>From: Tatsuo Ishii [mailto:ishii@sraoss.co.jp]
>
>> Personally, I don't find this hint particularly necessary. The
>> session was terminated because nothing was happening, so the real fix
>> on the application side is probably more involved than just retrying.
>> This is different from some of the other cases that were cited, such
>> as serialization conflicts, where you just got unlucky due to
>> concurrent events. In the case of idle-in-transaction-timeout, the
>> fault is entirely your own.
>
>Hum. Sounds like a fair argument. Ideriha-san, what do you think?
>
Hi.
As Peter mentioned, application code generally needs to handle more things
than retrying.
So I'm ok with not adding this hint.
Regards,
Takeshi Ideriha