Re: idle_in_transaction_timeout

On Tue, Jun 3, 2014 at 11:37 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> Robert Haas <robertmhaas(at)gmail(dot)com> writes:
>> I thought the reason why this hasn't been implemented before now is
>> that sending an ErrorResponse to the client will result in a loss of
>> protocol sync.
>
> Hmm ... you are right that this isn't as simple as an ereport(ERROR),
> but I'm not sure it's impossible. We could for instance put the backend
> into skip-till-Sync state so that it effectively ignored the next command
> message. Causing that to happen might be impracticably messy, though.

Another thing we could maybe do is AbortCurrentTransaction() and send
the client a NoticeResponse saying "hey, expect all of your future
commands to fail with complaints about the transaction being aborted".

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company