExistenceInCTAS().
>
> I changed it to IsCTASRelCreationAllowed
ass down
a new option.
> No ; that was from an earlier revision of the patch adding the tablespace,
> before Michael suggested a ReindexOptions struct, which subsumes 'options' and
> 'tablespaceOid'.
>
> I see now that 'skip_constraint_checks' is fro
until the
period of vacations is over before wrapping this up to give others the
occasion to comment.
--
Michael
signature.asc
Description: PGP signature
On Thu, Dec 24, 2020 at 10:50:34AM +0900, Michael Paquier wrote:
> FWIW, it still makes the most sense to me to keep the options that are
> extracted from the grammar or things that apply to all the
> sub-routines of REINDEX to be tracked in a single structure, so this
> should incl
ions is over before wrapping this up to give others the
>> occasion to comment.
>
> Thanks! Happy Vacations!
You too!
--
Michael
signature.asc
Description: PGP signature
ed_memory is affected by
> the huge_pages setting on operating systems where
> that is supported, and may be more likely to benefit from larger
> pages
> on operating systems where that is managed automatically.
The current formulation is not that confusing
d
either, as we should avoid incorrect computations or crash on OOM, not
to mention that this would fail the detection of bugs coming directly
from OpenSSL or any other SSL library this code plugs with. In short,
I think that there are more benefits in letting the caller control the
error handling
tension, though this is not really
something related to postgres_fdw. Or just refactor the code into
src/common/ and then consume the refactored pieces from libpq and the
backend.
--
Michael
signature.asc
Description: PGP signature
utton at the top. From there, you should
be able to control the status of all the existing commit fest and
patches.
--
Michael
signature.asc
Description: PGP signature
Second sentence sounds a bit strange by combining "the" and a GUC
name. I would just write "Default is default_table_a_m".
--
Michael
signature.asc
Description: PGP signature
with a cheap succession of commands by loading the data
in a transaction done after a TRUNCATE?
[1]:
https://www.postgresql.org/message-id/20201001.115136.288898409051085426.horikyota@gmail.com
[2]: https://www.postgresql.org/docs/current/populate.html#POPULATE-COPY-FROM
--
Michael
signature.asc
Description: PGP signature
e partitions inherit
> that value.
That's indeed more interesting from the user perspective. So +1 from
me.
--
Michael
signature.asc
Description: PGP signature
isted in the CF for this thread:
https://commitfest.postgresql.org/31/2857/
And it seems to me that waiting on author is most appropriate here, so
I have changed the entry to reflect that.
--
Michael
signature.asc
Description: PGP signature
e signature (pg_stat_statements_1_8 et al).
With the 1.8 flavor, it is possible to pass down a negative number
and it may not fail depending on the number of blocks of the relation,
so I think that you had better have a compatibility layer if a user
has the new binaries but is still on 1.8.
permissions of the app, but if that proves to be
necessary I am fine to switch the CF status. Just let me know if you
want me to do so when the time comes.
--
Michael
signature.asc
Description: PGP signature
I don't really know if I should
act on it or not :)
--
Michael
signature.asc
Description: PGP signature
On Tue, Dec 29, 2020 at 01:59:58PM +1300, Thomas Munro wrote:
> LGTM.
Thanks, I have done this one then.
--
Michael
signature.asc
Description: PGP signature
t, I have included catalogs.sgml, pgstatstatements.sgml,
explain.sgml, pg_verifybackup.sgml and wal.sgml in 13.
> doc/src/sgml/cube.sgml?
> doc/src/sgml/func.sgml?
These two are some beautification for the format of the function, so I
have left them out.
--
Michael
signature.asc
Description: PGP signature
tion Time: N.N ms
+ Execution Time: -N.N ms
(3 rows)
Not sure what to think about that, as this implies the calculation of
a negative execution time.
Regards,
--
Michael
signature.asc
Description: PGP signature
;
> since this Makefile has never actually linked libpq, it's hard to see why
> it'd be useful here.
Thanks David and Tom. I got around to apply this cleanup.
--
Michael
signature.asc
Description: PGP signature
ee that this looks like a monotonic
clock going backwards. Or could it be possible that it gave 0.0 as
result, still a minus sign was appended? That would mean an execution
that took less than 1us per the system clock.
