Rafia, Sergey,
Many thanks for working on this!
> I have incorporated most of the suggestions into the patch. I have also rebased and tested the patch on top of the
currentmaster
I noticed that this patch is marked as "Needs Review" and decided to
take a look.
I believe there is a bug in the implementation. Here is what I did:
```
57033 (master) =# select * from pg_stat_activity where pid = 57033;
...
total_active_time | 9.128
total_idle_in_transaction_time | 0
57033 (master) =# select * from pg_stat_activity where pid = 57033;
...
total_active_time | 10.626
total_idle_in_transaction_time | 0
57033 (master) =# BEGIN;
57033 (master) =# select * from pg_stat_activity where pid = 57033;
...
total_active_time | 17.443
total_idle_in_transaction_time | 2314.703
57033 (master) =# select * from pg_stat_activity where pid = 57033;
...
total_active_time | 2514.635
total_idle_in_transaction_time | 2314.703
57033 (master) =# COMMIT;
57033 (master) =# select * from pg_stat_activity where pid = 57033;
...
total_active_time | 22.048
total_idle_in_transaction_time | 7300.911
```
So it looks like total_active_time tracks seconds when a user executes
single expressions and milliseconds when running a transaction. It
should always track milliseconds.
Please use `git format-patch` for the next patch and provide a commit
message, as it was previously pointed out by Bharath. Please specify
the list of the authors and reviewers and add a note about
incrementing the catalog version.
--
Best regards,
Aleksander Alekseev