Libpq support to connect to standby server as priority

o the libpq. The 'read-only' server means Standby Server The 'read-write' server means Master Server support to 'read-write' -- Regards, Jing Wang Fujitsu Australia

Re: Libpq support to connect to standby server as priority

r not. If server is 'read-write' then it will try next server address. If all connections for 'read-only' get failed it will try to connect to the master server. -- Regards, Jing Wang Fujitsu Australia libpq_support_perfer-read_001.patch Description: Binary data

Re: Libpq support to connect to standby server as priority

Hi Takayuki, Thanks your reminder. I will have a look your patch and the review comments and update the patch soon. Please leave my current submitted patch now. -- Regards, Jing Wang Fujitsu Australia 2018-01-04 17:40 GMT+11:00 Tsunakawa, Takayuki < tsunakawa.ta...@jp.fujitsu.com>:

Re: [HACKERS] Support to COMMENT ON DATABASE CURRENT_DATABASE

ts for pg_dump are failing. The test case has been updated. >make makeDbSpec() return a DbSpec and then try to minimize the >forced-casting happening. Makes sense. It has been changed. Regards, Jing Wang Fujitsu Australia support_CURRENT_DATABASE_keyword_v4.6.patch Description: Binary data

Re: Libpq support to connect to standby server as priority

rg/15/1148/ ) which looks similar with this. But I would say this patch is more complex than my proposal. It is better separate these 2 patches to consider. Regards, Jing Wang Fujitsu Australia libpq_support_perfer-read_002.patch Description: Binary data

Re: [HACKERS] Support to COMMENT ON DATABASE CURRENT_DATABASE

taddress.c, and user.c, so there's more work to do besides >this Yes. fixed it. Enclosed please find the latest patch fixed above. Regards, Jing Wang Fujitsu Australia support_CURRENT_DATABASE_keyword_v4.7.patch Description: Binary data

Re: [HACKERS] WIP: Separate log file for extension

It may be better to be as following: int file_mode = log_streams[stream_id].file_mode -- Regards, Jing Wang Fujitsu Australia

Re: [HACKERS] Support to COMMENT ON DATABASE CURRENT_DATABASE

mentioned the database_name are also present in the GRANT/REVOKE/ALTER ROLE, so a patch will be present later for supporting CURRENT_DATABASE on these SQL statements. Regards, Jing Wang Fujitsu Australia comment_on_current_database_no_pgdump_v4.4.patch Description: Binary data

Re: [HACKERS] Support to COMMENT ON DATABASE CURRENT_DATABASE

OLE ... IN DATABASE CURRENT_DATABASE SET/RESET configuration_parameter GRANT ... ON DATABASE CURRENT_DATABASE TO role_specification ... REVOKE ... ON DATABASE CURRENT_DATABASE FROM ... Regards, Jing Wang Fujitsu Australia current_database_on_grant_revoke_role_v4.4.patch Description: Binary data

Re: [HACKERS] [PATCH] Vacuum: Update FSM more frequently

vacuumed_pages_at_fsm_vac = vacuumed_pages; + } vacuumed_pages_at_fsm_vac is initialised with value zero and seems no chance to go into the bracket. Regards, Jing Wang Fujitsu Australia

Re: [HACKERS] Support to COMMENT ON DATABASE CURRENT_DATABASE

N DATABASE CURRENT_DATABASE 8. ALTER ROLE ... IN DATABASE CURRENT_DATABASE SET/RESET configuration_parameter 9. GRANT ... ON DATABASE CURRENT_DATABASE TO role_specification ... 10. REVOKE ... ON DATABASE CURRENT_DATABASE FROM ... Regards, Jing Wang Fujitsu