rdelaage
/
postgresql
mirror of https://git.postgresql.org/git/postgresql.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
postgresql/contrib/sepgsql/expected/alter.out

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

317 lines
37 KiB
Plaintext
Raw Normal View History

sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
--
-- Test for various ALTER statements
--
-- clean-up in case a prior regression run failed
SET client_min_messages TO 'warning';
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
DROP DATABASE IF EXISTS sepgsql_test_regression_1;
DROP DATABASE IF EXISTS sepgsql_test_regression;
DROP USER IF EXISTS regress_sepgsql_test_user;
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
RESET client_min_messages;
SELECT sepgsql_getcon(); -- confirm client privilege
Fix sepgsql regression tests. The regression tests for sepgsql were broken by changes in the base distro as-shipped policies. Specifically, definition of unconfined_t in the system default policy was changed to bypass multi-category rules, which the regression test depended on. Fix that by defining a custom privileged domain (sepgsql_regtest_superuser_t) and using it instead of system's unconfined_t domain. The new sepgsql_regtest_superuser_t domain performs almost like the current unconfined_t, but restricted by multi-category policy as the traditional unconfined_t was. The custom policy module is a self defined domain, and so should not be affected by related future system policy changes. However, it still uses the unconfined_u:unconfined_r pair for selinux-user and role. Those definitions have not been changed for several years and seem less risky to rely on than the unconfined_t domain. Additionally, if we define custom user/role, they would need to be manually defined at the operating system level, adding more complexity to an already non-standard and complex regression test. Back-patch to 9.3. The regression tests will need more work before working correctly on 9.2. Starting with 9.2, sepgsql has had dependencies on libselinux versions that are only available on newer distros with the changed set of policies (e.g. RHEL 7.x). On 9.1 sepgsql works fine with the older distros with original policy set (e.g. RHEL 6.x), and on which the existing regression tests work fine. We might want eventually change 9.1 sepgsql regression tests to be more independent from the underlying OS policies, however more work will be needed to make that happen and it is not clear that it is worth the effort. Kohei KaiGai with review by Adam Brightwell and me, commentary by Stephen, Alvaro, Tom, Robert, and others.
2015-08-30 20:09:05 +02:00
sepgsql_getcon
----------------------------------------------------------
unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
(1 row)
--
-- CREATE Objects to be altered (with debug_audit being silent)
--
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
CREATE DATABASE sepgsql_test_regression_1;
CREATE USER regress_sepgsql_test_user;
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
CREATE SCHEMA regtest_schema_1;
CREATE SCHEMA regtest_schema_2;
GRANT ALL ON SCHEMA regtest_schema_1 TO public;
GRANT ALL ON SCHEMA regtest_schema_2 TO public;
SET search_path = regtest_schema_1, regtest_schema_2, public;
CREATE TABLE regtest_table_1 (a int, b text);
CREATE TABLE regtest_table_2 (c text) inherits (regtest_table_1);
CREATE TABLE regtest_table_3 (x int primary key, y text);
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
---
-- partitioned table parent
CREATE TABLE regtest_ptable_1 (o int, p text) PARTITION BY RANGE (o);
-- partitioned table children
CREATE TABLE regtest_ptable_1_ones PARTITION OF regtest_ptable_1 FOR VALUES FROM ('0') TO ('10');
CREATE TABLE regtest_ptable_1_tens PARTITION OF regtest_ptable_1 FOR VALUES FROM ('10') TO ('100');
---
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
CREATE SEQUENCE regtest_seq_1;
CREATE VIEW regtest_view_1 AS SELECT * FROM regtest_table_1 WHERE a > 0;
CREATE FUNCTION regtest_func_1 (text) RETURNS bool
AS 'BEGIN RETURN true; END' LANGUAGE 'plpgsql';
-- switch on debug_audit
SET sepgsql.debug_audit = true;
SET client_min_messages = LOG;
--
-- ALTER xxx OWNER TO
--
-- XXX: It should take db_xxx:{setattr} permission checks even if
-- owner is not actually changed.
