1. Topics
  2. CF CLI: how do you use `no-start` with `cf push`

CF CLI: how do you use `no-start` with `cf push`

Abby Chau
 

Hi everyone,

The CF CLI is soliciting feedback on how you use the `no-start` flag: 
  • what are your specific use cases? 
  • if the flag was no longer available, how might that affect you? (Would you strongly object to such a change, and why?) would you have a sensible workaround?

Feedback

Please contact us on the #cli channel on Cloud Foundry Slack. Alternatively, respond directly to this email.

Best,

Abby Chau
CF Product Manager, CF CLI


Note: The CF CLI team is currently working on updating the `cf v3-push` command to use the server side manifest work we've completed (with the Capi team), and to update the command such that it has as much feature parity with the current `cf-push` as possible. The eventual goal is to make the `cf push` command default to the enhanced v3-push command for users (on cc api 3.27 or higher). Since there is more granular control options in v3, `no-start` is a more nebulous flag.
Lee Porte
 

Hi, 

We use it when setting env vars that we don't want to commit. Or when committing then would harm the reuse of our code by others.


On Wed, 27 Jun 2018, 17:18 Abby Chau, <achau@...> wrote:
Hi everyone,

The CF CLI is soliciting feedback on how you use the `no-start` flag: 
  • what are your specific use cases? 
  • if the flag was no longer available, how might that affect you? (Would you strongly object to such a change, and why?) would you have a sensible workaround?

Feedback

Please contact us on the #cli channel on Cloud Foundry Slack. Alternatively, respond directly to this email.

Best,

Abby Chau
CF Product Manager, CF CLI


Note: The CF CLI team is currently working on updating the `cf v3-push` command to use the server side manifest work we've completed (with the Capi team), and to update the command such that it has as much feature parity with the current `cf-push` as possible. The eventual goal is to make the `cf push` command default to the enhanced v3-push command for users (on cc api 3.27 or higher). Since there is more granular control options in v3, `no-start` is a more nebulous flag.
Mike Youngstrom <youngm@...>
 

We mostly use it to do things to an application before starting it that cannot be done in a manifest.  For example, setting computed environment variables, or binding services that need service-parameters.  I think "--no-start" can be removed and we could use v3-create-app instead.

Mike

On Wed, Jun 27, 2018 at 10:18 AM Abby Chau <achau@...> wrote:
Hi everyone,

The CF CLI is soliciting feedback on how you use the `no-start` flag: 
  • what are your specific use cases? 
  • if the flag was no longer available, how might that affect you? (Would you strongly object to such a change, and why?) would you have a sensible workaround?

Feedback

Please contact us on the #cli channel on Cloud Foundry Slack. Alternatively, respond directly to this email.

Best,

Abby Chau
CF Product Manager, CF CLI


Note: The CF CLI team is currently working on updating the `cf v3-push` command to use the server side manifest work we've completed (with the Capi team), and to update the command such that it has as much feature parity with the current `cf-push` as possible. The eventual goal is to make the `cf push` command default to the enhanced v3-push command for users (on cc api 3.27 or higher). Since there is more granular control options in v3, `no-start` is a more nebulous flag.
Zach Robinson
 

Hi Lee,

Have you seen the new variable substitution feature for manifests?  It might really help your use-case.


-Zach

On Wed, Jun 27, 2018 at 10:09 AM Mike Youngstrom <youngm@...> wrote:
We mostly use it to do things to an application before starting it that cannot be done in a manifest.  For example, setting computed environment variables, or binding services that need service-parameters.  I think "--no-start" can be removed and we could use v3-create-app instead.

Mike

On Wed, Jun 27, 2018 at 10:18 AM Abby Chau <achau@...> wrote:
Hi everyone,

The CF CLI is soliciting feedback on how you use the `no-start` flag: 
  • what are your specific use cases? 
  • if the flag was no longer available, how might that affect you? (Would you strongly object to such a change, and why?) would you have a sensible workaround?

Feedback

Please contact us on the #cli channel on Cloud Foundry Slack. Alternatively, respond directly to this email.

Best,

Abby Chau
CF Product Manager, CF CLI


Note: The CF CLI team is currently working on updating the `cf v3-push` command to use the server side manifest work we've completed (with the Capi team), and to update the command such that it has as much feature parity with the current `cf-push` as possible. The eventual goal is to make the `cf push` command default to the enhanced v3-push command for users (on cc api 3.27 or higher). Since there is more granular control options in v3, `no-start` is a more nebulous flag.
Krannich, Bernd <bernd.krannich@...>
 

Hi Abby,

 

Thank you very much for reaching out. Indeed, we are frequent users of the `no-start` flag. A couple of use cases:

 

  • After a `cf ssh-enable APP_NAME`, you’ll need to restart your app to be able to SSH in. To prevent starting the app twice in that process, people do a `cf push --no-start` and then an explicit `cf start APP_NAME` afterwards.
  • The same applies for changes of security groups which also require an app restart.
  • Lastly, and maybe most importantly, people use `no-start` for pushing an app to which they want to bind services. They push the app with “no-start”, “cf create-service” their services, “cf bind-service” them to the app, and only then start the app.

 

So, I guess in summary, for all operations changing the app environment, doing those changes with the app not yet started is a common use case for us.

 

Regards,

Bernd

 

Bernd Krannich

SAP Cloud Platform

SAP SE

Dietmar-Hopp-Allee 16, 69190 Walldorf, Germany

 

E bernd.krannich@...

