| From: | Andrew Gierth <andrew(at)tao11(dot)riddles(dot)org(dot)uk> | 
|---|---|
| To: | Lars Vonk <lars(dot)vonk(at)gmail(dot)com> | 
| Cc: | pgsql-bugs(at)lists(dot)postgresql(dot)org | 
| Subject: | Re: BUG #15074: psql client never returns when creating index (long running operation) | 
| Date: | 2018-02-20 18:35:18 | 
| Message-ID: | 87sh9vo714.fsf@news-spur.riddles.org.uk | 
| Views: | Whole Thread | Raw Message | Download mbox | Resend email | 
| Thread: | |
| Lists: | pgsql-bugs | 
>>>>> "Lars" == Lars Vonk <lars(dot)vonk(at)gmail(dot)com> writes:
 Lars> Hi,
 Lars> Thanks for replying. I will double check the network settings,
 Lars> but couldn't find anything.
I've seen this so many times that I'd bet a fair sum on it.
 Lars> Are there any psql / postgres settings I can check, I couldn't
 Lars> find any that might be affecting this.
The only relevant pg settings are the tcp_keepalives_* ones, which (on
operating systems that support this, i.e. not Windows) cause pg to
request a different keepalive timeout from the OS's default. Note that
OS defaults themselves default to a bit over 2 hours unless you
explicitly configure something else.
Likewise on the client side, if you are using an OS that supports it,
you can use keepalives_idle= in the connection string.
-- 
Andrew (irc:RhodiumToad)
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Claudio Freire | 2018-02-20 21:13:26 | pg_upgrade and materialized views | 
| Previous Message | Lars Vonk | 2018-02-20 17:57:47 | Re: BUG #15074: psql client never returns when creating index (long running operation) |