Well, in that spirit:
- Cache-side processing: +1. I suppose given the fact that we wouldn't want
to rewrite the entire configuration generation logic at once, there's no
reason to prefer this being part of ORT immediately versus separate. Either
way, there's no real "extra step". Though I must admit, I am sad to see
this written in Go and not built into my baby: ORT.py
- Invert LoC for config update: +1 because this essentially lets you do
server configuration snapshots for free, in addition to the other benefits,
and that's a pretty requested feature, I think. For servers that are
unreachable from TO, there's a problem, and after backing off for some set
number of retries they should probably just be marked offline with reason
e.g. "unable to contact for updates". This is a bit off-topic, though, imo,
because it's sort of independent from what Rob's suggesting, and that
utility isn't even really designed with that in mind - at least just yet. A
conversation for after it's finished.
- Invert LoC for data selection: +1 (I think). Letting the cache server
decide what it needs to know allows you to decouple it from any particular
cache server, which would let us support things like Grove or NGinx more
easily. Or at least allow people to write their own config generators
(plugins?) for those servers. Though honestly, it's probably always going
to want the entire profile/parameter set and information for assigned
delivery services anyway, just then it'll decide what's important and
what's meaningless. This is somewhat more related, imo, since I _think_
(without looking at any code) that what Rob's thing does now is just
request the information it thinks it needs, and builds the configs with
that. I'd be interested to hear more about "fragility when out-of-sync"
thing. Or maybe I'm misunderstanding the concept? If what you mean is
something more like "the cache server selects what specific parameters it
needs" then I'm -1, but you should be able to get all of the parameters and
their modified dates with one call to `/profiles?name={{name}}` and then
decide from there. So the server still tells you everything that just
changed. Stuff like CG assignment/params and DS assignment/params would
likewise still need to be checked normally. So +1 for caches deciding what
API endpoints to call, -1 for big globs of "this is what was updated" being
pushed to the cache server, and -1 for cache servers trying to guess what
might have changed instead of checking everything.
- Direct Database Connection: -1
- Kafka: -1
> "This is true, but you can also run the cache-config generator to
visually inspect them as well"
yeah, but then you need to either expose that configuration generator's
output to the internet or you need to `ssh` into a cache server or
something to inspect it individually. I think his concern is not being able
to see it in Traffic Portal, which is arguably safer and much easier than
the other two options, respectively.
My response would be "but you can still see arbitrary configuration, so
maybe that just needs to be made easier to view and understand to
compensate. Like, instead of
{
"configFile": "records.config",
"name": "CONFIG proxy.config.http.insert_response_via_str",
"secure": false,
"value": "INT 3"
},
you see something like 'encoded Via header verbosity: none/low/med/high'"
On Wed, Jul 31, 2019 at 11:01 AM Chris Lemmons <[email protected]> wrote:
> This is true, but you can also run the cache-config generator to
> visually inspect them as well. That makes it easy to visually inspect
> them as well as to pipe them to diff and mechanically inspect them. So
> we don't lose the ability entirely, we just move it from one place to
> another.
>
> On Wed, Jul 31, 2019 at 10:47 AM Genz, Geoffrey
> <[email protected]> wrote:
> >
> > A small point, but TO currently allows one to visually inspect/validate
> the generated configuration files. I don't know how critical that
> functionality is (I personally found it invaluable when testing logging
> configuration changes), but it seems like we either have the generation
> logic in two places (ORT and TO), or we lose that ability in TO by moving
> all the logic to the cache.
> >
> > - Geoff
> >
> > On 7/31/19, 10:33 AM, "Jeremy Mitchell" <[email protected]> wrote:
> >
> > my feedback:
> >
> > 1. i like the idea of slimming down TO. It's gotten way too fat.
> Basically
> > deprecating these api endpoints at some point and letting "something
> else"
> > do the job of config generation:
> >
> > GET /api/$version/servers/#id/configfiles/ats
> > GET /api/$version/profiles/#id/configfiles/ats/#filename
> > GET /api/$version/servers/#id/configfiles/ats/#filename
> > GET /api/$version/cdns/#id/configfiles/ats/#filename
> >
> > 2. i don't really care if that "something else" is a sidecar to ORT
> or
> > actually ORT. will let you guys hash that out.
> >
> > 3. i like the idea of that "something else" eventually being able to
> handle
> > a push vs. a pull as rawlin suggested.
> >
> > 4. a bit curious how "cache snapshots" would work as rob suggested in
> >
> https://cwiki.apache.org/confluence/display/TC/Cache-Side+Config+Generation
> -
> > would you look at a cache snapshot diff and then snapshot (which
> would
> > queue updates in the background)?
