Re: Fix checkpoint skip logic on idle systems by tracking LSN progress

On Thu, Dec 22, 2016 at 3:10 AM, Andres Freund <andres(at)anarazel(dot)de> wrote:
> On 2016-12-21 16:35:28 -0500, Robert Haas wrote:
>> On Wed, Dec 21, 2016 at 4:28 PM, Andres Freund <andres(at)anarazel(dot)de> wrote:
>> > - Similarly I don't like the name "progress LSN" much. What does
>> > "progress" really mean in that". Maybe "consistency LSN"?
>>
>> Whoa. -1 from me for "consistency LSN". Consistency has to with
>> whether the cluster has recovered up to the minimum recovery point or
>> whatever -- that is -- questions like "am i going to run into torn
>> pages?" and "should I expect some heap tuples to maybe be missing
>> index tuples, or the other way around?".
>
> That's imo pretty much what progress LSN currently describes. Have there
> been any records which are important for durability/consistency and
> hence need to be archived and such.
>
>
>> What I think "progress LSN"
>> is getting at -- actually fairly well -- is whether we're getting
>> anything *important* done, not whether we are consistent. I don't
>> mind changing the name, but not to consistency LSN.
>
> Well, progress could just as well be replay. Or the actual insertion
> point. Or up to where we've written out. Or synced out. Or
> replicated....
>
> Open to other suggestions - I'm not really happy with consistency LSN,
> but definitely unhappy with progress LSN.
>

last_essential_LSN?

--
With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com