--
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
ALTER DATABASE sepgsql_test_regression_1 OWNER TO regress_sepgsql_test_user;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_db_t:s0 tclass=db_database name="sepgsql_test_regression_1" permissive=0
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
ALTER DATABASE sepgsql_test_regression_1 OWNER TO regress_sepgsql_test_user;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_db_t:s0 tclass=db_database name="sepgsql_test_regression_1" permissive=0
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
ALTER SCHEMA regtest_schema_1 OWNER TO regress_sepgsql_test_user;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_1" permissive=0
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
ALTER SCHEMA regtest_schema_1 OWNER TO regress_sepgsql_test_user;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_1" permissive=0
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
ALTER TABLE regtest_table_1 OWNER TO regress_sepgsql_test_user;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_1" permissive=0
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_2" permissive=0
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="public" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_1.regtest_table_1" permissive=0
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
ALTER TABLE regtest_table_1 OWNER TO regress_sepgsql_test_user;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_1.regtest_table_1" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_ptable_1 OWNER TO regress_sepgsql_test_user;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_1.regtest_ptable_1" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_ptable_1_ones OWNER TO regress_sepgsql_test_user;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_1.regtest_ptable_1_ones" permissive=0
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
ALTER SEQUENCE regtest_seq_1 OWNER TO regress_sepgsql_test_user;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_seq_t:s0 tclass=db_sequence name="regtest_schema_1.regtest_seq_1" permissive=0
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
ALTER SEQUENCE regtest_seq_1 OWNER TO regress_sepgsql_test_user;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_seq_t:s0 tclass=db_sequence name="regtest_schema_1.regtest_seq_1" permissive=0
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
ALTER VIEW regtest_view_1 OWNER TO regress_sepgsql_test_user;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_view_t:s0 tclass=db_view name="regtest_schema_1.regtest_view_1" permissive=0
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
ALTER VIEW regtest_view_1 OWNER TO regress_sepgsql_test_user;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_view_t:s0 tclass=db_view name="regtest_schema_1.regtest_view_1" permissive=0
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
ALTER FUNCTION regtest_func_1(text) OWNER TO regress_sepgsql_test_user;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_proc_exec_t:s0 tclass=db_procedure name="regtest_schema_1.regtest_func_1(pg_catalog.text)" permissive=0
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
ALTER FUNCTION regtest_func_1(text) OWNER TO regress_sepgsql_test_user;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_proc_exec_t:s0 tclass=db_procedure name="regtest_schema_1.regtest_func_1(pg_catalog.text)" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
--
-- ALTER xxx SET SCHEMA
--
ALTER TABLE regtest_table_1 SET SCHEMA regtest_schema_2;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { remove_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_1" permissive=0
LOG: SELinux: allowed { add_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_2" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_1.regtest_table_1" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_ptable_1 SET SCHEMA regtest_schema_2;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { remove_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_1" permissive=0
LOG: SELinux: allowed { add_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_2" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_1.regtest_ptable_1" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_ptable_1_ones SET SCHEMA regtest_schema_2;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { remove_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_1" permissive=0
LOG: SELinux: allowed { add_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_2" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_1.regtest_ptable_1_ones" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER SEQUENCE regtest_seq_1 SET SCHEMA regtest_schema_2;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { remove_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_1" permissive=0
LOG: SELinux: allowed { add_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_2" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_seq_t:s0 tclass=db_sequence name="regtest_schema_1.regtest_seq_1" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER VIEW regtest_view_1 SET SCHEMA regtest_schema_2;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { remove_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_1" permissive=0
LOG: SELinux: allowed { add_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_2" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_view_t:s0 tclass=db_view name="regtest_schema_1.regtest_view_1" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER FUNCTION regtest_func_1(text) SET SCHEMA regtest_schema_2;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { remove_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_1" permissive=0
