From: | Bharath Rupireddy <bharath(dot)rupireddyforpostgres(at)gmail(dot)com> |
---|---|
To: | Fujii Masao <masao(dot)fujii(at)oss(dot)nttdata(dot)com> |
Cc: | Justin Pryzby <pryzby(at)telsasoft(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>, Euler Taveira <euler(at)eulerto(dot)com>, Tomas Vondra <tomas(dot)vondra(at)enterprisedb(dot)com> |
Subject: | Re: Reword docs of feature "Remove temporary files after backend crash" |
Date: | 2021-10-11 07:20:28 |
Message-ID: | CALj2ACVs1DidQBNUwUGWD5v+cgC52=Gg+E3pMOudirqhp1YyWA@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Mon, Oct 11, 2021 at 11:37 AM Fujii Masao
<masao(dot)fujii(at)oss(dot)nttdata(dot)com> wrote:
> On 2021/10/10 22:33, Bharath Rupireddy wrote:
> >>> IIUC, the "server crash" includes any backend, auxiliary process,
> >>> postmaster crash i.e. database cluster/instance crash. The commit
> >>> cd91de0d1 especially added the temp file cleanup support if any
> >>> backend or auxiliary process (except syslogger and stats collector)
>
> We should mention not only a backend and an auxiliary processe
> but also background worker? Because, per Glossary, background worker
> is neither a backend nor an auxiliary process. Instead,
> maybe it's better to use "child processes" or something rather than
> mentioning those three processes.
If we were to use child processes (a term the glossary doesn't
define), we might end up saying postmaster child process crash, that's
not enough. We have to say things like "child process (except startup,
syslogger and stats collector) crash." IMO, let's not introduce
another term for the processes, the glossary defines many kinds of
processes already.
> > Also, I see that the restart_after_crash and
> > remove_temp_files_after_crash descriptions in guc.c say "Remove
> > temporary files after backend crash.". I think we can also modify them
> > to "Remove temporary files after the backend or auxiliary process
> > (except startup, syslogger and stats collector) crash.
>
> I'm not sure if we really need this long log message.
> IMO it's enough to add that information in the docs.
IMO let's be clear here as well for consistency reasons. I've seen
some of the long descriptions for GUCs [1]. And it seems like we don't
have any tex
So, the text for remove_temp_files_after_crash will be : "Remove
temporary files after backend or auxiliary process (except startup,
syslogger and stats collector) or background worker crash."
and for restart_after_crash: "Reinitialize server after backend crash
or auxiliary process (except startup, syslogger and stats collector)
or background worker crash."
I noticed another thing that the remove_temp_files_after_crash is
categorized as DEVELOPER_OPTIONS, shouldn't it be under
RESOURCES_DISK?
[1]
gettext_noop("Sets whether a WAL receiver should create a temporary
replication slot if no permanent slot is configured."),
gettext_noop("Writes full pages to WAL when first modified after a
checkpoint, even for a non-critical modification.")
gettext_noop("Enables backward compatibility mode for privilege checks
on large objects.")
gettext_noop("Forces a switch to the next WAL file if a "
"new file has not been started within N seconds."),
Regards,
Bharath Rupireddy.
From | Date | Subject | |
---|---|---|---|
Next Message | osumi.takamichi@fujitsu.com | 2021-10-11 07:27:43 | RE: Skipping logical replication transactions on subscriber side |
Previous Message | houzj.fnst@fujitsu.com | 2021-10-11 07:20:05 | RE: Added schema level support for publication. |