--
Michael
signature.asc
Description: PGP signature
, no?
Your change also means that we may call superuser() more times than
necessary, but last_roleid_is_super leverages that, and this code path
is not performance-critical either.
I am adding Stephen in CC, just in case he has some comments to
share.
--
Michael
signature.asc
Description: PGP signature
On Thu, Dec 24, 2020 at 01:23:40PM +0900, Michael Paquier wrote:
> Please note that I have added an entry in the CF app for the moment so
> as we don't lose track of it:
> https://commitfest.postgresql.org/31/2892/
I have been able to look at that again today, and applied it. I
Do you have some references?
Just a matter of curiosity :)
--
Michael
signature.asc
Description: PGP signature
#x27;s just necessary for the
decode routine that's the only thing using get_hex().
- Let's prefix the routines in src/common/ with "pg_", to be
consistent with base64.
- It would be good to document the top each routine in hex.c (see
base64.c for a similar reference).
--
Michael
signature.asc
Description: PGP signature
ion, or does anybody see a big problem with that
(compared to the others in that set)?
I initially pondered not introducing a new context but just filtering on
category, but as categories seem to be only descriptive in guc.c and not
used for any policy decisions so far, I have abandoned this pretty
qui
Hi,
Am Montag, den 28.12.2020, 20:41 +0900 schrieb Masahiko Sawada:
> On Sat, Nov 28, 2020 at 7:50 AM Michael Banck
> wrote:
> > https://www.postgresql.org/docs/current/default-roles.html mentions the
> > "Administrator" several times, but does not specify it furth
checksums these days that it warrants to
be moved into the "normal part", like in the attached.
Michael
--
Michael Banck
Projektleiter / Senior Berater
Tel.: +49 2166 9901-171
Fax: +49 2166 9901-100
Email: michael.ba...@credativ.de
credativ GmbH, HRB Mönchengladbach 12080
USt
TH is a good idea either, I'd like to get
rid of it in the long-term) and ECPG, so that's clearly a gain.
I don't have a Windows host at hand, though I think that it should
work there correctly. What do you think about the ideas in the
attached patch?
--
Michael
From 8952fb2d5e5a7a0
On Fri, Jan 01, 2021 at 08:34:34PM +0100, Michael Banck wrote:
> I think enough people use data checksums these days that it warrants to
> be moved into the "normal part", like in the attached.
+1. Let's see first what others think about this change.
--
Michael
signature
On Tue, Dec 29, 2020 at 06:22:43PM +0900, Michael Paquier wrote:
> Yes, I have done an extra effort on those fixes where needed. On top
> of that, I have included catalogs.sgml, pgstatstatements.sgml,
> explain.sgml, pg_verifybackup.sgml and wal.sgml in 13.
Justin, I got to look at th
) xtCxt->xmlerrcxt,
> xml_errorHandler);
I quite like your suggestion to be a maximum simple here, and the docs
of upstream also give a lot of context:
http://xmlsoft.org/html/libxml-xmlerror.html#xmlSetStructuredErrorFunc
So let's use this version and call it a day for this part.
--
Michae
Heya,
(changing the subject as we're moving the goalposts)
Am Samstag, den 02.01.2021, 10:47 -0500 schrieb Stephen Frost:
> * Michael Paquier (mich...@paquier.xyz) wrote:
> > On Fri, Jan 01, 2021 at 08:34:34PM +0100, Michael Banck wrote:
> > > I think enough people use da
e case you are looking for frontend tools
able to do logging. Merging ECPG and the backend requires switching
to a logic where we return more than one error code so we could just
use an enum for the result result à-la-PQping like in libpq.
--
Michael
signature.asc
Description: PGP signature
ou may want to document as a comment the reason
why this gets sent even if (wal_level < logical).
--
Michael
signature.asc
Description: PGP signature
etched on startup by configuring primary_conninfo or
restore_command." ("or fetched on startup by configuring
recovery.conf" in some older branches). This confusion has been
introduced by 878bd9a, down to 9.6.
Heikki, what do you think?
--
Michael
signature.asc
Description: PGP signature
nline activation of checksusm, so could pretty much be
committed verbatim right now, except for the "either as an offline
operation or in a running cluster" bit which would have to be rewritten.
