Re: Introduction of open tracing (otter)

This looks really good. Just checking are we planning to integrate this for the 3.x release or 4.x? On Wed, Sep 11, 2019 at 12:50 AM Paul Davis wrote: > +1 > > On Tue, Sep 10, 2019 at 4:51 PM Adam Kocoloski > wrote: > > > > Yeah, I meant the latter — joining CouchDB’s span information to spans

Re: Pay-to-play /Sponsored Features

Hi Jan, Makes perfect sense, many thanks! Reddy De : Jan Lehnardt Envoyé : mardi 10 septembre 2019 13:42 À : dev@couchdb.apache.org Objet : Re: Pay-to-play /Sponsored Features Hi Reddy, Final decisions about inclusion of code developed however lies with this

Re: Introduction of open tracing (otter)

Hi Ilya, I like this quite a bit. It came by IRC or Slack the other day and it felt like a natural fit, modulo the perf questions, but I’d be surprised if that was a large problem, given what this is meant to do. That said, some folks disable logging for performance reasons *chuckle* Best Jan

Re: Compaction daemon changes

Cool. PR for the scheduling windows is up: https://github.com/apache/couchdb-smoosh/pull/3 Adam > On Sep 6, 2019, at 1:07 PM, Jan Lehnardt wrote: > > Heya Adam, that sounds all great. FWIW, I think the "do not compact during > this time" is b

Re: Introduction of open tracing (otter)

> For me the main question is how much overhead is associated with tracing. Bellow is a breakdown on cost of various operations > Can an admin safely configure it to run in production? I believe it can be used in production if we are careful with filters. We need to configure prefilter_rules t

Re: Proposed deprecations / removals (was: Re: Getting started on CouchDB 3.0, and an introduction)

Hi Jan - Do you happen to have the ticket/link for this? > I remembered one last deprecation we wanted in 3.0: security tightening, > which included the deprecation of admin party. > Thanks! Deni On Mon, Sep 9, 2019 at 2:14 PM Joan Touzet wrote: > I remembered one last deprecation we wanted i

IO queueing defaults

A few months ago a bunch of code landed on master around IO QoS and prioritization. I think we need to have a conversation about the defaults for that system and what we want to allow users to enable. First topic - there are actually two different generations of the IOQ system: IOQ and IOQ2. On