#codefresh (2019-07)
Archive: https://archive.sweetops.com/codefresh/
2019-07-02
Dockerhub registry issue - Impact on your Codefresh builds Jul 2, 14:09 UTC Investigating - We are currently investigating an issue with pulling images from dockerhub (502 errors). This seems like an overall issue in Dockerhub and may affect your builds in Codefresh even if your own images aren’t hosted in Dockerhub. We’ll provide more updates.
Dockerhub registry issue - Impact on your Codefresh builds Jul 2, 14:42 UTC Monitoring - Issue seems to be resolved, it was related to an issue on cloudflare that impacted services like dockerhub, npm and others. We’re currently monitoring the system to make sure everything is up.Jul 2, 14:09 UTC Investigating - We are currently investigating an issue with pulling images from dockerhub (502 errors). This seems like an overall issue in Dockerhub and may affect your builds in Codefresh even if your own images aren’t hosted in Dockerhub. We’ll provide…
Dockerhub registry issue - Impact on your Codefresh builds Jul 2, 19:20 UTC Resolved - According to Cloudflare’s Status Page, the issue is now resolved. Thus, all indirectly impacted services (DockerHub, NPM, among others) should be functional as well.
More info here: https://www.cloudflarestatus.com/incidents/9wjyx63y2xsyJul 2, 14:42 UTC Monitoring - Issue seems to be resolved, it was related to an issue on cloudflare that impacted services like dockerhub, npm and others. We’re currently monitoring the system to make sure everything is up.Jul 2,…
2019-07-10
Public #office-hours starting now! Join me here: https://zoom.us/meeting/register/dd2072a53834b30a7c24e00bf0acd2b8
Have any questions? This is your chance to ask us anything.
@Erik Osterman (Cloud Posse) Do you have details on use of the iam auth in a Codefresh pipeline? I am in a pairing call discussing the use and wondered if you have steps somewhere as an example of this in a pipeline. I am looking through your GitHub now.
I don’t have an example of using “STS” tokens
I can show you how it works, but not a codefresh pipeline
Mind helping me out on a call real fast with customer and talk me through how it would look in a pipeline and I’ll put the steps together post call?
sure
I believe it will be a freestyle step with the auth
is this for terraform by any chance?
(here’s how we’re running terratest)
we have 40+ repos updated so far
No this is using aws authenticator to pass in the ARN for use during the annotation creation for the ingress around the Amazon ALB.
Maybe I am not understanding this but the ARN is needed in the Kube config for the role to be able to use alb annotations.
oh, we haven’t tried using aws-iam-authenticator with codefresh.
Okay that’s cool. Then never mind the request. I think I’ll just need to write the steps.
yea, we’ve taken a different approach for deploying the ALB ingress.
we provision backing services with terraform under atlantis
then we write the arn to SSM
when we deploy the ALB ingress chart, then we pull the ARN out of SSM using chamber.
Thanks think they’re going with a different approach will gather more information but right now per example they have alb annotations defined in the values of a Helm chart that they pass to the ingress of the service to update alb.
and we deploy the alb ingress chart with helmfile
that said, I’m not keen on the ALB ingress.
it’s lame. it creates a new ALB for every ingress.
provisioning ALBs is slow (and they are expensive) relatively speaking.
you’ll quickly run out of ALBs (per the limits) and need to request a limit increase
per https://docs.aws.amazon.com/general/latest/gr/aws_service_limits.html isn’t the limit the same for ELBs? In regards to pricing, ELBs and ALBs are about the same price
You could create a single ingress for all services running in a single namespace
In our case, we setup environments for every pull request. Every pull request is in a new namespace.
It’s so nice how nginx-ingress can just reuse the load balancer
yes/no, it means that you need to rearchitect how the ingresses are defined. I think it’s nicer to define the ingress rules along side the service than globally in the namespace.
yeah I agree. Decoupling the services from the ingress is just an added PITA for whoever is configuring the service
are you able to limit which pipelines/projects have access to which k8s clusters?
i.e. only prod pipelines have access to the prod cluster or only pipelines in a specific project can have access to a project specific cluster
2019-07-11
ah enterprise feature
2019-07-17
2019-07-18
Docker Hub has reported an incident Jul 18, 17:37 UTC Investigating - Docker Hub has reported an incident affecting their Docker Hub Registry and Docker Hub Web components. This could affect Codefresh builds.
More information here: https://status.docker.com/pages/incident/533c6539221ae15e3f000031/5d30ac0eca873b5c9dfe7768
Docker Hub has reported an incident Jul 18, 17:47 UTC Monitoring - Docker Hub has reported the affected components as operational.
More information here: https://status.docker.com/pages/incident/533c6539221ae15e3f000031/5d30ac0eca873b5c9dfe7768Jul 18, 17:37 UTC Investigating - Docker Hub has reported an incident affecting their Docker Hub Registry and Docker Hub Web components. This could affect Codefresh builds.
More information here: https://status.docker.com/pages/incident/533c6539221ae15e3f000031/5d30ac0eca873b5c9dfe7768
GitHub has reported an incident Jul 18, 18:07 UTC Investigating - GitHub has reported an incident affecting Web and API requests.