LOG: SELinux: allowed { add_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_1" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_proc_exec_t:s0 tclass=db_procedure name="regtest_schema_1.regtest_func_1(pg_catalog.text)" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
--
-- ALTER xxx RENAME TO
--
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
ALTER DATABASE sepgsql_test_regression_1 RENAME TO sepgsql_test_regression;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_db_t:s0 tclass=db_database name="sepgsql_test_regression_1" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER SCHEMA regtest_schema_1 RENAME TO regtest_schema;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_1" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER TABLE regtest_table_1 RENAME TO regtest_table;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_2" permissive=0
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="public" permissive=0
LOG: SELinux: allowed { add_name remove_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_2" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_2.regtest_table_1" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
---
-- partitioned table parent
ALTER TABLE regtest_ptable_1 RENAME TO regtest_ptable;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { add_name remove_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_2" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_2.regtest_ptable_1" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
-- partitioned table child
ALTER TABLE regtest_ptable_1_ones RENAME TO regtest_table_part;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { add_name remove_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_2" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_2.regtest_ptable_1_ones" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
---
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER SEQUENCE regtest_seq_1 RENAME TO regtest_seq;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { add_name remove_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_2" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_seq_t:s0 tclass=db_sequence name="regtest_schema_2.regtest_seq_1" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER VIEW regtest_view_1 RENAME TO regtest_view;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { add_name remove_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_2" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_view_t:s0 tclass=db_view name="regtest_schema_2.regtest_view_1" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER FUNCTION regtest_func_1(text) RENAME TO regtest_func;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { add_name remove_name } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_2" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_proc_exec_t:s0 tclass=db_procedure name="regtest_schema_2.regtest_func_1(pg_catalog.text)" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
SET search_path = regtest_schema, regtest_schema_2, public;
--
-- misc ALTER commands
--
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
ALTER DATABASE sepgsql_test_regression CONNECTION LIMIT 999;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_db_t:s0 tclass=db_database name="sepgsql_test_regression" permissive=0
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
ALTER DATABASE sepgsql_test_regression SET search_path TO regtest_schema, public; -- not supported yet
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER TABLE regtest_table ADD COLUMN d float;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema" permissive=0
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_2" permissive=0
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="public" permissive=0
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="pg_catalog" permissive=0
Fix sepgsql regression tests. The regression tests for sepgsql were broken by changes in the base distro as-shipped policies. Specifically, definition of unconfined_t in the system default policy was changed to bypass multi-category rules, which the regression test depended on. Fix that by defining a custom privileged domain (sepgsql_regtest_superuser_t) and using it instead of system's unconfined_t domain. The new sepgsql_regtest_superuser_t domain performs almost like the current unconfined_t, but restricted by multi-category policy as the traditional unconfined_t was. The custom policy module is a self defined domain, and so should not be affected by related future system policy changes. However, it still uses the unconfined_u:unconfined_r pair for selinux-user and role. Those definitions have not been changed for several years and seem less risky to rely on than the unconfined_t domain. Additionally, if we define custom user/role, they would need to be manually defined at the operating system level, adding more complexity to an already non-standard and complex regression test. Back-patch to 9.3. The regression tests will need more work before working correctly on 9.2. Starting with 9.2, sepgsql has had dependencies on libselinux versions that are only available on newer distros with the changed set of policies (e.g. RHEL 7.x). On 9.1 sepgsql works fine with the older distros with original policy set (e.g. RHEL 6.x), and on which the existing regression tests work fine. We might want eventually change 9.1 sepgsql regression tests to be more independent from the underlying OS policies, however more work will be needed to make that happen and it is not clear that it is worth the effort. Kohei KaiGai with review by Adam Brightwell and me, commentary by Stephen, Alvaro, Tom, Robert, and others.