 

Pflichtangaben/Mandatory Disclosure Statement: www.sap.com/impressum

 

Diese E-Mail kann Betriebs- oder Geschäftsgeheimnisse oder sonstige vertrauliche Informationen enthalten. Sollten Sie diese E-Mail irrtümlich erhalten haben, ist Ihnen eine Kenntnisnahme des Inhalts, eine Vervielfältigung oder Weitergabe der E-Mail ausdrücklich untersagt. Bitte benachrichtigen Sie uns und vernichten Sie die empfangene E-Mail. Vielen Dank.

 

This e-mail may contain trade secrets or privileged, undisclosed, or otherwise confidential information. If you have received this e-mail in error, you are hereby notified that any review, copying, or distribution of it is strictly prohibited. Please inform us immediately and destroy the original transmittal. Thank you for your cooperation.

 

 

From: <cf-dev@...> on behalf of Abby Chau <achau@...>
Reply-To: "cf-dev@..." <cf-dev@...>
Date: Wednesday, 27. June 2018 at 18:18
To: "Discussions about Cloud Foundry projects and the system overall." <cf-dev@...>
Subject: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

 

Hi everyone,

 

The CF CLI is soliciting feedback on how you use the `no-start` flag: 

  • what are your specific use cases? 
  • if the flag was no longer available, how might that affect you? (Would you strongly object to such a change, and why?) would you have a sensible workaround?

 

Feedback

 

Please contact us on the #cli channel on Cloud Foundry Slack. Alternatively, respond directly to this email.



Best,



Abby Chau

CF Product Manager, CF CLI



 

Note: The CF CLI team is currently working on updating the `cf v3-push` command to use the server side manifest work we've completed (with the Capi team), and to update the command such that it has as much feature parity with the current `cf-push` as possible. The eventual goal is to make the `cf push` command default to the enhanced v3-push command for users (on cc api 3.27 or higher). Since there is more granular control options in v3, `no-start` is a more nebulous flag.
Hjortshoj, Julian <Julian.Hjortshoj@...>
 

+1 to that last use case.  In particular, if a service requires bind configuration, then it's not possible to put the binding in an app manifest (because manifests don't support service binding configuration) so it is very helpful to keep your app from starting until the service is bound.

Adding the ability to specify bind configuration in an application manifest might be a reasonable mitigation if --no-start were to disappear.

HTH,
-Julian

From: cf-dev@... [cf-dev@...] on behalf of Krannich, Bernd [bernd.krannich@...]
Sent: Wednesday, June 27, 2018 9:55 AM
To: cf-dev@...
Subject: Re: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

Hi Abby,

 

Thank you very much for reaching out. Indeed, we are frequent users of the `no-start` flag. A couple of use cases:

 

  • After a `cf ssh-enable APP_NAME`, you’ll need to restart your app to be able to SSH in. To prevent starting the app twice in that process, people do a `cf push --no-start` and then an explicit `cf start APP_NAME` afterwards.
  • The same applies for changes of security groups which also require an app restart.
  • Lastly, and maybe most importantly, people use `no-start` for pushing an app to which they want to bind services. They push the app with “no-start”, “cf create-service” their services, “cf bind-service” them to the app, and only then start the app.

 

So, I guess in summary, for all operations changing the app environment, doing those changes with the app not yet started is a common use case for us.

 

Regards,

Bernd

 

Bernd Krannich

SAP Cloud Platform

SAP SE

Dietmar-Hopp-Allee 16, 69190 Walldorf, Germany

 

E bernd.krannich@...

 

Pflichtangaben/Mandatory Disclosure Statement: www.sap.com/impressum

 

Diese E-Mail kann Betriebs- oder Geschäftsgeheimnisse oder sonstige vertrauliche Informationen enthalten. Sollten Sie diese E-Mail irrtümlich erhalten haben, ist Ihnen eine Kenntnisnahme des Inhalts, eine Vervielfältigung oder Weitergabe der E-Mail ausdrücklich untersagt. Bitte benachrichtigen Sie uns und vernichten Sie die empfangene E-Mail. Vielen Dank.

 

This e-mail may contain trade secrets or privileged, undisclosed, or otherwise confidential information. If you have received this e-mail in error, you are hereby notified that any review, copying, or distribution of it is strictly prohibited. Please inform us immediately and destroy the original transmittal. Thank you for your cooperation.

 

 

From: <cf-dev@...> on behalf of Abby Chau <achau@...>
Reply-To: "cf-dev@..." <cf-dev@...>
Date: Wednesday, 27. June 2018 at 18:18
To: "Discussions about Cloud Foundry projects and the system overall." <cf-dev@...>
Subject: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

 

Hi everyone,

 

The CF CLI is soliciting feedback on how you use the `no-start` flag: 

  • what are your specific use cases? 
  • if the flag was no longer available, how might that affect you? (Would you strongly object to such a change, and why?) would you have a sensible workaround?

 

Feedback

 

Please contact us on the #cli channel on Cloud Foundry Slack. Alternatively, respond directly to this email.



Best,



Abby Chau

CF Product Manager, CF CLI



 

Note: The CF CLI team is currently working on updating the `cf v3-push` command to use the server side manifest work we've completed (with the Capi team), and to update the command such that it has as much feature parity with the current `cf-push` as possible. The eventual goal is to make the `cf push` command default to the enhanced v3-push command for users (on cc api 3.27 or higher). Since there is more granular control options in v3, `no-start` is a more nebulous flag.
Abby Chau
 

Hey Bernd and Julian,

Many thanks for your responses. 