> +
> +
> + Off-line Enabling of Checksums
> +
> +
> +The linkend="app-
DatachecksumsWorkerShmem->target = enable_checksums;
This one is especially confusing as it looks like we're unconditionally
enabling checksums but instead we're just setting the target based on
the bool (see above). It might be our standard notation though.
[...]
> +/*
&
On Sun, Jan 03, 2021 at 09:05:09PM +0900, Michael Paquier wrote:
> So let's use this version and call it a day for this part.
This has been done as of b49154b.
--
Michael
signature.asc
Description: PGP signature
On Mon, Jan 04, 2021 at 07:11:43PM +0100, Michael Banck wrote:
> Am Samstag, den 02.01.2021, 10:47 -0500 schrieb Stephen Frost:
>> * Michael Paquier (mich...@paquier.xyz) wrote:
>>> On Fri, Jan 01, 2021 at 08:34:34PM +0100, Michael Banck wrote:
>>> > I think enough p
nto shared libs is fine by me at the end
because we don't have a case for libpq yet, but I object to the lack
of protection against overflows.
--
Michael
signature.asc
Description: PGP signature
Hi,
Am Mittwoch, den 06.01.2021, 10:52 +0900 schrieb Michael Paquier:
> On Mon, Jan 04, 2021 at 07:11:43PM +0100, Michael Banck wrote:
> > So maybe we should switch on wal_compression if we enable data checksums
> > by default.
>
> I don't agree with this assump
On Sat, Dec 19, 2020 at 03:13:50PM +0900, Michael Paquier wrote:
> At the same time, I have taken care of your comment from upthread to
> return a failure if the caller passes NULL for the context, and
> adjusted some comments. What do you think of the attached?
I have looked agai
Dump transaction origin information only for
> abort prepared. We need this during recovery to update the replication
> origin progress."?
That sounds fine.
--
Michael
signature.asc
Description: PGP signature
Hi,
On Wed, Jan 06, 2021 at 09:55:08AM -0800, Andres Freund wrote:
> On 2021-01-06 12:02:40 -0500, Stephen Frost wrote:
> > * Andres Freund (and...@anarazel.de) wrote:
> > > On 2021-01-04 19:11:43 +0100, Michael Banck wrote:
> > > > This looks much better from the
instance around handily right now to check.
Micael
--
Michael Banck
Projektleiter / Senior Berater
Tel.: +49 2166 9901-171
Fax: +49 2166 9901-100
Email: michael.ba...@credativ.de
credativ GmbH, HRB Mönchengladbach 12080
USt-ID-Nummer: DE204566209
Trompeterallee 108, 41189 Mönchengladbach
Gesc
Am Mittwoch, den 06.01.2021, 19:07 +0100 schrieb Michael Banck:
> Am Mittwoch, den 06.01.2021, 09:58 -0800 schrieb Andres Freund:
> > It still requires running a binary locally on the DB server, no? Which
> > means it'll not be an option on most cloud providers...
>
> A
RDS use --data-checksums during
> initdb?
At least data_checksums=on for Azure Managed Postgres, Amazon RDS and
Google Cloud SQL Postgres. It might not be on for all types (I just
checked the basic one each earlier today), but I guess it is.
Michael
--
Michael Banck
Projektleiter /
Am Mittwoch, den 06.01.2021, 13:08 -0800 schrieb Peter Geoghegan:
> On Wed, Jan 6, 2021 at 1:04 PM Michael Banck
> wrote:
> > At least data_checksums=on for Azure Managed Postgres, Amazon RDS and
> > Google Cloud SQL Postgres. It might not be on for all types (I just
> &g
s
1GB for the smallest possible one and 32GB for a production instance.
It's 1GB for the small Google Cloud SQL Postgres and 2GB for the small
RDS test instance. It is user-changeable everywhere (at least to some
degree).
Michael
--
Michael Banck
Projektleiter / Senior Berater
Tel.: +49
On Mon, Dec 14, 2020 at 12:48:15PM +0900, Michael Paquier wrote:
> This is a nice cleanup, so I have moved ahead and applied it. A
> rebased version of the SHA1 business is attached.
Rebased version attached to address the conflicts caused by 55fe26a.