2015-08-30 20:09:05 +02:00
LINE 1: ALTER TABLE regtest_table ADD COLUMN d float;
^
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="pg_catalog" permissive=0
LOG: SELinux: allowed { create } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table.d" permissive=0
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="pg_catalog" permissive=0
LOG: SELinux: allowed { create } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema.regtest_table_2.d" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER TABLE regtest_table DROP COLUMN d;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { drop } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema.regtest_table_2.d" permissive=0
LOG: SELinux: allowed { drop } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table.d" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER TABLE regtest_table ALTER b SET DEFAULT 'abcd'; -- not supported yet
ALTER TABLE regtest_table ALTER b SET DEFAULT 'XYZ'; -- not supported yet
ALTER TABLE regtest_table ALTER b DROP DEFAULT; -- not supported yet
ALTER TABLE regtest_table ALTER b SET NOT NULL;
Revert "Adjust contrib/sepgsql regression test expected outputs." This reverts commit 76c111a7f166; should have been included in 9ce04b50e120. Noted by Joe Conway
2023-04-12 20:22:17 +02:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table.b" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema.regtest_table_2.b" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER TABLE regtest_table ALTER b DROP NOT NULL;
Revert "Adjust contrib/sepgsql regression test expected outputs." This reverts commit 76c111a7f166; should have been included in 9ce04b50e120. Noted by Joe Conway
2023-04-12 20:22:17 +02:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table.b" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema.regtest_table_2.b" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER TABLE regtest_table ALTER b SET STATISTICS -1;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table.b" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema.regtest_table_2.b" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER TABLE regtest_table ALTER b SET (n_distinct = 999);
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table.b" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER TABLE regtest_table ALTER b SET STORAGE PLAIN;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table.b" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema.regtest_table_2.b" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER TABLE regtest_table ADD CONSTRAINT test_fk FOREIGN KEY (a) REFERENCES regtest_table_3(x); -- not supported
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="pg_catalog" permissive=0
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="pg_catalog" permissive=0
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="pg_catalog" permissive=0
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="pg_catalog" permissive=0
LOG: SELinux: allowed { select } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema.regtest_table_3" permissive=0
LOG: SELinux: allowed { select } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="column x of table regtest_table_3" permissive=0
Fix sepgsql expected test output Commit f75cec4fff87 changed the order in which the relations are permission-checked in RI_Initial_Check, which the sepgsql test is sensitive to. Adapt. Discussion: https://postgr.es/m/3468125.1683238309@sss.pgh.pa.us
2023-05-05 11:10:25 +02:00
LOG: SELinux: allowed { select } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_2.regtest_table" permissive=0
LOG: SELinux: allowed { select } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="column a of table regtest_table" permissive=0
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema_2" permissive=0
sepgsql: Enforce db_schema:search permission. KaiGai Kohei, with comment and doc wordsmithing by me
2013-04-05 14:51:31 +02:00
LINE 1: SELECT fk."a" FROM ONLY "regtest_schema_2"."regtest_table" f...
^
QUERY: SELECT fk."a" FROM ONLY "regtest_schema_2"."regtest_table" fk LEFT OUTER JOIN ONLY "regtest_schema"."regtest_table_3" pk ON ( pk."x" OPERATOR(pg_catalog.=) fk."a") WHERE pk."x" IS NULL AND (fk."a" IS NOT NULL)
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="regtest_schema" permissive=0
sepgsql: Enforce db_schema:search permission. KaiGai Kohei, with comment and doc wordsmithing by me
2013-04-05 14:51:31 +02:00
LINE 1: ...schema_2"."regtest_table" fk LEFT OUTER JOIN ONLY "regtest_s...
^
QUERY: SELECT fk."a" FROM ONLY "regtest_schema_2"."regtest_table" fk LEFT OUTER JOIN ONLY "regtest_schema"."regtest_table_3" pk ON ( pk."x" OPERATOR(pg_catalog.=) fk."a") WHERE pk."x" IS NULL AND (fk."a" IS NOT NULL)
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="pg_catalog" permissive=0
Fix sepgsql regression tests. The regression tests for sepgsql were broken by changes in the base distro as-shipped policies. Specifically, definition of unconfined_t in the system default policy was changed to bypass multi-category rules, which the regression test depended on. Fix that by defining a custom privileged domain (sepgsql_regtest_superuser_t) and using it instead of system's unconfined_t domain. The new sepgsql_regtest_superuser_t domain performs almost like the current unconfined_t, but restricted by multi-category policy as the traditional unconfined_t was. The custom policy module is a self defined domain, and so should not be affected by related future system policy changes. However, it still uses the unconfined_u:unconfined_r pair for selinux-user and role. Those definitions have not been changed for several years and seem less risky to rely on than the unconfined_t domain. Additionally, if we define custom user/role, they would need to be manually defined at the operating system level, adding more complexity to an already non-standard and complex regression test. Back-patch to 9.3. The regression tests will need more work before working correctly on 9.2. Starting with 9.2, sepgsql has had dependencies on libselinux versions that are only available on newer distros with the changed set of policies (e.g. RHEL 7.x). On 9.1 sepgsql works fine with the older distros with original policy set (e.g. RHEL 6.x), and on which the existing regression tests work fine. We might want eventually change 9.1 sepgsql regression tests to be more independent from the underlying OS policies, however more work will be needed to make that happen and it is not clear that it is worth the effort. Kohei KaiGai with review by Adam Brightwell and me, commentary by Stephen, Alvaro, Tom, Robert, and others.
