At 2012-12-29 14:23:45 -0500, sfrost(at)snowman(dot)net wrote:
>
> Regarding the actual comment, here's the wording that I'd use:

Sorry for nitpicking, but "we can't long jumps" made me cringe.
Here's a slightly more condensed version:

    /*
     * We can't use ereport(ERROR) here, because any longjmps
     * in DoingCommandRead state run the risk of violating our
     * protocol or the SSL protocol, by interrupting OpenSSL in
     * the middle of changing its internal state.
     *
     * Currently, the only option is to promote ERROR to FATAL
     * until we figure out a better way to handle errors in this
     * state.
     */

Patch along these lines attached, which also removes trailing
whitespace from the original patch.

-- Abhijit

In response to

• Re: Hot Standby conflict resolution handling at 2012-12-29 19:23:45 from Stephen Frost

Responses

• Re: Hot Standby conflict resolution handling at 2013-01-17 05:47:40 from Pavan Deolasee
• Re: Hot Standby conflict resolution handling at 2013-01-17 06:38:31 from Tom Lane

pgsql-hackers by date

Next:	From: Craig Ringer	Date: 2013-01-17 04:38:51
	Subject: Re: CF3+4 (was Re: Parallel query execution)
Previous:	From: Tom Lane	Date: 2013-01-17 03:40:07
	Subject: Re: CF3+4