Wondering if you folks have tried using the `v3-create-app` command as Mike has pointed out (and Zach elsewhere). The command is basically the equivalent of `cf push test --no-start` in that it creates an app, leaves it in a `stopped` state, and users can then apply configurations, bind services, set-groups, enable ssh, etc. This command appears to be a workaround for your use cases, but please let us know if that's incorrect. 

Best,

Abby

On Wed, Jun 27, 2018 at 3:04 PM, Hjortshoj, Julian <Julian.Hjortshoj@...> wrote:
+1 to that last use case.  In particular, if a service requires bind configuration, then it's not possible to put the binding in an app manifest (because manifests don't support service binding configuration) so it is very helpful to keep your app from starting until the service is bound.

Adding the ability to specify bind configuration in an application manifest might be a reasonable mitigation if --no-start were to disappear.

HTH,
-Julian
From: cf-dev@... [cf-dev@...] on behalf of Krannich, Bernd [bernd.krannich@...]
Sent: Wednesday, June 27, 2018 9:55 AM
To: cf-dev@...
Subject: Re: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

Hi Abby,

 

Thank you very much for reaching out. Indeed, we are frequent users of the `no-start` flag. A couple of use cases:

 

  • After a `cf ssh-enable APP_NAME`, you’ll need to restart your app to be able to SSH in. To prevent starting the app twice in that process, people do a `cf push --no-start` and then an explicit `cf start APP_NAME` afterwards.
  • The same applies for changes of security groups which also require an app restart.
  • Lastly, and maybe most importantly, people use `no-start` for pushing an app to which they want to bind services. They push the app with “no-start”, “cf create-service” their services, “cf bind-service” them to the app, and only then start the app.

 

So, I guess in summary, for all operations changing the app environment, doing those changes with the app not yet started is a common use case for us.

 

Regards,

Bernd

 

Bernd Krannich

SAP Cloud Platform

SAP SE

Dietmar-Hopp-Allee 16, 69190 Walldorf, Germany

 

E bernd.krannich@...

 

Pflichtangaben/Mandatory Disclosure Statement: www.sap.com/impressum

 

Diese E-Mail kann Betriebs- oder Geschäftsgeheimnisse oder sonstige vertrauliche Informationen enthalten. Sollten Sie diese E-Mail irrtümlich erhalten haben, ist Ihnen eine Kenntnisnahme des Inhalts, eine Vervielfältigung oder Weitergabe der E-Mail ausdrücklich untersagt. Bitte benachrichtigen Sie uns und vernichten Sie die empfangene E-Mail. Vielen Dank.

 

This e-mail may contain trade secrets or privileged, undisclosed, or otherwise confidential information. If you have received this e-mail in error, you are hereby notified that any review, copying, or distribution of it is strictly prohibited. Please inform us immediately and destroy the original transmittal. Thank you for your cooperation.

 

 

From: <cf-dev@...> on behalf of Abby Chau <achau@...>
Reply-To: "cf-dev@..." <cf-dev@...>
Date: Wednesday, 27. June 2018 at 18:18
To: "Discussions about Cloud Foundry projects and the system overall." <cf-dev@...>
Subject: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

 

Hi everyone,

 

The CF CLI is soliciting feedback on how you use the `no-start` flag: 

  • what are your specific use cases? 
  • if the flag was no longer available, how might that affect you? (Would you strongly object to such a change, and why?) would you have a sensible workaround?

 

Feedback

 

Please contact us on the #cli channel on Cloud Foundry Slack. Alternatively, respond directly to this email.



Best,



Abby Chau

CF Product Manager, CF CLI



 

Note: The CF CLI team is currently working on updating the `cf v3-push` command to use the server side manifest work we've completed (with the Capi team), and to update the command such that it has as much feature parity with the current `cf-push` as possible. The eventual goal is to make the `cf push` command default to the enhanced v3-push command for users (on cc api 3.27 or higher). Since there is more granular control options in v3, `no-start` is a more nebulous flag.
Zach Robinson
 

thanks to all that have replied.

i'd like to ask a clarifying question.  it sounds like most of these use-cases are "set up some environment for the app".  this implies to me that these workflows are for new apps and not for updating existing ones.  does that sound right?  are their use-cases for "no-start" when updating an app?

thanks!
-zach

On Wed, Jun 27, 2018 at 4:30 PM Abby Chau <achau@...> wrote:
Hey Bernd and Julian,

Many thanks for your responses. 

Wondering if you folks have tried using the `v3-create-app` command as Mike has pointed out (and Zach elsewhere). The command is basically the equivalent of `cf push test --no-start` in that it creates an app, leaves it in a `stopped` state, and users can then apply configurations, bind services, set-groups, enable ssh, etc. This command appears to be a workaround for your use cases, but please let us know if that's incorrect. 

Best,

Abby

On Wed, Jun 27, 2018 at 3:04 PM, Hjortshoj, Julian <Julian.Hjortshoj@...> wrote:
+1 to that last use case.  In particular, if a service requires bind configuration, then it's not possible to put the binding in an app manifest (because manifests don't support service binding configuration) so it is very helpful to keep your app from starting until the service is bound.

Adding the ability to specify bind configuration in an application manifest might be a reasonable mitigation if --no-start were to disappear.

