RE: A new function to wait for the backend exit after termination

Поиск
Список
Период
Сортировка
От Hou, Zhijie
Тема RE: A new function to wait for the backend exit after termination
Дата
Msg-id 2a63dccfc72b44e982627d625092908d@G08CNEXMBPEKD05.g08.fujitsu.local
обсуждение исходный текст
Ответ на Re: A new function to wait for the backend exit after termination  (Muhammad Usama <m.usama@gmail.com>)
Ответы Re: A new function to wait for the backend exit after termination  (Tom Lane <tgl@sss.pgh.pa.us>)
RE: A new function to wait for the backend exit after termination  ("Hou, Zhijie" <houzj.fnst@cn.fujitsu.com>)
Re: A new function to wait for the backend exit after termination  (Bharath Rupireddy <bharath.rupireddyforpostgres@gmail.com>)
Список pgsql-hackers
Hi

I take a look into the patch, and here some comments.

1.
+
+    ereport(WARNING,
+            (errmsg("could not wait for the termination of the backend with PID %d within %ld milliseconds",
+                    pid, timeout)));
+

The code use %ld to print int64 type.
How about use INT64_FORMAT, which looks more appropriate. 

2.
+    if (timeout <= 0)
+    {
+        ereport(WARNING,
+                (errmsg("timeout cannot be negative or zero: %ld", timeout)));
+        PG_RETURN_BOOL(r);
+    }

The same as 1.

3.
+pg_terminate_backend_and_wait(PG_FUNCTION_ARGS)
+{
+    int     pid = PG_GETARG_DATUM(0);

+pg_wait_backend(PG_FUNCTION_ARGS)
+{
+    int        pid = PG_GETARG_INT32(0);

The code use different macro to get pid,
How about use PG_GETARG_INT32(0) for each one.


I changed the status to 'wait on anthor'.
The others of the patch LGTM, 
I think it can be changed to Ready for Committer again, when this comment is confirmed.


Best regards,
houzj







В списке pgsql-hackers по дате отправления:

Предыдущее
От: Michael Paquier
Дата:
Сообщение: Re: Deprecate custom encoding conversions
Следующее
От: Tom Lane
Дата:
Сообщение: Re: A new function to wait for the backend exit after termination