Oh this is fun doing archeology on other projects (CouchDB in this case). I recognize one of the typical PR reviewers from elsewhere, and it appears that the DNS SRV based cluster discovery feature hasn't been merged upstream yet. This might turn into a neat opportunity for us to connect more with the CouchDB PMC. I'll link back any upstream issues.
On Sat, 20 Jul 2019 at 14:02, Matt Sicker <boa...@gmail.com> wrote: > > Thanks for the link! That looks like an even easier way to support > CouchDB for sure. I'll play around with each option to see which makes > more sense to prototype. > > On Sat, 20 Jul 2019 at 13:41, Michele Sciabarra <mich...@sciabarra.com> wrote: > > > > A critical aspect to consider is backup. There is an operator in the work > > (that is the primary reason so far to use an external CouchDB) There is an > > operator in the work that may be useful > > https://github.com/nicolai86/couchdb-operator > > > > -- > > Michele Sciabarra > > mich...@sciabarra.com > > > > > > > > ----- Original message ----- > > From: Matt Sicker <boa...@gmail.com> > > To: dev@openwhisk.apache.org > > Subject: Re: Looking to contribute; which areas need some love? > > Date: Saturday, July 20, 2019 8:03 PM > > > > Continuing discussion in > > https://github.com/apache/incubator-openwhisk-deploy-kube/issues/498 > > > > On Sat, 20 Jul 2019 at 12:48, Matt Sicker <boa...@gmail.com> wrote: > > > > > > So it seems that there's an incubating chart for CouchDB already which > > > might be adaptable: > > > https://github.com/helm/charts/tree/master/incubator/couchdb > > > > > > Main difference I already see is that they offer a StatefulSet for > > > clustering the CouchDB instances. Might not be too complicated. I'll > > > file an issue since I wasn't able to find any existing ones. > > > > > > On Fri, 19 Jul 2019 at 14:41, Matt Sicker <boa...@gmail.com> wrote: > > > > > > > > That does sound like an interesting thing to do. I've been looking for > > > > an excuse to practice more Kubernetes things, so this could be rather > > > > practical there. > > > > > > > > On Fri, 19 Jul 2019 at 13:21, Michele Sciabarra <mich...@sciabarra.com> > > > > wrote: > > > > > > > > > > In my opinion, given your experience with Kubernetes, a long needed > > > > > feature is to add support for clustered CouchDb in the helm chart. It > > > > > is the only missing pieces to allow for a true ‘production’ CouchDb. > > > > > Kafka is already clustered, CouchDb is not. David recommend to use an > > > > > external one but it is not always an option when you have an internal > > > > > deployment for example. > > > > > > > > > > -- > > > > > Michele Sciabarra > > > > > mich...@sciabarra.com > > > > > > > > > > > > > > > > > > > > ----- Original message ----- > > > > > From: Matt Sicker <boa...@gmail.com> > > > > > To: dev@openwhisk.apache.org > > > > > Subject: Looking to contribute; which areas need some love? > > > > > Date: Friday, July 19, 2019 7:10 PM > > > > > > > > > > Hey all, I've been working on some other personal project the past few > > > > > weekends, but I've been itching to explore this project more and make > > > > > some code contributions. I see the main openwhisk repo is Scala/Akka > > > > > which is tech that I've used before. However, I'm also interested in > > > > > hearing which areas feel neglected right now and could use some > > > > > attention. > > > > > > > > > > As a bonus, this can also start an idea of adding entries to > > > > > https://helpwanted.apache.org/ to recruit more contributors. I see > > > > > there's already some GSoC projects going on, so this is an additional > > > > > resource that can be useful here. > > > > > > > > > > -- > > > > > Matt Sicker <boa...@gmail.com> > > > > > > > > > > > > > > > > > -- > > > > Matt Sicker <boa...@gmail.com> > > > > > > > > > > > > -- > > > Matt Sicker <boa...@gmail.com> > > > > > > > > -- > > Matt Sicker <boa...@gmail.com> > > > > > -- > Matt Sicker <boa...@gmail.com> -- Matt Sicker <boa...@gmail.com>