HTH,
-Julian
From: cf-dev@... [cf-dev@...] on behalf of Krannich, Bernd [bernd.krannich@...]
Sent: Wednesday, June 27, 2018 9:55 AM
To: cf-dev@...
Subject: Re: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

Hi Abby,

 

Thank you very much for reaching out. Indeed, we are frequent users of the `no-start` flag. A couple of use cases:

 

  • After a `cf ssh-enable APP_NAME`, you’ll need to restart your app to be able to SSH in. To prevent starting the app twice in that process, people do a `cf push --no-start` and then an explicit `cf start APP_NAME` afterwards.
  • The same applies for changes of security groups which also require an app restart.
  • Lastly, and maybe most importantly, people use `no-start` for pushing an app to which they want to bind services. They push the app with “no-start”, “cf create-service” their services, “cf bind-service” them to the app, and only then start the app.

 

So, I guess in summary, for all operations changing the app environment, doing those changes with the app not yet started is a common use case for us.

 

Regards,

Bernd

 

Bernd Krannich

SAP Cloud Platform

SAP SE

Dietmar-Hopp-Allee 16, 69190 Walldorf, Germany

 

E bernd.krannich@...

 

Pflichtangaben/Mandatory Disclosure Statement: www.sap.com/impressum

 

Diese E-Mail kann Betriebs- oder Geschäftsgeheimnisse oder sonstige vertrauliche Informationen enthalten. Sollten Sie diese E-Mail irrtümlich erhalten haben, ist Ihnen eine Kenntnisnahme des Inhalts, eine Vervielfältigung oder Weitergabe der E-Mail ausdrücklich untersagt. Bitte benachrichtigen Sie uns und vernichten Sie die empfangene E-Mail. Vielen Dank.

 

This e-mail may contain trade secrets or privileged, undisclosed, or otherwise confidential information. If you have received this e-mail in error, you are hereby notified that any review, copying, or distribution of it is strictly prohibited. Please inform us immediately and destroy the original transmittal. Thank you for your cooperation.

 

 

From: <cf-dev@...> on behalf of Abby Chau <achau@...>
Reply-To: "cf-dev@..." <cf-dev@...>
Date: Wednesday, 27. June 2018 at 18:18
To: "Discussions about Cloud Foundry projects and the system overall." <cf-dev@...>
Subject: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

 

Hi everyone,

 

The CF CLI is soliciting feedback on how you use the `no-start` flag: 

  • what are your specific use cases? 
  • if the flag was no longer available, how might that affect you? (Would you strongly object to such a change, and why?) would you have a sensible workaround?

 

Feedback

 

Please contact us on the #cli channel on Cloud Foundry Slack. Alternatively, respond directly to this email.



Best,



Abby Chau

CF Product Manager, CF CLI



 

Note: The CF CLI team is currently working on updating the `cf v3-push` command to use the server side manifest work we've completed (with the Capi team), and to update the command such that it has as much feature parity with the current `cf-push` as possible. The eventual goal is to make the `cf push` command default to the enhanced v3-push command for users (on cc api 3.27 or higher). Since there is more granular control options in v3, `no-start` is a more nebulous flag.
John Mcteague <john@...>
 

We have multiple foundries in our org and occasionally we would want to push an app in the stopped state to a pool for DR purposes (active/active across foundries is always desirable but simply not always practical for older applications transitioning to CF). 


On Thu, 28 Jun 2018, 00:58 Zach Robinson, <zrobinson@...> wrote:
thanks to all that have replied.

i'd like to ask a clarifying question.  it sounds like most of these use-cases are "set up some environment for the app".  this implies to me that these workflows are for new apps and not for updating existing ones.  does that sound right?  are their use-cases for "no-start" when updating an app?

thanks!
-zach

On Wed, Jun 27, 2018 at 4:30 PM Abby Chau <achau@...> wrote:
Hey Bernd and Julian,

Many thanks for your responses. 

Wondering if you folks have tried using the `v3-create-app` command as Mike has pointed out (and Zach elsewhere). The command is basically the equivalent of `cf push test --no-start` in that it creates an app, leaves it in a `stopped` state, and users can then apply configurations, bind services, set-groups, enable ssh, etc. This command appears to be a workaround for your use cases, but please let us know if that's incorrect. 

Best,

Abby

On Wed, Jun 27, 2018 at 3:04 PM, Hjortshoj, Julian <Julian.Hjortshoj@...> wrote:
+1 to that last use case.  In particular, if a service requires bind configuration, then it's not possible to put the binding in an app manifest (because manifests don't support service binding configuration) so it is very helpful to keep your app from starting until the service is bound.

Adding the ability to specify bind configuration in an application manifest might be a reasonable mitigation if --no-start were to disappear.

HTH,
-Julian
From: cf-dev@... [cf-dev@...] on behalf of Krannich, Bernd [bernd.krannich@...]
Sent: Wednesday, June 27, 2018 9:55 AM
To: cf-dev@...
Subject: Re: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

Hi Abby,

 

Thank you very much for reaching out. Indeed, we are frequent users of the `no-start` flag. A couple of use cases:

 

  • After a `cf ssh-enable APP_NAME`, you’ll need to restart your app to be able to SSH in. To prevent starting the app twice in that process, people do a `cf push --no-start` and then an explicit `cf start APP_NAME` afterwards.
  • The same applies for changes of security groups which also require an app restart.
  • Lastly, and maybe most importantly, people use `no-start` for pushing an app to which they want to bind services. They push the app with “no-start”, “cf create-service” their services, “cf bind-service” them to the app, and only then start the app.

 

So, I guess in summary, for all operations changing the app environment, doing those changes with the app not yet started is a common use case for us.

 

Regards,

Bernd

 

Bernd Krannich

SAP Cloud Platform

SAP SE

Dietmar-Hopp-Allee 16, 69190 Walldorf, Germany

 

E bernd.krannich@...

