Control: tags -1 unreproducable
Hi Santiago,
thanks for your bug report. I tried to reproduce the build problem
that you describe but I was not able to do so. Looking at your build logs
I can see that it is always test 247 that fails. Test 247 starts a postgresql
database server and then tries to read data from this database.
Looking at your log file the critical section is here:
===== Snippet from Santiagos failure log ====
## ------------------------ ##
## Summary of the failures. ##
## ------------------------ ##
Failed tests:
GNU PSPP 0.10.2 test suite test groups:
NUM: FILE-NAME:LINE TEST-GROUP-NAME
KEYWORDS
247: get-data-psql.at:124 GET DATA /TYPE=PSQL
## ---------------------- ##
## Detailed failed tests. ##
## ---------------------- ##
# -*- compilation -*-
247. get-data-psql.at:124: testing GET DATA /TYPE=PSQL ...
./get-data-psql.at:125: initdb -A trust
--- /dev/null 2017-01-05 19:44:44.564000000 +0000
+++ /<<PKGBUILDDIR>>/tests/testsuite.dir/at-groups/247/stderr 2017-01-06
02:09:40.476955912 +0000
@@ -0,0 +1 @@
+locale: Cannot set LC_MESSAGES to default locale: No such file or directory
stdout:
The files belonging to this database system will be owned by user "buildd".
This user must also own the server process.
The database cluster will be initialized with locale "C".
The default database encoding has accordingly been set to "SQL_ASCII".
The default text search configuration will be set to "english".
Data page checksums are disabled.
creating directory /<<PKGBUILDDIR>>/tests/testsuite.dir/0247/cluster ... ok
creating subdirectories ... ok
selecting default max_connections ... 100
selecting default shared_buffers ... 128MB
selecting dynamic shared memory implementation ... posix
creating configuration files ... ok
running bootstrap script ... ok
performing post-bootstrap initialization ... ok
syncing data to disk ... ok
Success. You can now start the database server using:
pg_ctl -D /<<PKGBUILDDIR>>/tests/testsuite.dir/0247/cluster -l logfile start
247. get-data-psql.at:124: 247. GET DATA /TYPE=PSQL (get-data-psql.at:124):
FAILED (get-data-psql.at:125)
====== End of snippet =====
My clue is that the problem is related to the line
+locale: Cannot set LC_MESSAGES to default locale: No such file or directory
so it would be a hint that there is some problem on your build
system with the locale mechanism.
If I build the pspp package from source as you did with
dpkg-buildpackage -B
and then run test 247 with verbose output it looks like this:
======== Snippet from Friedrich =======
fritz@debian:~/pspp/debian/pspp-0.10.2$ XTERM_LOCALE='' /bin/bash
'./tests/testsuite' 247 -v -C tests
AUTOTEST_PATH=tests/data:tests/language/lexer:tests/libpspp:tests/output:src/ui/terminal:utilities
RUNNER='' -j1
## --------------------------- ##
## GNU PSPP 0.10.2 test suite. ##
## --------------------------- ##
247. get-data-psql.at:124: testing GET DATA /TYPE=PSQL ...
./get-data-psql.at:125: initdb -A trust
stdout:
The files belonging to this database system will be owned by user "fritz".
This user must also own the server process.
The database cluster will be initialized with locale "C".
The default database encoding has accordingly been set to "SQL_ASCII".
The default text search configuration will be set to "english".
Data page checksums are disabled.