> >
> > otherwise, thanks for taking the initiative, rob. and looking
> forward to
> > seeing what comes of this that will make TC safer/more efficient.
> >
> > jeremy
> >
> >
> >
> >
> >
> > On Wed, Jul 31, 2019 at 9:20 AM Gray, Jonathan <
> [email protected]>
> > wrote:
> >
> > > Smaller, simpler pieces closer to the cache that do one job are far
> > > simpler to maintain, triage, and build. I'm not a fan of trying
> to inject
> > > a message bus in the middle of everything.
> > >
> > > Jonathan G
> > >
> > >
> > > On 7/31/19, 8:48 AM, "Genz, Geoffrey" <[email protected]>
> wrote:
> > >
> > > To throw a completely different idea out there . . . some time
> ago
> > > Matt Mills was talking about using Kafka as the configuration
> transport
> > > mechanism for Traffic Control. The idea is to use a Kafka
> compacted topic
> > > as the configuration source. TO would write database updates to
> Kafka, and
> > > the ORT equivalent would pull its configuration from Kafka.
> > >
> > > To explain compacted topics a bit, a standard Kafka message is
> a key
> > > and a payload; in a compacted topics, only the most recent
> message/payload
> > > with a particular key is kept. As a result, reading all the
> messages from
> > > a topic will give you the current state of what's basically a key
> value
> > > store, with the benefit of not doing actual mutations of data. So
> a cache
> > > could get the full expected configuration by reading all the
> existing
> > > messages on the appropriate topic, as well as get new updates to
> > > configuration by listening for new Kafka messages.
> > >
> > > This leaves the load on the Kafka brokers, which I can assure
> you
> > > given recent experience, is minimal. TO would only have the
> responsibility
> > > of writing database updates to Kafka, ORT only would need to read
> > > individual updates (and be smart enough to know how and when to
> apply them
> > > -- perhaps hints could be provided in the payload?). The result
> is TO is
> > > "pushing" updates to the caches (via Kafka) as Rawlin was
> proposing, and
> > > ORT could still pull the full configuration whenever necessary
> with no hit
> > > to Postgres or TO.
> > >
> > > Now this is obviously a radical shift (and there are no doubt
> other
> > > ways to implement the basic idea), but It seemed worth bringing up.
> > >
> > > - Geoff
> > >
> > > On 7/31/19, 8:30 AM, "Lavanya Bathina" <[email protected]>
> wrote:
> > >
> > > +1 on this
> > >
> > > On Jul 30, 2019, at 6:01 PM, Rawlin Peters <
> > > [email protected]> wrote:
> > >
> > > I've been thinking for a while now that ORT's current
> pull-based
> > > model
> > > of checking for queued updates is not really ideal, and I
> was
> > > hoping
> > > with "ORT 2.0" that we would switch that paradigm around
> to where
> > > TO
> > > itself would push updates out to queued caches. That way
> TO would
> > > never get overloaded because we could tune the level of
> concurrency
> > > for pushing out updates (based on server capacity/specs),
> and we
> > > would
> > > eliminate the "waiting period" between the time updates
> are queued
> > > and
> > > the time ORT actually updates the config on the cache.
> > >
> > > I think cache-side config generation is a good idea in
> terms of
> > > enabling canary deployments, but as CDNs continue to scale
> by
> > > adding
> > > more and more caches, we might want to get out ahead of
> the ORT
> > > load/waiting problem by flipping that paradigm from "pull"
> to
> > > "push"
> > > somehow. Then instead of 1000 caches all asking TO the same
> > > question
> > > and causing 1000 duplicated reads from the DB, TO would
> just read
> > > the
> > > one answer from the DB and send it to all the caches,
> further
> > > reducing
> > > load on the DB as well. The data in the "push" request
> from TO to
> > > ORT
> > > 2.0 would contain all the information ORT would request
> from the
> > > API
> > > itself, not the actual config files.
> > >
> > > With the API transition from Perl to Go, I think we're
> eliminating
> > > the
> > > Perl CPU bottleneck from TO, but the next bottleneck seems
> like it
> > > would be reading from the DB due to the constantly growing
> number
> > > of
> > > concurrent ORT requests as a CDN scales up. We should keep
> that in
> > > mind for whatever "ORT 2.0"-type changes we're making so
> that it
> > > won't
> > > make flipping that paradigm around even harder.
> > >
> > > - Rawlin
> > >
> > > > On Tue, Jul 30, 2019 at 4:23 PM Robert Butts <
> [email protected]>
> > > wrote:
> > > >
> > > >> I'm confused why this is separate from ORT.
> > > >
> > > > Because ORT does a lot more than just fetching config
> files.
> > > Rewriting all
> > > > of ORT in Go would be considerably more work.