 

Pflichtangaben/Mandatory Disclosure Statement: www.sap.com/impressum

 

Diese E-Mail kann Betriebs- oder Geschäftsgeheimnisse oder sonstige vertrauliche Informationen enthalten. Sollten Sie diese E-Mail irrtümlich erhalten haben, ist Ihnen eine Kenntnisnahme des Inhalts, eine Vervielfältigung oder Weitergabe der E-Mail ausdrücklich untersagt. Bitte benachrichtigen Sie uns und vernichten Sie die empfangene E-Mail. Vielen Dank.

 

This e-mail may contain trade secrets or privileged, undisclosed, or otherwise confidential information. If you have received this e-mail in error, you are hereby notified that any review, copying, or distribution of it is strictly prohibited. Please inform us immediately and destroy the original transmittal. Thank you for your cooperation.

 

 

From: <cf-dev@...> on behalf of Abby Chau <achau@...>
Reply-To: "cf-dev@..." <cf-dev@...>
Date: Wednesday, 27. June 2018 at 18:18
To: "Discussions about Cloud Foundry projects and the system overall." <cf-dev@...>
Subject: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

 

Hi everyone,

 

The CF CLI is soliciting feedback on how you use the `no-start` flag: 

  • what are your specific use cases? 
  • if the flag was no longer available, how might that affect you? (Would you strongly object to such a change, and why?) would you have a sensible workaround?

 

Feedback

 

Please contact us on the #cli channel on Cloud Foundry Slack. Alternatively, respond directly to this email.



Best,



Abby Chau

CF Product Manager, CF CLI



 

Note: The CF CLI team is currently working on updating the `cf v3-push` command to use the server side manifest work we've completed (with the Capi team), and to update the command such that it has as much feature parity with the current `cf-push` as possible. The eventual goal is to make the `cf push` command default to the enhanced v3-push command for users (on cc api 3.27 or higher). Since there is more granular control options in v3, `no-start` is a more nebulous flag.
Krannich, Bernd <bernd.krannich@...>
 

Hi Abby, Zach,

 

I don’t think we have looked into v3 commands in too much detail. Please forgive my ignorance in this topic, are the v3 commands already considered a stable API so that I could check if people can replace their existing “--no-start” scenarios with it?

 

> this implies to me that these workflows are for new apps and not for updating existing ones.

IIRC, some of those workflows are done as part of blue-/green-updates. I am not sure that people would completely delete their currently inactive app completely, or if they would just keep updating the inactive app and then switch over to it.

 

Thanks in advance,

Bernd

 

From: <cf-dev@...> on behalf of Zach Robinson <zrobinson@...>
Reply-To: "cf-dev@..." <cf-dev@...>
Date: Thursday, 28. June 2018 at 01:58
To: "cf-dev@..." <cf-dev@...>
Subject: Re: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

 

thanks to all that have replied.

 

i'd like to ask a clarifying question.  it sounds like most of these use-cases are "set up some environment for the app".  this implies to me that these workflows are for new apps and not for updating existing ones.  does that sound right?  are their use-cases for "no-start" when updating an app?

 

thanks!

-zach

 

On Wed, Jun 27, 2018 at 4:30 PM Abby Chau <achau@...> wrote:

Hey Bernd and Julian,

 

Many thanks for your responses. 

 

Wondering if you folks have tried using the `v3-create-app` command as Mike has pointed out (and Zach elsewhere). The command is basically the equivalent of `cf push test --no-start` in that it creates an app, leaves it in a `stopped` state, and users can then apply configurations, bind services, set-groups, enable ssh, etc. This command appears to be a workaround for your use cases, but please let us know if that's incorrect. 



Best,



Abby

 

On Wed, Jun 27, 2018 at 3:04 PM, Hjortshoj, Julian <Julian.Hjortshoj@...> wrote:

+1 to that last use case.  In particular, if a service requires bind configuration, then it's not possible to put the binding in an app manifest (because manifests don't support service binding configuration) so it is very helpful to keep your app from starting until the service is bound.

 

Adding the ability to specify bind configuration in an application manifest might be a reasonable mitigation if --no-start were to disappear.

 

HTH,

-Julian

From: cf-dev@... [cf-dev@...] on behalf of Krannich, Bernd [bernd.krannich@...]
Sent: Wednesday, June 27, 2018 9:55 AM
To: cf-dev@...
Subject: Re: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

Hi Abby,

 

Thank you very much for reaching out. Indeed, we are frequent users of the `no-start` flag. A couple of use cases:

 

  • After a `cf ssh-enable APP_NAME`, you’ll need to restart your app to be able to SSH in. To prevent starting the app twice in that process, people do a `cf push --no-start` and then an explicit `cf start APP_NAME` afterwards.
  • The same applies for changes of security groups which also require an app restart.
  • Lastly, and maybe most importantly, people use `no-start` for pushing an app to which they want to bind services. They push the app with “no-start”, “cf create-service” their services, “cf bind-service” them to the app, and only then start the app.

 

So, I guess in summary, for all operations changing the app environment, doing those changes with the app not yet started is a common use case for us.

 

Regards,

Bernd

 

Bernd Krannich

SAP Cloud Platform

SAP SE

Dietmar-Hopp-Allee 16, 69190 Walldorf, Germany

 

E bernd.krannich@...

 

Pflichtangaben/Mandatory Disclosure Statement: www.sap.com/impressum

 

Diese E-Mail kann Betriebs- oder Geschäftsgeheimnisse oder sonstige vertrauliche Informationen enthalten. Sollten Sie diese E-Mail irrtümlich erhalten haben, ist Ihnen eine Kenntnisnahme des Inhalts, eine Vervielfältigung oder Weitergabe der E-Mail ausdrücklich untersagt. Bitte benachrichtigen Sie uns und vernichten Sie die empfangene E-Mail. Vielen Dank.