creating directory
/home/fritz/pspp/debian/pspp-0.10.2/tests/testsuite.dir/0247/cluster ... ok
creating subdirectories ... ok
selecting default max_connections ... 100
selecting default shared_buffers ... 128MB
selecting dynamic shared memory implementation ... posix
creating configuration files ... ok
running bootstrap script ... ok
performing post-bootstrap initialization ... ok
syncing data to disk ... ok
Success. You can now start the database server using:
pg_ctl -D
/home/fritz/pspp/debian/pspp-0.10.2/tests/testsuite.dir/0247/cluster -l logfile
start
./get-data-psql.at:125: pg_ctl start -w -o "-k $socket_dir -h ''"
stdout:
waiting for server to start....LOG: database system was shut down at
2017-01-08 10:50:11 CET
LOG: MultiXact member wraparound protections are now enabled
LOG: database system is ready to accept connections
LOG: autovacuum launcher started
done
server started
./get-data-psql.at:125: createdb -h "$socket_dir" -p $PG_PORT $PG_DBASE
stderr:
stdout:
./get-data-psql.at:125: psql -h "$socket_dir" -p $PG_PORT $PG_DBASE <
populate.sql
stdout:
CREATE TABLE
CREATE TABLE
INSERT 0 1000
CREATE TABLE
INSERT 0 1
INSERT 0 1
INSERT 0 1
./get-data-psql.at:128: cat > ordinary-query.sps <<EOF
GET DATA /TYPE=psql
/CONNECT="host=$socket_dir port=$PGPORT dbname=$PG_DBASE"
/UNENCRYPTED
/SQL="select * from thing".
DISPLAY DICTIONARY.
LIST.
EOF
./get-data-psql.at:139: pspp -o pspp.csv ordinary-query.sps
./get-data-psql.at:140: cat pspp.csv
./get-data-psql.at:173: cat > empty-result.sps <<EOF
GET DATA /TYPE=psql
/CONNECT="host=$socket_dir port=$PGPORT dbname=$PG_DBASE"
/UNENCRYPTED
/SQL="select * from empty".
DISPLAY DICTIONARY.
LIST.
EOF
./get-data-psql.at:184: pspp -o pspp.csv empty-result.sps
./get-data-psql.at:185: cat pspp.csv
./get-data-psql.at:193: cat > large-result.sps <<EOF
GET DATA /TYPE=psql
/CONNECT="host=$socket_dir port=$PGPORT dbname=$PG_DBASE"
/UNENCRYPTED
/SQL="select * from large".
NUMERIC diff.
COMPUTE diff = x - lag (x).
TEMPORARY.
SELECT IF (diff <> 1).
LIST.
TEMPORARY.
N OF CASES 6.
LIST.
SORT CASES BY x (D).
TEMPORARY.
N OF CASES 6.
LIST.
EOF
./get-data-psql.at:217: pspp -o pspp.csv large-result.sps
./get-data-psql.at:218: cat pspp.csv
./get-data-psql.at:250: psql -h "$socket_dir" -p $PG_PORT $PG_DBASE <
all-null-string.sql
stdout:
CREATE TABLE
INSERT 0 1
INSERT 0 1
./get-data-psql.at:253: cat > get-data.sps <<EOF
GET DATA /TYPE=psql
/CONNECT="host=$socket_dir port=$PGPORT dbname=$PG_DBASE"
/UNENCRYPTED
/SQL="select * from foo".
DISPLAY DICTIONARY.
LIST.
EOF
./get-data-psql.at:264: pspp -o pspp.csv get-data.sps
server shutting down
247. get-data-psql.at:124: ok
## ------------- ##
## Test results. ##
## ------------- ##
1 test was successful.
fritz@debian:~/pspp/debian/pspp-0.10.2$
======== End of snippet ===============
So the difference is the locale message at the very beginning of the
test.
I did the build in various ways:
a) dpgk-buildpackage -B
b) debuild
c) pdebuild
d) from git with make; make check
but I was not able to reproduce the problem. I did my tests in
debian sid in a vmware virtual machine on a Macbook. I have no
idea how I could investigate further into this problem.
Regards
Friedrich
> Am 06.01.2017 um 19:00 schrieb Santiago Vila <sanv...@debian.org>:
>
> Package: src:pspp
> Version: 0.10.2-1
> Severity: important
>
> Dear maintainer:
>
> I tried to build this package in stretch with "dpkg-buildpackage -A"
> but it failed:
>
> ------------------------------------------------------------------------------
>
> I've put several build logs here:
>
> https://people.debian.org/~sanvila/build-logs/pspp/
>
> If this is really a bug in one of the build-depends, please use reassign and
> affects,
> so that this is still visible in the page for this package.
>
> The bug should be reproducible with sbuild on a single CPU virtual machine,
> provided you try enough times (as the failure happens randomly).
>
> Thanks.
