couchdb-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Russell Branca <>
Subject Re: Migration Strategy [was Re: Git Repository Creation]
Date Thu, 23 Jan 2014 18:54:11 GMT
> 3. R15* & at least R16B01 and under are still prone to scheduler collapse.
> > I am not sure if the
> > “+st <msecs> " hack in R16B02+ eases the pain or not, nor how badly this
> > impacts heavy
> > CouchDB users. This is the “wake up all schedulers every <milliseconds>
> to
> > prevent collapse"
> > issue. Read Scott Fritchie’s threads in Erlang-Questions [1] for details
> &
> > erl man[2].
> >
> >
> Mmm the link you refer say it is still better than in R14. Anyway this is
> the right path to follow, we should collect any issues we have in mind in
> using latest supported versions of Erlang and see how it impact us.
I'm not sure where you see anything in DCH's link [1] that says R14 is
worse. The relevant bits from that thread are "R15B0x's schedulers are
broken", and "R16B's schedulers appear to be even more broken". The only
mention of R14 I see on that thread is: "The CPU usage is noticeably lower
than R14, but worryingly so.".

I think we should support R14 until their is a viable stable alternative,
or there is a sufficiently interesting upgrades to justify the switch. If
we need to run forked versions of 3rd party dependencies until such time
that we can upgrade in full, that seems like the logical path. Ideally we
can figure out a way to use upstream versions directly without needing to


  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message