 

This e-mail may contain trade secrets or privileged, undisclosed, or otherwise confidential information. If you have received this e-mail in error, you are hereby notified that any review, copying, or distribution of it is strictly prohibited. Please inform us immediately and destroy the original transmittal. Thank you for your cooperation.

 

 

From: <cf-dev@...> on behalf of Abby Chau <achau@...>
Reply-To: "cf-dev@..." <cf-dev@...>
Date: Wednesday, 27. June 2018 at 18:18
To: "Discussions about Cloud Foundry projects and the system overall." <cf-dev@...>
Subject: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

 

Hi everyone,

 

The CF CLI is soliciting feedback on how you use the `no-start` flag: 

  • what are your specific use cases? 
  • if the flag was no longer available, how might that affect you? (Would you strongly object to such a change, and why?) would you have a sensible workaround?

 

Feedback

 

Please contact us on the #cli channel on Cloud Foundry Slack. Alternatively, respond directly to this email.

 

Best,

 

Abby Chau

CF Product Manager, CF CLI

 

 

Note: The CF CLI team is currently working on updating the `cf v3-push` command to use the server side manifest work we've completed (with the Capi team), and to update the command such that it has as much feature parity with the current `cf-push` as possible. The eventual goal is to make the `cf push` command default to the enhanced v3-push command for users (on cc api 3.27 or higher). Since there is more granular control options in v3, `no-start` is a more nebulous flag.

 
Chip Childers <cchilders@...>
 

On Wed, Jun 27, 2018 at 7:30 PM Abby Chau <achau@...> wrote:
Wondering if you folks have tried using the `v3-create-app` command as Mike has pointed out (and Zach elsewhere). The command is basically the equivalent of `cf push test --no-start` in that it creates an app, leaves it in a `stopped` state, and users can then apply configurations, bind services, set-groups, enable ssh, etc. This command appears to be a workaround for your use cases, but please let us know if that's incorrect. 

My 2 cents here, FWIW... the v3-foo$ syntax feels like it should be only temporary and should eventually be merged back into the more standard commands / feature flag combos.
--
Chip Childers
CTO, Cloud Foundry Foundation
1.267.250.0815
Christian Brinker
 

Hi,

+1 on chips idea. ;-)

v3-foo was fun but we should get rid of that syntax.

Greetings
Christian

On Thu, 28 Jun 2018 at 16:50 Chip Childers <cchilders@...> wrote:
On Wed, Jun 27, 2018 at 7:30 PM Abby Chau <achau@...> wrote:
Wondering if you folks have tried using the `v3-create-app` command as Mike has pointed out (and Zach elsewhere). The command is basically the equivalent of `cf push test --no-start` in that it creates an app, leaves it in a `stopped` state, and users can then apply configurations, bind services, set-groups, enable ssh, etc. This command appears to be a workaround for your use cases, but please let us know if that's incorrect. 

My 2 cents here, FWIW... the v3-foo$ syntax feels like it should be only temporary and should eventually be merged back into the more standard commands / feature flag combos.
--
Chip Childers
CTO, Cloud Foundry Foundation
1.267.250.0815

--
E-Mail: cbrinker@...

evoila GmbH
Weberstraße 21
55130 Mainz
Germany

Geschäftsführer: Johannes Hiemer

Amtsgericht Mainz HRB 42719

Diese E-Mail enthält vertrauliche und/oder rechtlich geschützte Informationen. Wenn Sie nicht der richtige Adressat sind oder diese E-Mail irrtümlich erhalten haben, informieren Sie bitte sofort den Absender und vernichten Sie diese Mail. Das unerlaubte Kopieren sowie die unbefugte Weitergabe dieser Mail ist nicht gestattet.

This e-mail may contain confidential and/or privileged information. If You are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and destroy this e-mail. Any unauthorised copying, disclosure or distribution of the material in this e-mail is strictly forbidden.
Abby Chau
 

Hey Bernd,

Zach can speak more to this but, I believe, the CC v3 API has been stable for some time, with v3.27. 

The v3 prefixed CF CLI commands are relatively lesser known so apologies if it's been relatively unclear. All the CF CLI v3 prefixed commands are all still considered experimental (the CLI commands such as `cf v3-push`, `cf v3-app`, `cf v3-set-health-check`, `cf v3-droplets`, etc) . A few reasons for this are that the CF CLI team wanted to garner feedback about them from users, and we wanted the freedom to change the commands (output, etc) based on user feedback. An example of this is the `cf v3-app` command, which we will be changing soon based on user feedback. (There is a survey out currently about the UX). 

Starting with the `cf v3-app` command, (which we will remove the experimental status for soon, and which the current `cf app` command will default to using for users on CC API 3.27. or higher) the CF CLI team have embarked on a concerted effort to remove experimental status for all the other v3 prefixed commands. We realise this may have an impact on users, and we want to mitigate for this as much as possible - we are doing so by asking the community for feedback, and updating them as much as possible regarding our current efforts. 

After we release the work for the `cf v3-app`, we will be looking at other CF CLI v3 prefixed commands to remove the experimental status for - again we are keen on making sure impact is minimal for users whilst simultaneously trying to make sure all the benefits of the v3 CC API functionality is fully realised as intended. 

I hope this makes sense, please do not hesitate to reach out if you have any questions or feedback. We will be communicating with the community each step of the way. 

