Re: Nicer error when connecting to standby with hot_standby=off

On 2021/03/24 16:59, Alvaro Herrera wrote:
> On 2021-Mar-24, Fujii Masao wrote:
>
>> On 2021/03/24 5:59, Tom Lane wrote:
>>> Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org> writes:
>>>> FATAL: the database system is starting up
>>>> DETAIL: WAL is being applied to recover from a system crash.
>>>> or
>>>> DETAIL: The system is applying WAL to recover from a system crash.
>>>> or
>>>> DETAIL: The startup process is applying WAL to recover from a system crash.
>>>
>>> I don't think the postmaster has enough context to know if that's
>>> actually true. It just launches the startup process and waits for
>>> results. If somebody saw this during a normal (non-crash) startup,
>>> they'd be justifiably alarmed.
>>
>> Yes, so logging "the database system is starting up" seems enough to me.
>
> No objection.

Thanks! So I changed the message reported at PM_STARTUP to that one,
based on v8 patch that James posted upthread. I also ran pgindent for
the patch. Attached is the updated version of the patch.

Barring any objection, I will commit this.

Regards,

--
Fujii Masao
Advanced Computing Technology Center
Research and Development Headquarters
NTT DATA CORPORATION