Re: Corruption during WAL replay

Robert Haas <robertmhaas@gmail.com> writes:
> On Fri, Sep 24, 2021 at 3:42 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> I think the basic idea is about right, but I'm not happy with the
>> three-way delayChkpt business; that seems too cute by three-quarters.

> Nobody, but the version of the patch that I was looking at uses a
> separate bit for each one:

> +/* symbols for PGPROC.delayChkpt */
> +#define DELAY_CHKPT_START (1<<0)
> +#define DELAY_CHKPT_COMPLETE (1<<1)

Hm, that's not in the patch version that the CF app claims to be
latest [1].  It does this:

+/* type for PGPROC.delayChkpt */
+typedef enum DelayChkptType
+{
+    DELAY_CHKPT_NONE = 0,
+    DELAY_CHKPT_START,
+    DELAY_CHKPT_COMPLETE
+} DelayChkptType;

which seems like a distinct disimprovement over what you're quoting.

            regards, tom lane

[1] https://www.postgresql.org/message-id/20210106.173327.1444585955309078930.horikyota.ntt@gmail.com