2015-08-30 20:09:05 +02:00
LINE 1: ..."regtest_schema"."regtest_table_3" pk ON ( pk."x" OPERATOR(p...
^
QUERY: SELECT fk."a" FROM ONLY "regtest_schema_2"."regtest_table" fk LEFT OUTER JOIN ONLY "regtest_schema"."regtest_table_3" pk ON ( pk."x" OPERATOR(pg_catalog.=) fk."a") WHERE pk."x" IS NULL AND (fk."a" IS NOT NULL)
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="pg_catalog" permissive=0
LOG: SELinux: allowed { select } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_2.regtest_table" permissive=0
LOG: SELinux: allowed { select } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="column a of table regtest_table" permissive=0
LOG: SELinux: allowed { select } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema.regtest_table_3" permissive=0
LOG: SELinux: allowed { select } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="column x of table regtest_table_3" permissive=0
LOG: SELinux: allowed { execute } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_proc_exec_t:s0 tclass=db_procedure name="pg_catalog.int4eq(integer,integer)" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER TABLE regtest_table ADD CONSTRAINT test_ck CHECK (b like '%abc%') NOT VALID; -- not supported
ALTER TABLE regtest_table VALIDATE CONSTRAINT test_ck; -- not supported
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { execute } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_proc_exec_t:s0 tclass=db_procedure name="pg_catalog.textlike(pg_catalog.text,pg_catalog.text)" permissive=0
LOG: SELinux: allowed { execute } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_proc_exec_t:s0 tclass=db_procedure name="pg_catalog.textlike(pg_catalog.text,pg_catalog.text)" permissive=0
Blindly attempt to fix sepgsql tests #2.
2017-03-26 05:54:23 +02:00
ALTER TABLE regtest_table DROP CONSTRAINT test_ck; -- not supported
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
CREATE TRIGGER regtest_test_trig BEFORE UPDATE ON regtest_table
FOR EACH ROW EXECUTE PROCEDURE suppress_redundant_updates_trigger();
ALTER TABLE regtest_table DISABLE TRIGGER regtest_test_trig; -- not supported
ALTER TABLE regtest_table ENABLE TRIGGER regtest_test_trig; -- not supported
CREATE RULE regtest_test_rule AS ON INSERT TO regtest_table_3 DO ALSO NOTHING;
ALTER TABLE regtest_table_3 DISABLE RULE regtest_test_rule; -- not supported
ALTER TABLE regtest_table_3 ENABLE RULE regtest_test_rule; -- not supported
Blind attempt at fixing sepgsql output for 578b22.
2018-11-21 06:02:38 +01:00
ALTER TABLE regtest_table SET (fillfactor = 75);
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_2.regtest_table" permissive=0
Blind attempt at fixing sepgsql output for 578b22.
