This page is not created by, affiliated with, or supported by Slack Technologies, Inc.
2018-06-19
Channels
- # aws-lambda (1)
- # beginners (35)
- # cider (7)
- # cljsjs (2)
- # clojure (48)
- # clojure-austria (1)
- # clojure-conj (9)
- # clojure-dev (8)
- # clojure-india (6)
- # clojure-italy (12)
- # clojure-nl (8)
- # clojure-norway (3)
- # clojure-spec (9)
- # clojure-uk (92)
- # clojurescript (103)
- # community-development (7)
- # cursive (15)
- # datomic (75)
- # devcards (3)
- # emacs (3)
- # events (1)
- # fulcro (129)
- # hoplon (4)
- # immutant (2)
- # jobs (10)
- # leiningen (9)
- # off-topic (4)
- # onyx (2)
- # re-frame (45)
- # reagent (39)
- # reitit (40)
- # remote-jobs (4)
- # ring (2)
- # ring-swagger (9)
- # shadow-cljs (17)
- # tools-deps (31)
@jaret by removing every line/namespace and testing i finally found what is that breaks, if you have a namespace called "datomic.db" in your project, codedeploy fails at validate service
Does anybody know whether communication with an on-prem transactor is secure by default? I'm trying to get one running in Heroku, backed with Heroku Postgres; all comms with the DB are over SSL and authenticated, so I'm not worried about the storage, so if comms between peers and the transactor are secure then I'm good to go, but I can't find any mention of the transactor's network protocol anywhere. I'd be very grateful if anyone knows the answer or can point me to a helpful resource, thanks
@gabriele.carrettoni good catch! You cannot, and should never, use namespaces owned by some other organization in your code
@conan look at your transactor properties file
## Set to false to disable SSL between the peers and the transactor.
# Default: true
# encrypt-channel=true
@stuarthalloway now i get namespace 'cheshire.factory' not found
java.lang.Exception: namespace 'cheshire.factory' not found, compiling:(cheshire/core.clj:1:1)
what is going on now 😣@gabriele.carrettoni I doubt cheshire will work at all until we update jackson in Datomic Cloud, planned for the next release
Cheshire requires a version of Jackson that uses methods not present in the version currently shipping with Datomic Cloud
@stuarthalloway i see, i'll remove the dependency and use data.json
that is a priority fix for us, you are (at least) the second person to hit it
There is a roadmap to update on-prem deps too? By default, it has issues with clojurescript and tons of other projects, like onyx
can’t connect to datomic cloud:
Caused by: clojure.lang.ExceptionInfo: Unable to connect to system: #:cognitect.anomalies{:category :cognitect.anomalies/unavailable, :message "Connection refused"}
@denik you’re getting that error in your local repl? have you restarted your SOCKS proxy? can you run the proxy test (https://docs.datomic.com/cloud/getting-started/connecting.html#test-bastion)
I can’t run the test, we shut down out bastion bc we develop with a machine inside the VPC
I might have had an orphan jvm running that was holding a connection but after restart that should be wiped, too
is it possible you opened up SG ingress to a specific IP (the original dev system) and now you have a new IP
(using ions) i have renamed a namespace and deployed a new version, now the lambda throws that it doesn't find the namespace
That's a known issue with Aws marketplace. You can unsubscribe and resubscribe to get it to go away or you can just choose the one that has the latest
Is there a way to pass in pull selector to a query? Like:
(d/q '[:find (pull ?e ?fields)
:in $ ?fields
:where [?e :game/name _]]
(d/db conn) [:db/id :game/name])
This fails when using the datomic client apiHello, has anyone been able to configure logging inside of ions so that the log entries show up in cloudwatch?
@okocim you should be able to use CloudWatch like any other API. That said, The ion API does not yet provide a way to put your logs in the same log stream that Datomic uses. That is planned for a future release.
@stuarthalloway Thanks Stu. I was trying to avoid using the API directly and instead make use of either the LambdaAppender or the System.out.println functionality, but I was having a hard time determining exactly which log stream this ends up in. Appreciate the heads up.
@okocim remember that ion code is not running in a Lambda
the Lambda is just a proxy
got it. Well that explains why I wasn't seeing what I expected.. I'm going to re-familiarize myself with the docs. Thanks so much 🙂
@lockdown- that doesn’t sound right.
@joe.lane which part? I didn't include the bastion but included the cost of the ec2 nodes
2 i3.large instances for 30 days runs you ~$224.64 https://aws.amazon.com/ec2/pricing/on-demand/
Oh, you know what, I think you’re right @lockdown-.
Ultimately I think it depends on the “per node” part of your question. Sorry for adding confusion.
Any guidance on running tests with the cloud client lib? I have a rather big hammer approach right now that involves creating/deleting DBs for tests that need clean data, and it's pretty slow.
Ive done that and found it to be pretty fast