More information here: https://www.githubstatus.com/ / https://www.githubstatus.com/incidents/jzcjznsv759m
Docker Hub has reported an incident Jul 18, 18:11 UTC Resolved - Incident reported as fully Resolved.
More information here: https://status.docker.com/pages/incident/533c6539221ae15e3f000031/5d30ac0eca873b5c9dfe7768Jul 18, 17:47 UTC Monitoring - Docker Hub has reported the affected components as operational.
More information here: https://status.docker.com/pages/incident/533c6539221ae15e3f000031/5d30ac0eca873b5c9dfe7768Jul 18, 17:37 UTC Investigating - Docker Hub has reported an incident affecting their Docker Hub Registry…
GitHub has reported an incident Jul 18, 18:54 UTC Monitoring - GitHub has reported that Web and API error rates are back to normal levels.Jul 18, 18:07 UTC Investigating - GitHub has reported an incident affecting Web and API requests.
More information here: https://www.githubstatus.com/ / https://www.githubstatus.com/incidents/jzcjznsv759m
GitHub has reported an incident Jul 18, 19:26 UTC Resolved - GitHub has reported this incident as ResolvedJul 18, 18:54 UTC Monitoring - GitHub has reported that Web and API error rates are back to normal levels.Jul 18, 18:07 UTC Investigating - GitHub has reported an incident affecting Web and API requests.
More information here: https://www.githubstatus.com/ / https://www.githubstatus.com/incidents/jzcjznsv759m
Codefresh issue affecting r.cfcr.io registry Jul 18, 22:03 UTC Investigating - We are investigating an issue affecting our r.cfcr.io registry component
Codefresh issue affecting r.cfcr.io registry Jul 18, 22:29 UTC Monitoring - There was a temporary network issue. The r.cfcr.io registry is operational again. We’ll keep monitoring it.Jul 18, 22:03 UTC Investigating - We are investigating an issue affecting our r.cfcr.io registry component
Codefresh issue affecting r.cfcr.io registry Jul 18, 23:25 UTC Resolved - This incident has been resolved.Jul 18, 22:29 UTC Monitoring - There was a temporary network issue. The r.cfcr.io registry is operational again. We’ll keep monitoring it.Jul 18, 22:03 UTC Investigating - We are investigating an issue affecting our r.cfcr.io registry component
2019-07-22
GitHub has reported an incident Jul 22, 15:58 UTC Investigating - GitHub has reported an incident This might impact your builds.
More information here: https://www.githubstatus.com/incidents/tq7pm7mn9b1k
GitHub has reported an incident Jul 22, 22:54 UTC Resolved - GH has reported this incident is now resolved.Jul 22, 15:58 UTC Investigating - GitHub has reported an incident This might impact your builds.
More information here: https://www.githubstatus.com/incidents/tq7pm7mn9b1k
Internal Incident Jul 23, 03:20 UTC Investigating - We’re experiencing an intermittent issue with our API. We’re still trying to identify the root cause.
Internal Incident Jul 23, 06:00 UTC Resolved - We have resolved the issue which caused intermediate network issues in CodefreshJul 23, 03:20 UTC Investigating - We’re experiencing an intermittent issue with our API. We’re still trying to identify the root cause.
2019-07-23
Codefresh UI outage Jul 23, 12:32 UTC Resolved - This incident has been resolved.Jul 23, 12:30 UTC Investigating - We are currently investigating this issue.
anyone use pipelines to create new projects/pipelines w/codefresh cli
see http://github.com/cloudposse/codefresh for example
2019-07-24
thanks Erik you guys always have an example for everything I need to start looking there first
1
they might not always be the “latest” (since usually it percolates from customer engagements)
2019-07-25
Platform Latency Jul 26, 03:10 UTC Update - We are continuing to investigate this issue.Jul 26, 03:09 UTC Investigating - Engineering is investigating reports of latency in platform.
Platform Latency Jul 26, 04:45 UTC Resolved - We resolved the issue and the platform performance is back to normalJul 26, 03:10 UTC Update - We are continuing to investigate this issue.Jul 26, 03:09 UTC Investigating - Engineering is investigating reports of latency in platform.
2019-07-26
@dustinvb have you seen any “auto discovery” feature for codefresh?
e.g. a repo could just define something like this:
version: "1.0"
kind: "pipeline"
metadata:
name: foobar/tests
isPublic: false
labels:
tags: []
deprecate:
applicationPort: '8080'
repoPipeline: true
project: foobar
spec:
triggers:
- branchRegex: /.*/gi
branchRegexInput: regex
commentRegex: /.*/gi
context: github-webhooks
events:
- push
provider: github
pullRequestAllowForkEvents: false
repo: foobar
type: git
contexts:
- terraform modules
variables: []
specTemplate:
location: url
url: >-
<https://raw.githubusercontent.com/cloudposse/foobar/master/codefresh/terraform.yml>
stages: []
steps: {}
it would automatically see a new foobar repo was created, detect the config above, and execute it.
this would tie in to org webhooks (not repo webhooks)
No I haven’t seen anything like this in use. I had a request just today to have a codefresh terraform module to perform something similar. Can you ping our shared channel and double check with Amir and Kostis there?
ok
thatd be awesome
this would be great with github templates
Wow, interesting idea
I hadn’t considered that