On 2021/10/01 15:27, Michael Paquier wrote:
On Wed, Sep 29, 2021 at 10:11:53PM +0900, Fujii Masao wrote:
BTW, when logfile fails to be opened, pgbench gets stuck due to commit
aeb57af8e6. So even if we decided not to back-patch those changes,
we should improve the handling of logfile open failure, to fix the issue.
There is an entry in the CF for this thread:
https://commitfest.postgresql.org/34/3219/
I have moved that to the next one as some pieces are missing. If you
are planning to handle the rest, could you register your name as a
committer?
Thanks for letting me know that!
I registered myself as a committer of the patch again.
pg_time_usec_t conn_duration; /* cumulated connection and deconnection
*
delays */
BTW, while reading the patch, I found the above comment in pgbench.c.
"deconnection" seems a valid word in French (?), but isn't it better to
replace it with "disconnection"? Patch attached.
Regards,
--
Fujii Masao
Advanced Computing Technology Center
Research and Development Headquarters
NTT DATA CORPORATION
diff --git a/src/bin/pgbench/pgbench.c b/src/bin/pgbench/pgbench.c
index d6b31bbf53..81693e1765 100644
--- a/src/bin/pgbench/pgbench.c
+++ b/src/bin/pgbench/pgbench.c
@@ -509,7 +509,7 @@ typedef struct
pg_time_usec_t create_time; /* thread creation time */
pg_time_usec_t started_time; /* thread is running */
pg_time_usec_t bench_start; /* thread is benchmarking */
- pg_time_usec_t conn_duration; /* cumulated connection and deconnection
+ pg_time_usec_t conn_duration; /* cumulated connection and
disconnection
*
delays */
StatsData stats;
