building HEAD on macos fails with #error no source of random numbers configured
Looking at src/port/pg_strong_random.c this would be a bug in autoconf
#else
/* The autoconf script should not have allowed this */
#error no source of random numbers configured
My configure line is:
./configure --prefix=/usr/local/pgsql/10 --enable-debug --with-python
--with-openssl --with-libraries=/usr/local/opt/openssl/lib
--with-includes=/usr/local/opt/openssl/include/ --no-create --no-recursion
I am using openssl-1.0.2j
Regards,
Dave Cramer
Dave Cramer <davecramer@gmail.com> writes:
Looking at src/port/pg_strong_random.c this would be a bug in autoconf
It looks more like self-inflicted damage from here:
./configure --prefix=/usr/local/pgsql/10 --enable-debug --with-python
--with-openssl --with-libraries=/usr/local/opt/openssl/lib
--with-includes=/usr/local/opt/openssl/include/ --no-create --no-recursion
Why are you using either --no-create or --no-recursion? The former
*definitely* breaks things:
$ ./configure --help | grep create
-n, --no-create do not create output files
Presumably the proximate cause of that error message is that configure
hasn't updated pg_config.h from some ancient version thereof, as a
consequence of this switch.
I'm not sure what --no-recursion does, but I would say that we'd
consider that unsupported as well.
regards, tom lane
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
That will teach me to copy and paste a config from somewhere ...
Thanks
Dave Cramer
On 9 December 2016 at 10:43, Tom Lane <tgl@sss.pgh.pa.us> wrote:
Show quoted text
Dave Cramer <davecramer@gmail.com> writes:
Looking at src/port/pg_strong_random.c this would be a bug in autoconf
It looks more like self-inflicted damage from here:
./configure --prefix=/usr/local/pgsql/10 --enable-debug --with-python
--with-openssl --with-libraries=/usr/local/opt/openssl/lib
--with-includes=/usr/local/opt/openssl/include/ --no-create--no-recursion
Why are you using either --no-create or --no-recursion? The former
*definitely* breaks things:$ ./configure --help | grep create
-n, --no-create do not create output filesPresumably the proximate cause of that error message is that configure
hasn't updated pg_config.h from some ancient version thereof, as a
consequence of this switch.I'm not sure what --no-recursion does, but I would say that we'd
consider that unsupported as well.regards, tom lane
On 12/09/2016 05:43 PM, Tom Lane wrote:
Dave Cramer <davecramer@gmail.com> writes:
Looking at src/port/pg_strong_random.c this would be a bug in autoconf
It looks more like self-inflicted damage from here:
./configure --prefix=/usr/local/pgsql/10 --enable-debug --with-python
--with-openssl --with-libraries=/usr/local/opt/openssl/lib
--with-includes=/usr/local/opt/openssl/include/ --no-create --no-recursionWhy are you using either --no-create or --no-recursion? The former
*definitely* breaks things:$ ./configure --help | grep create
-n, --no-create do not create output filesPresumably the proximate cause of that error message is that configure
hasn't updated pg_config.h from some ancient version thereof, as a
consequence of this switch.I'm not sure what --no-recursion does, but I would say that we'd
consider that unsupported as well.
Interesting. Running config.status adds those --no-create --no-recursion
flags automatically. You can see them in the command-line at the top of
config.log, too. I never bothered to check what they do...
- Heikki
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
On 12/9/16 9:53 AM, Heikki Linnakangas wrote:
I'm not sure what --no-recursion does, but I would say that we'd
consider that unsupported as well.Interesting. Running config.status adds those --no-create --no-recursion
flags automatically. You can see them in the command-line at the top of
config.log, too. I never bothered to check what they do...
AIUI they have to do with config dependency checking (where a simple
make will detect if config needs to run again or not). I've been bitten
by this in the past as well. Maybe there's a way to get config to spit
out a warning for those options and have make filter the warning out.
--
Jim Nasby, Data Architect, Blue Treble Consulting, Austin TX
Experts in Analytics, Data Architecture and PostgreSQL
Data in Trouble? Get it in Treble! http://BlueTreble.com
855-TREBLE2 (855-873-2532)
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
On 12/11/16 1:52 PM, Jim Nasby wrote:
On 12/9/16 9:53 AM, Heikki Linnakangas wrote:
I'm not sure what --no-recursion does, but I would say that we'd
consider that unsupported as well.Interesting. Running config.status adds those --no-create --no-recursion
flags automatically. You can see them in the command-line at the top of
config.log, too. I never bothered to check what they do...AIUI they have to do with config dependency checking (where a simple
make will detect if config needs to run again or not). I've been bitten
by this in the past as well. Maybe there's a way to get config to spit
out a warning for those options and have make filter the warning out.
When config.status is out of date against configure, then
Makefile.global runs config.status --recheck, which internally runs the
original configure line with --no-create and --no-recursion added.
--no-create means not to create any output files, because the makefile
rules will create those. --no-recursion means not to run any configure
scripts in subdirectories (we don't use that functionality).
It's arguably a bit confusing that config.log then records the configure
line with --no-create and --no-recursion added. But other than that,
everything works as intended.
--
Peter Eisentraut http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers