work-offline.html.md.erb provides information about setting up dnsmasq on linux for working with pcf dev. However, Ubuntu 14.04 and newer generally come with dnsmasq already set up and managed by NetworkManager.

On those systems the user should not need to install dnsmasq, and should put the configuration in /etc/NetworkManager/dnsmasq.d/ instead of /etc/dnsmasq.d. The config can be reloaded via service network-manager restart (for Ubuntu 14.04) or systemctl restart NetworkManager.service (for Ubuntu 15.04 and up)

I tried to reach out to the team there for help with a pcf dev rabbitmq issue but it said that I needed to contact the administrator for an invite

Hi,
I am trying to install pcf dev in windows 10 pro.
Pre-requisites for the installation are installed. But when I run the below command, Installation failed and gave below error.
Command: cf install-plugin cfdev
Error: Plugin cfdev v0.0.13 could not be installed as it contains commands with names that are already used: dev.

The link for the Windows installer points to the internal releases, where the latest was in 2016.

In my opinion would be better to point to the release page.

Hi, I tried the steps mentioned in the link - https://docs.pivotal.io/pcf-dev/install-windows-legacy.html

When I tried to do cf start dev, I am gettting the below error.

Please sign in with your Pivotal Network account.
Need an account? Join Pivotal Network: https://network.pivotal.io

Email←[36;1m>←[0m xxx

Password←[36;1m>←[0m
←[31;1mFAILED←[0m
Error: Get https://network.pivotal.io/api/v2/api_token?password=xxx&username=xxx:
x509: certificate signed by unknown authority.

I am running the setup behind my company firewall. I have already set http_proxy.

Can you please assist?

https://docs.pivotal.io/pcf-dev/install-osx.html

Uninstall section:
The plugin name should be cfdev.

It takes long time to start pcf dev.
Are there a quick way to start pcf dev, after 1st time started successfully?
Or we need to run the command "cf dev start -f FILEPATH/FILE" each time start?

Under the chapter "Insecure Docker Registries" this file contains the following:
"The -r flag adds support for custom insecure Docker registries."

This is either wrong or there is a bug in PCF DEV. When I setup a local "insecure" registry, i.e. a registry listening on HTTP and not HTTPS, and add this to my PCF DEV instance using the -r flag, I am getting the following error message:

...
Staging process started ...
Failed to talk to docker registry: Get https://host.pcfdev.io:5000/v2/: http: server gave HTTP response to HTTPS client
Staging process finished
...

This tells me that either PCF DEV does not honor the "-r" flag or it may expect a "secure" registry but with a self-signed certificate. I suspect the latter because the documentation of the "-r" flag in PCF DEV (with cf dev --help) seems to indicate that:

[-r registry1,registry2,...] Docker registries that PCF Dev will use without SSL validation. Specify in 'host:port' format.

It would be great if the documentation describes the correct way of doing it. It also would be great if developers would be given a recipe on how to setup a local Docker registry correctly since dealing with self-signed certificates is absolutely none-trivial!

Trying to follow Use Spring Cloud Services and install the Cook App and Config. The Config Server has the following error: The Config Server cannot initialize using the configuration that has been provided. Please double-check the configuration and correct any mistakes. Is there an updated config URL? Here's the one in the documentation, which returns a 404:

{
"count": 1,
"git": {
"label": "master",
"uri": "https://github.com/spring-cloud-samples/cook-config"
}
}

I've setup CF Dev correctly & can browse to the Spring demo app. However if I try to navigate to https://apps.dev.cfdev.sh, I keep getting a:

404 Not Found: Requested route ('apps.dev.cfdev.sh') does not exist.

rgds
SJ

Hi, I've tried to follow this approach to connect to the registry (https://docs.pivotal.io/pcf-dev/configuring.html) but pcf is still trying to connect via https. Here's what I've tried:

cf dev start -r 192.168.11.1:5000

cf login -a https://api.local.pcfdev.io --skip-ssl-validation

cf enable-feature-flag diego_docker

cf push my-app-kotlin -o 192.168.11.1:5000/my-app-kotlin:1.0

and then I got this:

Pushing app my-app-kotlin to org pcfdev-org / space pcfdev-space as admin...
Getting app info...
Creating app with these attributes...
+ name:           my-app-kotlin
+ docker image:   192.168.11.1:5000/my-app-kotlin:1.0
  routes:
+   my-app-kotlin.local.pcfdev.io

Creating app my-app-kotlin...
Mapping routes...

Staging app and tracing logs...
   Creating container
   Successfully created container
   Staging...
   Staging process started ...
   Failed to talk to docker registry: Get https://192.168.11.1:5000/v2/: http: server gave HTTP response to HTTPS client
   Staging process finished
   Exit status 0
   Staging Complete
   Destroying container
   Successfully destroyed container

Waiting for app to start...
Start unsuccessful

Just for the sake of it I've also connected via ssh to pcf to check if I had connection to my private registry and it actually works:

vcap@agent-id-pcfdev-0:~$ curl -X GET http://192.168.11.1:5000/v2/_catalog
{"repositories":["my-app-kotlin"]}

Am I missing something? How to bypass insecure connections between PCF and Docker Registry?

Thanks.

https://docs.pivotal.io/pcf-dev/usage-scs.html: step 5 (cf create service): command seems to be invalid.
Wrong syntax/quoting of JSON-arg and missing/wrong parameters.
After adding the missing quote at the JSON-arg I still get:

cf create-service -c '{ "git": { "uri": "https://github.com/spring-cloud-services-samples/cook-config", "label": "master" } }' p-config-server standard config-server

FAILED Incorrect Usage. Requires service, service plan, service instance as arguments

As I'm just doing my first steps in PCF dev, I don't know where to get this arguments from.

Hi,

I'm playing with pcfdev on my linux box and want to experiment with SCS, more specific, spring cloud server to store secrets.
The installation failes with exitcode 1.
I had this issue before and that was due to expired certificates. I think the same issue is back.
Error in the install log:

Service broker error: Could not find an available leader from [https://10.144.0.13:8443]

When I bosh ssh into my dedicated-mysql-broker and grep certificates in /var/vcap/sys/logs/*/*log I see:

broker/broker.log:[dedicated-mysql-adapter] 2021/01/20 17:31:29 Couldn't get read only information about URL https://10.144.0.13:8443, error: &errors.errorString{s:"GET https://10.144.0.13:8443/status failed: Get https://10.144.0.13:8443/status: x509: certificate has expired or is not yet valid"}

Is there still maintenance on the product?

The cf dev start command asks for the user's API token. It would be nice to mention this in the documentation so it's not a surprise to the user, and explain why it's needed.