Best,

Abby


On Thu, Jun 28, 2018 at 7:50 AM, Chip Childers <cchilders@...> wrote:
On Wed, Jun 27, 2018 at 7:30 PM Abby Chau <achau@...> wrote:
Wondering if you folks have tried using the `v3-create-app` command as Mike has pointed out (and Zach elsewhere). The command is basically the equivalent of `cf push test --no-start` in that it creates an app, leaves it in a `stopped` state, and users can then apply configurations, bind services, set-groups, enable ssh, etc. This command appears to be a workaround for your use cases, but please let us know if that's incorrect. 

My 2 cents here, FWIW... the v3-foo$ syntax feels like it should be only temporary and should eventually be merged back into the more standard commands / feature flag combos.
--
Chip Childers
CTO, Cloud Foundry Foundation
1.267.250.0815
Hjortshoj, Julian <Julian.Hjortshoj@...>
 

We have not tried the v3 api, but yes, that sounds correct to me, and v3-create-app seems like a reasonable substitute.
From: cf-dev@... [cf-dev@...] on behalf of Zach Robinson [zrobinson@...]
Sent: Wednesday, June 27, 2018 4:58 PM
To: cf-dev@...
Subject: Re: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

thanks to all that have replied.

i'd like to ask a clarifying question.  it sounds like most of these use-cases are "set up some environment for the app".  this implies to me that these workflows are for new apps and not for updating existing ones.  does that sound right?  are their use-cases for "no-start" when updating an app?

thanks!
-zach

On Wed, Jun 27, 2018 at 4:30 PM Abby Chau <achau@...> wrote:

Hey Bernd and Julian,

Many thanks for your responses. 

Wondering if you folks have tried using the `v3-create-app` command as Mike has pointed out (and Zach elsewhere). The command is basically the equivalent of `cf push test --no-start` in that it creates an app, leaves it in a `stopped` state, and users can then apply configurations, bind services, set-groups, enable ssh, etc. This command appears to be a workaround for your use cases, but please let us know if that's incorrect. 

Best,

Abby

On Wed, Jun 27, 2018 at 3:04 PM, Hjortshoj, Julian <Julian.Hjortshoj@...> wrote:
+1 to that last use case.  In particular, if a service requires bind configuration, then it's not possible to put the binding in an app manifest (because manifests don't support service binding configuration) so it is very helpful to keep your app from starting until the service is bound.

Adding the ability to specify bind configuration in an application manifest might be a reasonable mitigation if --no-start were to disappear.

HTH,
-Julian
From: cf-dev@... [cf-dev@...] on behalf of Krannich, Bernd [bernd.krannich@...]
Sent: Wednesday, June 27, 2018 9:55 AM
To: cf-dev@...
Subject: Re: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

Hi Abby,

 

Thank you very much for reaching out. Indeed, we are frequent users of the `no-start` flag. A couple of use cases:

 

  • After a `cf ssh-enable APP_NAME`, you’ll need to restart your app to be able to SSH in. To prevent starting the app twice in that process, people do a `cf push --no-start` and then an explicit `cf start APP_NAME` afterwards.
  • The same applies for changes of security groups which also require an app restart.
  • Lastly, and maybe most importantly, people use `no-start` for pushing an app to which they want to bind services. They push the app with “no-start”, “cf create-service” their services, “cf bind-service” them to the app, and only then start the app.

 

So, I guess in summary, for all operations changing the app environment, doing those changes with the app not yet started is a common use case for us.

 

Regards,

Bernd

 

Bernd Krannich

SAP Cloud Platform

SAP SE

Dietmar-Hopp-Allee 16, 69190 Walldorf, Germany

 

E bernd.krannich@...

 

Pflichtangaben/Mandatory Disclosure Statement: www.sap.com/impressum

 

Diese E-Mail kann Betriebs- oder Geschäftsgeheimnisse oder sonstige vertrauliche Informationen enthalten. Sollten Sie diese E-Mail irrtümlich erhalten haben, ist Ihnen eine Kenntnisnahme des Inhalts, eine Vervielfältigung oder Weitergabe der E-Mail ausdrücklich untersagt. Bitte benachrichtigen Sie uns und vernichten Sie die empfangene E-Mail. Vielen Dank.

 

This e-mail may contain trade secrets or privileged, undisclosed, or otherwise confidential information. If you have received this e-mail in error, you are hereby notified that any review, copying, or distribution of it is strictly prohibited. Please inform us immediately and destroy the original transmittal. Thank you for your cooperation.

 

 

From: <cf-dev@...> on behalf of Abby Chau <achau@...>
Reply-To: "cf-dev@..." <cf-dev@...>
Date: Wednesday, 27. June 2018 at 18:18
To: "Discussions about Cloud Foundry projects and the system overall." <cf-dev@...>
Subject: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

 

Hi everyone,

 

The CF CLI is soliciting feedback on how you use the `no-start` flag: 

  • what are your specific use cases? 
  • if the flag was no longer available, how might that affect you? (Would you strongly object to such a change, and why?) would you have a sensible workaround?

 

Feedback

 

Please contact us on the #cli channel on Cloud Foundry Slack. Alternatively, respond directly to this email.



Best,



Abby Chau

CF Product Manager, CF CLI



 

Note: The CF CLI team is currently working on updating the `cf v3-push` command to use the server side manifest work we've completed (with the Capi team), and to update the command such that it has as much feature parity with the current `cf-push` as possible. The eventual goal is to make the `cf push` command default to the enhanced v3-push command for users (on cc api 3.27 or higher). Since there is more granular control options in v3, `no-start` is a more nebulous flag.
Dr Nic Williams <drnicwilliams@...>
 

I use no-start because ‘cf push’ doesn’t have flags to set env vars/service bindings.