> Contrawise, if we
> > > were to put
> > > > the config generation in the ORT script itself, we would
> have to
> > > write it
> > > > all from scratch in Perl (the old config gen used the
> database
> > > directly,
> > > > it'd still have to be rewritten) or Python. This was
> just the
> > > easiest path
> > > > forward.
> > > >
> > > >> I feel like this logic should just be replacing the
> config
> > > fetching logic
> > > > of ORT
> > > >
> > > > That's exactly what it does: the PR changes ORT to call
> this app
> > > instead of
> > > > calling Traffic Ops over HTTP:
> > > >
> > >
> https://github.com/apache/trafficcontrol/pull/3762/files#diff-fe8a3eac71ee592a7170f2bdc7e65624R1485
> > > >
> > > >> Is that the eventual plan? Or does our vision of the
> future
> > > include this
> > > > *and* ORT?
> > > >
> > > > I reserve the right to develop a strong opinion about
> that in
> > > the future.
> > > >
> > > >
> > > > On Tue, Jul 30, 2019 at 3:17 PM ocket8888 <
> [email protected]>
> > > wrote:
> > > >
> > > >>> "I'm just looking for consensus that this is the right
> > > approach."
> > > >>
> > > >> Umm... sort of. I think moving cache configuration to
> the cache
> > > itself
> > > >> is a great idea,
> > > >>
> > > >> but I'm confused why this is separate from ORT. Like if
> this is
> > > going to
> > > >> be generating the
> > > >>
> > > >> configs and it's already right there on the server, I
> feel like
> > > this
> > > >> logic should just be
> > > >>
> > > >> replacing the config fetching logic of ORT (and
> personally I
> > > think a
> > > >> neat place to try it
> > > >>
> > > >> out would be in ORT.py).
> > > >>
> > > >>
> > > >> Is that the eventual plan? Or does our vision of the
> future
> > > include this
> > > >> *and* ORT?
> > > >>
> > > >>
> > > >>> On 7/30/19 2:15 PM, Robert Butts wrote:
> > > >>> Hi all! I've been working on moving the ATS config
> generation
> > > from
> > > >> Traffic
> > > >>> Ops to a standalone app alongside ORT, that queries the
> > > standard TO API
> > > >> to
> > > >>> generate its data. I just wanted to put it here, and
> get some
> > > feedback,
> > > >> to
> > > >>> make sure the community agrees this is the right
> direction.
> > > >>>
> > > >>> There's a (very) brief spec here: (I might put more
> detail
> > > into it later,
> > > >>> let me know if that's important to anyone)
> > > >>>
> > > >>
> > >
> https://cwiki.apache.org/confluence/display/TC/Cache-Side+Config+Generation
> > > >>>
> > > >>> And the Draft PR is here:
> > > >>> https://github.com/apache/trafficcontrol/pull/3762
> > > >>>
> > > >>> This has a number of advantages:
> > > >>> 1. TO is a monolith, this moves a significant amount
> of logic
> > > out of it,
> > > >>> into a smaller per-cache app/library that's easier to
> test,
> > > validate,
> > > >>> rewrite, deploy, canary, rollback, etc.
> > > >>> 2. Deploying cache config changes is much smaller and
> safer.
> > > Instead of
> > > >>> having to deploy (and potentially roll back) TO, you
> can
> > > canary deploy on
> > > >>> one cache at a time.
> > > >>> 3. This makes TC more cache-agnostic. It moves cache
> config
> > > generation
> > > >>> logic out of TO, and into an independent app/library.
> The app
> > > (atstccfg)
> > > >> is
> > > >>> actually very similar to Grove's config generator
> > > (grovetccfg). This
> > > >> makes
> > > >>> it easier and more obvious how to write config
> generators for
> > > other
> > > >> proxies.
> > > >>> 4. By using the API and putting the generator
> functions in a
> > > library,
> > > >> this
> > > >>> really gives a lot more flexibility to put the config
> gen
> > > anywhere you
> > > >> want
> > > >>> without too much work. You could easily put it in an
> HTTP
> > > service, or
> > > >> even
> > > >>> put it back in TO via a Plugin. That's not something
> that's
> > > really
> > > >> possible
> > > >>> with the existing system, generating directly from the
> > > database.
> > > >>>
> > > >>> Right now, I'm just looking for consensus that this is
> the
> > > right
> > > >> approach.
> > > >>> Does the community agree this is the right direction?
> Are
> > > there concerns?
> > > >>> Would anyone like more details about anything in
> particular?
> > > >>>
> > > >>> Thanks,
> > > >>>
> > > >>
> > >
> > >
> > >
> > >
> > >
> >
> >
>