I have fixed three places in pgcrypto
On Wed, Jan 06, 2021 at 03:27:03PM +0200, Heikki Linnakangas wrote:
> Looks fine to me.
Thanks, I have been able to get this part done as of 55fe26a.
--
Michael
signature.asc
Description: PGP signature
l except if we use more switch/case
to set up the various variables. What about something like the
attached?
--
Michael
diff --git a/src/include/common/checksum_helper.h b/src/include/common/checksum_helper.h
index ebdf1ccf44..cac7570ea1 100644
--- a/src/include/common/checksum_helper.h
+++ b/s
On Tue, Jan 05, 2021 at 11:54:42AM +0300, Amine Tengilimoglu wrote:
> Thank you Michael. I agree with you. Relevant part can be removed from the
> document and eliminate the confusion at least.
Okay, I got around this stuff, and committed a fix for 9.6~12. Thanks
for the report,
Hi,
On Thu, Jan 07, 2021 at 03:05:44PM +0100, Daniel Gustafsson wrote:
> > On 5 Jan 2021, at 18:19, Michael Banck wrote:
> >> +
> >> + Data pages are not checksum protected by default, but this can
> >> optionally be
> >> + enabled for a cluster.
ink more about that.
>> Using separate fields looked cleaner to me if it came to debugging,
>> and the cleanup was rather minimal except if we use more switch/case
>> to set up the various variables. What about something like the
>> attached?
>
> +1
Thanks, I have com
On Fri, Dec 18, 2020 at 03:46:42PM +0900, Michael Paquier wrote:
> This has been tested on Windows and Linux across all the versions of
> OpenSSL we support on HEAD. I am also attaching a small module called
> hmacfuncs that I used as a way to validate this patch across all the
>
sentences? Maybe something like "invalid block
number %u referring to meta page" and "block number out of range for
relation %s: %llu"?
--
Michael
signature.asc
Description: PGP signature
On the ohter
hand, we're just returning early from print_aligned_vertical() right now
and never get to the part where we usually print the title if there are
zero results.
Michael
--
Michael Banck
Projektleiter / Senior Berater
Tel.: +49 2166 9901-171
Fax: +49 2166 9901-100
Email: micha
down result_size() and down to the cryptohash
functoins, meaning that there is no need to take about it more than
that IMO. The second one would be tied to the HMAC refactoring. This
would be valuable in the case of pgcrypto when building with OpenSSL,
meaning that the code would go through the
h is: Ready for Committer
So that's this patch: https://commitfest.postgresql.org/32/2941/.
Alvaro is most likely going to take care of that, so let's wait for
him.
--
Michael
signature.asc
Description: PGP signature
On Thu, Feb 11, 2021 at 02:50:13PM -0500, Robert Haas wrote:
> +1 for not back-patching whatever we do here.
+1.
--
Michael
signature.asc
Description: PGP signature
creating a bogus rbtxn. THus it is wrong to set create=true and
> create_as_top=true. We could change the assertion like Assert (txn &&
> txn->base_snapshot) to make things clearer.
I don't see much the point to change this code. The result would be
the same: a PANIC at th
gram.y, and you are
covering all of them. So this looks good to me. I'll apply and
backpatch in a bit. It is worth noting that tab-complete.c does a bad
job in completing those clauses.
--
Michael
signature.asc
Description: PGP signature
In short, any *future*
potential bug fix that would require to be backpatched in need of
using this function or touching its area would result in a conflict.
This changes makes no sense.
--
Michael
signature.asc
Description: PGP signature
On Sat, Feb 13, 2021 at 09:33:48PM -0300, Ranier Vilela wrote:
> Em sáb., 13 de fev. de 2021 às 20:32, Michael Paquier
> escreveu:
>
>> You are missing the point here. What you are proposing here would not
>> be backpatched. However, reusing the same words as upthread,
could
finish by being inlined, allowing the planner to apply optimizations
as it would know the function body. Going with SPI has the advantage
to have code able to work without any changes across major versions,
which is a no-brainer when it comes to long-term maintenance.
--
Michael
signature.asc
Description: PGP signature
checksum type is added.
I have reviewed my patch this morning, tweaked a comment, and applied
it.
--
Michael
signature.asc
Description: PGP signature
ave patched the top of the
function if it were me, btw.