If the response to that is “use manifest.yml” then perhaps get rid off all the push flags and force ppl to use manifest.yml. As it stands now, we can cf push most attributes of the manifest.yml, except env vars and service bindings.

From: cf-dev@... <cf-dev@...> on behalf of Hjortshoj, Julian <Julian.Hjortshoj@...>
Sent: Friday, June 29, 2018 3:06:03 AM
To: cf-dev@...
Subject: Re: [cf-dev] CF CLI: how do you use `no-start` with `cf push`
 
We have not tried the v3 api, but yes, that sounds correct to me, and v3-create-app seems like a reasonable substitute.
From: cf-dev@... [cf-dev@...] on behalf of Zach Robinson [zrobinson@...]
Sent: Wednesday, June 27, 2018 4:58 PM
To: cf-dev@...
Subject: Re: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

thanks to all that have replied.

i'd like to ask a clarifying question.  it sounds like most of these use-cases are "set up some environment for the app".  this implies to me that these workflows are for new apps and not for updating existing ones.  does that sound right?  are their use-cases for "no-start" when updating an app?

thanks!
-zach

On Wed, Jun 27, 2018 at 4:30 PM Abby Chau <achau@...> wrote:
Hey Bernd and Julian,

Many thanks for your responses. 

Wondering if you folks have tried using the `v3-create-app` command as Mike has pointed out (and Zach elsewhere). The command is basically the equivalent of `cf push test --no-start` in that it creates an app, leaves it in a `stopped` state, and users can then apply configurations, bind services, set-groups, enable ssh, etc. This command appears to be a workaround for your use cases, but please let us know if that's incorrect. 

Best,

Abby

On Wed, Jun 27, 2018 at 3:04 PM, Hjortshoj, Julian <Julian.Hjortshoj@...> wrote:
+1 to that last use case.  In particular, if a service requires bind configuration, then it's not possible to put the binding in an app manifest (because manifests don't support service binding configuration) so it is very helpful to keep your app from starting until the service is bound.

Adding the ability to specify bind configuration in an application manifest might be a reasonable mitigation if --no-start were to disappear.

HTH,
-Julian
From: cf-dev@... [cf-dev@...] on behalf of Krannich, Bernd [bernd.krannich@...]
Sent: Wednesday, June 27, 2018 9:55 AM
To: cf-dev@...
Subject: Re: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

Hi Abby,

 

Thank you very much for reaching out. Indeed, we are frequent users of the `no-start` flag. A couple of use cases:

 

  • After a `cf ssh-enable APP_NAME`, you’ll need to restart your app to be able to SSH in. To prevent starting the app twice in that process, people do a `cf push --no-start` and then an explicit `cf start APP_NAME` afterwards.
  • The same applies for changes of security groups which also require an app restart.
  • Lastly, and maybe most importantly, people use `no-start` for pushing an app to which they want to bind services. They push the app with “no-start”, “cf create-service” their services, “cf bind-service” them to the app, and only then start the app.

 

So, I guess in summary, for all operations changing the app environment, doing those changes with the app not yet started is a common use case for us.

 

Regards,

Bernd

 

Bernd Krannich

SAP Cloud Platform

SAP SE

Dietmar-Hopp-Allee 16, 69190 Walldorf, Germany

 

E bernd.krannich@...

 

Pflichtangaben/Mandatory Disclosure Statement: www.sap.com/impressum

 

Diese E-Mail kann Betriebs- oder Geschäftsgeheimnisse oder sonstige vertrauliche Informationen enthalten. Sollten Sie diese E-Mail irrtümlich erhalten haben, ist Ihnen eine Kenntnisnahme des Inhalts, eine Vervielfältigung oder Weitergabe der E-Mail ausdrücklich untersagt. Bitte benachrichtigen Sie uns und vernichten Sie die empfangene E-Mail. Vielen Dank.

 

This e-mail may contain trade secrets or privileged, undisclosed, or otherwise confidential information. If you have received this e-mail in error, you are hereby notified that any review, copying, or distribution of it is strictly prohibited. Please inform us immediately and destroy the original transmittal. Thank you for your cooperation.

 

 

From: <cf-dev@...> on behalf of Abby Chau <achau@...>
Reply-To: "cf-dev@..." <cf-dev@...>
Date: Wednesday, 27. June 2018 at 18:18
To: "Discussions about Cloud Foundry projects and the system overall." <cf-dev@...>
Subject: [cf-dev] CF CLI: how do you use `no-start` with `cf push`

 

Hi everyone,

 

The CF CLI is soliciting feedback on how you use the `no-start` flag: 

  • what are your specific use cases? 
  • if the flag was no longer available, how might that affect you? (Would you strongly object to such a change, and why?) would you have a sensible workaround?

 

Feedback

 

Please contact us on the #cli channel on Cloud Foundry Slack. Alternatively, respond directly to this email.



Best,



Abby Chau

CF Product Manager, CF CLI



 

Note: The CF CLI team is currently working on updating the `cf v3-push` command to use the server side manifest work we've completed (with the Capi team), and to update the command such that it has as much feature parity with the current `cf-push` as possible. The eventual goal is to make the `cf push` command default to the enhanced v3-push command for users (on cc api 3.27 or higher). Since there is more granular control options in v3, `no-start` is a more nebulous flag.
1 - 15 of 15