2018-11-21 06:02:38 +01:00
ALTER TABLE regtest_table RESET (fillfactor);
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_2.regtest_table" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER TABLE regtest_table_2 NO INHERIT regtest_table; -- not supported
ALTER TABLE regtest_table_2 INHERIT regtest_table; -- not supported
ALTER TABLE regtest_table SET TABLESPACE pg_default;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_2.regtest_table" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
---
-- partitioned table parent
ALTER TABLE regtest_ptable ADD COLUMN d float;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="pg_catalog" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
LINE 1: ALTER TABLE regtest_ptable ADD COLUMN d float;
^
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="pg_catalog" permissive=0
LOG: SELinux: allowed { create } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_ptable.d" permissive=0
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="pg_catalog" permissive=0
LOG: SELinux: allowed { create } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table_part.d" permissive=0
LOG: SELinux: allowed { search } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_schema_t:s0 tclass=db_schema name="pg_catalog" permissive=0
LOG: SELinux: allowed { create } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema.regtest_ptable_1_tens.d" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_ptable DROP COLUMN d;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { drop } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table_part.d" permissive=0
LOG: SELinux: allowed { drop } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema.regtest_ptable_1_tens.d" permissive=0
LOG: SELinux: allowed { drop } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_ptable.d" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_ptable ALTER p SET DEFAULT 'abcd'; -- not supported by sepgsql
ALTER TABLE regtest_ptable ALTER p SET DEFAULT 'XYZ'; -- not supported by sepgsql
ALTER TABLE regtest_ptable ALTER p DROP DEFAULT; -- not supported by sepgsql
ALTER TABLE regtest_ptable ALTER p SET NOT NULL;
Revert "Adjust contrib/sepgsql regression test expected outputs." This reverts commit 76c111a7f166; should have been included in 9ce04b50e120. Noted by Joe Conway
2023-04-12 20:22:17 +02:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_ptable.p" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table_part.p" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema.regtest_ptable_1_tens.p" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_ptable ALTER p DROP NOT NULL;
Revert "Adjust contrib/sepgsql regression test expected outputs." This reverts commit 76c111a7f166; should have been included in 9ce04b50e120. Noted by Joe Conway
2023-04-12 20:22:17 +02:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_ptable.p" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table_part.p" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema.regtest_ptable_1_tens.p" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_ptable ALTER p SET STATISTICS -1;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_ptable.p" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table_part.p" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema.regtest_ptable_1_tens.p" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_ptable ALTER p SET (n_distinct = 999);
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_ptable.p" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_ptable ALTER p SET STORAGE PLAIN;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_ptable.p" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table_part.p" permissive=0
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema.regtest_ptable_1_tens.p" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_ptable ADD CONSTRAINT test_ck CHECK (p like '%abc%') NOT VALID; -- not supported by sepgsql
ALTER TABLE regtest_ptable DROP CONSTRAINT test_ck; -- not supported by sepgsql
ALTER TABLE regtest_ptable SET TABLESPACE pg_default;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_2.regtest_ptable" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
-- partitioned table child
ALTER TABLE regtest_table_part ALTER p SET DEFAULT 'abcd'; -- not supported by sepgsql
ALTER TABLE regtest_table_part ALTER p SET DEFAULT 'XYZ'; -- not supported by sepgsql
ALTER TABLE regtest_table_part ALTER p DROP DEFAULT; -- not supported by sepgsql
ALTER TABLE regtest_table_part ALTER p SET NOT NULL;
Revert "Adjust contrib/sepgsql regression test expected outputs." This reverts commit 76c111a7f166; should have been included in 9ce04b50e120. Noted by Joe Conway
2023-04-12 20:22:17 +02:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table_part.p" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_table_part ALTER p DROP NOT NULL;
Revert "Adjust contrib/sepgsql regression test expected outputs." This reverts commit 76c111a7f166; should have been included in 9ce04b50e120. Noted by Joe Conway
2023-04-12 20:22:17 +02:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table_part.p" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_table_part ALTER p SET STATISTICS -1;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table_part.p" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_table_part ALTER p SET (n_distinct = 999);
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table_part.p" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_table_part ALTER p SET STORAGE PLAIN;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_column name="regtest_schema_2.regtest_table_part.p" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_table_part ADD CONSTRAINT test_ck CHECK (p like '%abc%') NOT VALID; -- not supported by sepgsql