--
Michael
signature.asc
Description: PGP signature
aves 090_reindexdb.pl,
but I was not really convinced that this was worth the extra test
cycles (I am aware of the --tablespace option missing in reindexdb,
someone I know was trying to get that done for the next CF).
--
Michael
signature.asc
Description: PGP signature
On Tue, Feb 09, 2021 at 04:27:34PM +0900, Michael Paquier wrote:
> Putting sanity checks within all the table_* functions of tableam.h
> would not be a good idea, as nothing prevents the call of what's
> stored in rel->rd_tableam.
I have been playing with this idea, and
egrity checks.
It seems to me that the first two sentences of this comment should be
combined together.
--
Michael
signature.asc
Description: PGP signature
of atomics, we would
prepare a spinlock for nothing.
--
Michael
signature.asc
Description: PGP signature
On Sat, Jan 23, 2021 at 01:43:20PM +0900, Michael Paquier wrote:
> Rebased patch is attached wiht SHA1 added as of a8ed6bb. Now that
> SHA1 is part of the set of options for cryptohashes, a lot of code of
> pgcrypto can be cleaned up thanks to the refactoring done here, but
> I am lea
rg/gitweb/?p=postgresql.git;a=commit;h=cc072641d41c55c6aa24a331fc1f8029e0a8d799
While the default is still superuser-only, it becomes possible to
grant access to this stuff to other roles that have no need to be
superusers.
--
Michael
signature.asc
Description: PGP signature
ot;INDEX", MatchAny, "DEPENDS"))
- COMPLETE_WITH("ON EXTENSION");
The part, if removed, means that typing "alter index my_index no " is
not able to complete with "DEPENDS ON EXTENSION" anymore. So it seems
to me that ALTER INDEX got that right, and that the other commands had
better do the same.
--
Michael
signature.asc
Description: PGP signature
ith a write(0) when starting the backend. There are two of
them in proc.c, one in twophase.c. Do you mind if I add an open item
for this one?
--
Michael
signature.asc
Description: PGP signature
On Tue, Feb 16, 2021 at 07:54:32AM +, Zohar Gofer wrote:
> Thanks. This seems to be the fix we need.
> Would it be possible to push it to previous versions? 12 or 13?
New features don't go into stable branches, only bug fixes do. And
this is not a bug fix, but a feature.
discussed as of 1e324cb:
https://www.postgresql.org/message-id/caltqxtcogrfevp9uou5vftngsn+vhzuu9+9a0inarfyvohs...@mail.gmail.com
Anyway, after sleeping on it, I have just applied your original patch
as that's simpler, and will cover the cases people would care for.
--
Michael
signature.asc
Description: PGP signature
On Tue, Feb 16, 2021 at 11:47:52PM +0900, Fujii Masao wrote:
> On 2021/02/16 15:50, Michael Paquier wrote:
>> + /*
>> +* Read "writtenUpto" without holding a spinlock. So it may not be
>> +* consistent with other WAL receiver's shared variables protec
command is working
on and on which relation the work is currently done. But I am not
really sure that we need that as long a we have a VERBOSE option that
lets us know via the logs what already happened in a single command.
I see no bug here.
--
Michael
signature.asc
Description: PGP signature
On Wed, Feb 17, 2021 at 12:10:43AM -0600, Justin Pryzby wrote:
> On Wed, Feb 17, 2021 at 02:55:04PM +0900, Michael Paquier wrote:
>> I see no bug here.
>
> pg_stat_progress_create_index includes partitions_{done,total} for
> CREATE INDEX p, so isn't it strange if it w
ons, for objectaddress functions, etc. That
would make this patch set simpler, switching
get_collation_version_for_oid() to just use a missing_ok argument.
And that would be more consistent with the other syscache lookup
functions we have here and there in the tree.
--
Michael
signature.asc
Description: PGP signature
can use the crypto part properly, as pqsecure_initialize
gets only called now when libpq negotiates SSL with the postmaster.
It seems to me that we should make sure of a proper reset of the
crypto part within pqDropConnection(), while the initialization needs
to happen in PQconnectPoll().
--
Michael
signature.asc
Description: PGP signature
201 - 300 of 11799 matches
Mail list logo
