Re: reducing statistics write overhead

Martin Pihlak wrote:
> Magnus Hagander wrote:
>> I wrote a patch for this some time back, that was actually applied.
>> Turns out it didn't work, and I ran out of time to fix it, so it was
>> backed out again. And then I forgot about it :-) If you look through the
>> cvs history of pgstat you should be able to find it - maybe it can give
>> you some further ideas.
>
> Got it - this was 1.126. Looks very familiar indeed :)

:-)

> I had also previously experimented with stat() based polling but ran into
> the same issues - no portable high resolution timestamp on files. I guess
> stat() is unusable unless we can live with 1 second update interval for the
> stats (eg. backend reads the file if it is within 1 second of the request).

If the filesystem has inodes, noticing that the inode changes should
usually be enough, no? Since we always create a new file and rename it
into place, there is no way that inode can be reused right away. But it
does require that the stat info is not cached somewhere in between...

//Magnus