Thread: honour TEMP_CONFIG in pg_upgrade tests

I propose to have the pg_upgrade test honour TEMP_CONFIG as pg_regress 
(or its Makefile) does, by applying the patch below. There might be 
opther places this should also be done, but this would be a start. The 
motivation is to see if it stops the errors we often see on the somewhat 
slow axolotl buildfarm member when checking pg_upgrade - axolotl's 
TEMP_CONFIG sets up the stats temp directory on a ramdisk, as well as 
turning off fsync.

This would be backpatched to 9.2.


cheers

andrew


diff --git a/src/bin/pg_upgrade/test.sh b/src/bin/pg_upgrade/test.sh
index ec3a7ed..c0dcf8b 100644
--- a/src/bin/pg_upgrade/test.sh
+++ b/src/bin/pg_upgrade/test.sh
@@ -21,6 +21,10 @@ unset MAKELEVEL # authentication configuration. standard_initdb() {    "$1" -N
+   if [ -n "$TEMP_CONFIG" -a -f "$TEMP_CONFIG" ]
+   then
+       cat "$TEMP_CONFIG" >> "$PGDATA/postgresql.conf"
+   fi    ../../test/regress/pg_regress --config-auth "$PGDATA" }

Andrew Dunstan <andrew@dunslane.net> writes:
> I propose to have the pg_upgrade test honour TEMP_CONFIG as pg_regress 
> (or its Makefile) does, by applying the patch below. There might be 
> opther places this should also be done, but this would be a start.

No objection to the concept, but I wonder if the -f test should
instead be a -r test (in the original place too, I guess).
        regards, tom lane

On 09/17/2015 09:42 AM, Tom Lane wrote:
> Andrew Dunstan <andrew@dunslane.net> writes:
>> I propose to have the pg_upgrade test honour TEMP_CONFIG as pg_regress
>> (or its Makefile) does, by applying the patch below. There might be
>> opther places this should also be done, but this would be a start.
> No objection to the concept, but I wonder if the -f test should
> instead be a -r test (in the original place too, I guess).
>
>             

In the original place permissions are in effect tested in pg_regress.c.

I'll go ahead with -r.

cheers

andrew