ALTER TABLE regtest_table_part VALIDATE CONSTRAINT test_ck; -- not supported by sepgsql
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { execute } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=system_u:object_r:sepgsql_proc_exec_t:s0 tclass=db_procedure name="pg_catalog.textlike(pg_catalog.text,pg_catalog.text)" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_table_part DROP CONSTRAINT test_ck; -- not supported by sepgsql
CREATE TRIGGER regtest_part_test_trig BEFORE UPDATE ON regtest_table_part
FOR EACH ROW EXECUTE PROCEDURE suppress_redundant_updates_trigger();
ALTER TABLE regtest_table_part DISABLE TRIGGER regtest_part_test_trig; -- not supported by sepgsql
ALTER TABLE regtest_table_part ENABLE TRIGGER regtest_part_test_trig; -- not supported by sepgsql
ALTER TABLE regtest_table_part SET (fillfactor = 75);
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_2.regtest_table_part" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_table_part RESET (fillfactor);
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_2.regtest_table_part" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
ALTER TABLE regtest_table_part SET TABLESPACE pg_default;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_table_t:s0 tclass=db_table name="regtest_schema_2.regtest_table_part" permissive=0
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
---
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER VIEW regtest_view SET (security_barrier);
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_view_t:s0 tclass=db_view name="regtest_schema_2.regtest_view" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
ALTER SEQUENCE regtest_seq INCREMENT BY 10 START WITH 1000;
Include permissive/enforcing state in sepgsql log messages. SELinux itself does this (at least in modern releases), and it seems like a good idea to reduce confusion. Dave Page Discussion: https://postgr.es/m/CA+OCxowsQoLEYc=jN7OtNvOdX0Jg5L7nMYt++=k0X78HGq-sXg@mail.gmail.com
2022-01-12 20:23:13 +01:00
LOG: SELinux: allowed { setattr } scontext=unconfined_u:unconfined_r:sepgsql_regtest_superuser_t:s0 tcontext=unconfined_u:object_r:sepgsql_seq_t:s0 tclass=db_sequence name="regtest_schema_2.regtest_seq" permissive=0
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
--
-- clean-up objects
--
RESET sepgsql.debug_audit;
RESET client_min_messages;
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
DROP DATABASE sepgsql_test_regression;
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
DROP SCHEMA regtest_schema CASCADE;
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
NOTICE: drop cascades to 4 other objects
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
DETAIL: drop cascades to table regtest_table_2
drop cascades to table regtest_table_3
drop cascades to constraint test_fk on table regtest_table
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
drop cascades to table regtest_ptable_1_tens
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
DROP SCHEMA regtest_schema_2 CASCADE;
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
NOTICE: drop cascades to 5 other objects
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
DETAIL: drop cascades to table regtest_table
Add partitioned table support to sepgsql The new partitioned table capability added a new relkind, namely RELKIND_PARTITIONED_TABLE. Update sepgsql to treat this new relkind exactly the same way it does RELKIND_RELATION. In addition, add regression test coverage for partitioned tables. Issue raised by Stephen Frost and initial patch by Mike Palmiotto. Review by Tom Lane and Robert Haas, and editorializing by me. Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
2017-04-09 23:01:58 +02:00
drop cascades to table regtest_ptable
sepgsql: Support for new post-ALTER access hook. KaiGai Kohei
2013-03-27 13:10:14 +01:00
drop cascades to sequence regtest_seq
drop cascades to view regtest_view
drop cascades to function regtest_func(text)
Establish conventions about global object names used in regression tests. To ensure that "make installcheck" can be used safely against an existing installation, we need to be careful about what global object names (database, role, and tablespace names) we use; otherwise we might accidentally clobber important objects. There's been a weak consensus that test databases should have names including "regression", and that test role names should start with "regress_", but we didn't have any particular rule about tablespace names; and neither of the other rules was followed with any consistency either. This commit moves us a long way towards having a hard-and-fast rule that regression test databases must have names including "regression", and that test role and tablespace names must start with "regress_". It's not completely there because I did not touch some test cases in rolenames.sql that test creation of special role names like "session_user". That will require some rethinking of exactly what we want to test, whereas the intent of this patch is just to hit all the cases in which the needed renamings are cosmetic. There is no enforcement mechanism in this patch either, but if we don't add one we can expect that the tests will soon be violating the convention again. Again, that's not such a cosmetic change and it will require discussion. (But I did use a quick-hack enforcement patch to find these cases.) Discussion: <16638.1468620817@sss.pgh.pa.us>
2016-07-18 00:42:31 +02:00
DROP USER regress_sepgsql_test_user;