from:"ashish.jain14"


Hello,


I have got a nodejs application which is deployed on openstack provisioned 
resources. Somehow whenever my response has a large payload I am unable to 
receive the response to my web browser however I am able to print the response 
at the nodejs server end which is running in openstack. I am using LBAAS from 
neutron and use the IP of LBAAS to hit nodejs.


Scenario1: I send a request for an id which exist in DB. DB returns json data 
which is delivered to nodejs. However unable to send it to browser running on 
my host. I get a 504 error


Scenario2: I send a request for an id which DOES NOT exist in DB. This time I 
get back the json response all the way till browser.


To summarize all the communications are working fine, only difference between 
the 2 scenarios is payload size. I have been playing around with MTU on network 
as well as the compute node for br-tun and eth1 but till now have not succeeded 
. I am also trying out few things with tcpdump but nothing substantial I have 
found till now.


Need some help on how to debug this out.


Regards
Ashish

The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Deploying a cluster using openstack heat

Hello,


Does heat provide capabilities to deploy a cluster? The use case is to deploy a 
mongo db cluster. Any pointers would be helpful.


Regards

Ashish

The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] [Potential Spoofed] Re: [Potential Spoofed] Issue with receiving response with large payload from a node js application running on openstack

Michael,


Thanks for your response and help with steps to debug the issue.. The issue is 
resolved now by increasing the MTU size of GRE tunnel.


Regards

Ashish


From: Michael Krotscheck krotsch...@gmail.com
Sent: Tuesday, May 12, 2015 8:44 PM
To: Ashish Jain (WT01 - BAS); openstack@lists.openstack.org
Subject: Re: [Openstack] [Potential Spoofed] Re: [Potential Spoofed] Issue 
with receiving response with large payload from a node js application running 
on openstack

Hey there, Ashish-

I'm afraid I have never encountered this issue before myself, but I can assist 
with debugging. The first step would be to isolate whether the problem is with 
OpenStack's Nova, LBaaS, NodeJS, your application code, or the browser itself. 
This could be as easy as trying it from a different browser - firefox vs. 
chrome for instance - and if the response in one is correct and the other is 
not, then you have your culprit. If that doesn't work, you'll need to capture 
the entire HTTP request (I recommend Chrome's Network Debugging tools), and 
replay the request in a tool like curl or Postman. With this method, you may be 
able to recreate the issue by selectively adding/removing HTTP request headers.

At this point, you've got some clues. If there is a specific header that causes 
the request to behave oddly, then it's likely that the problem is on the server 
side rather than the client side. If the issue persists only in the browser, 
and not curl/postman, then your issue is with your client runtime and/or 
application code.

From that point forward, continue isolating components to see if they 
contribute to the issue. Armed with the knowledge of which component is 
causing your issue, you can solicit specific answers from the appropriate 
community :). Good luck!

Michael

On Tue, May 12, 2015 at 2:23 AM 
ashish.jai...@wipro.commailto:ashish.jai...@wipro.com wrote:
Another point to add If I use curl I can see the total response size I could 
receive is 906 bytes which means 227 bytes for header and  payload size of 679 
bytes. Any clues anyone?


From: ashish.jai...@wipro.commailto:ashish.jai...@wipro.com 
ashish.jai...@wipro.commailto:ashish.jai...@wipro.com
Sent: Tuesday, May 12, 2015 2:05 PM
To: openstack@lists.openstack.orgmailto:openstack@lists.openstack.org
Subject: [Potential Spoofed] Re: [Openstack] [Potential Spoofed] Issue with 
receiving response with large payload from a node js application running on 
openstack

I just figured out that if the payload size is greater than 623 , I do not get 
any response back. The size of the header is 227 bytes. So the total response 
size sent all the way to my browser is 850 bytes. Anything above this is not 
received  by browser.


From: ashish.jai...@wipro.commailto:ashish.jai...@wipro.com 
ashish.jai...@wipro.commailto:ashish.jai...@wipro.com
Sent: Tuesday, May 12, 2015 11:29 AM
To: openstack@lists.openstack.orgmailto:openstack@lists.openstack.org
Subject: [Potential Spoofed] [Openstack] Issue with receiving response with 
large payload from a node js application running on openstack

Hello,


I have got a nodejs application which is deployed on openstack provisioned 
resources. Somehow whenever my response has a large payload I am unable to 
receive the response to my web browser however I am able to print the response 
at the nodejs server end which is running in openstack. I am using LBAAS from 
neutron and use the IP of LBAAS to hit nodejs.


Scenario1: I send a request for an id which exist in DB. DB returns json data 
which is delivered to nodejs. However unable to send it to browser running on 
my host. I get a 504 error


Scenario2: I send a request for an id which DOES NOT exist in DB. This time I 
get back the json response all the way till browser.


To summarize all the communications are working fine, only difference between 
the 2 scenarios is payload size. I have been playing around with MTU on network 
as well as the compute node for br-tun and eth1 but till now have not succeeded 
. I am also trying out few things with tcpdump but nothing substantial I have 
found till now.


Need some help on how to debug this out.


Regards
Ashish

The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.comhttp://www.wipro.com

Re: [Openstack] [Potential Spoofed] Issue with receiving response with large payload from a node js application running on openstack

I just figured out that if the payload size is greater than 623 , I do not get 
any response back. The size of the header is 227 bytes. So the total response 
size sent all the way to my browser is 850 bytes. Anything above this is not 
received  by browser.

From: ashish.jai...@wipro.com ashish.jai...@wipro.com
Sent: Tuesday, May 12, 2015 11:29 AM
To: openstack@lists.openstack.org
Subject: [Potential Spoofed] [Openstack] Issue with receiving response with 
large payload from a node js application running on openstack

Hello,

I have got a nodejs application which is deployed on openstack provisioned 
resources. Somehow whenever my response has a large payload I am unable to 
receive the response to my web browser however I am able to print the response 
at the nodejs server end which is running in openstack. I am using LBAAS from 
neutron and use the IP of LBAAS to hit nodejs.

Scenario1: I send a request for an id which exist in DB. DB returns json data 
which is delivered to nodejs. However unable to send it to browser running on 
my host. I get a 504 error

Scenario2: I send a request for an id which DOES NOT exist in DB. This time I 
get back the json response all the way till browser.

To summarize all the communications are working fine, only difference between 
the 2 scenarios is payload size. I have been playing around with MTU on network 
as well as the compute node for br-tun and eth1 but till now have not succeeded 
. I am also trying out few things with tcpdump but nothing substantial I have 
found till now.

Need some help on how to debug this out.

Regards
Ashish

The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] [Potential Spoofed] Re: [Potential Spoofed] Issue with receiving response with large payload from a node js application running on openstack

Another point to add If I use curl I can see the total response size I could 
receive is 906 bytes which means 227 bytes for header and  payload size of 679 
bytes. Any clues anyone?

From: ashish.jai...@wipro.com ashish.jai...@wipro.com
Sent: Tuesday, May 12, 2015 2:05 PM
To: openstack@lists.openstack.org
Subject: [Potential Spoofed] Re: [Openstack] [Potential Spoofed] Issue with 
receiving response with large payload from a node js application running on 
openstack

I just figured out that if the payload size is greater than 623 , I do not get 
any response back. The size of the header is 227 bytes. So the total response 
size sent all the way to my browser is 850 bytes. Anything above this is not 
received  by browser.

From: ashish.jai...@wipro.com ashish.jai...@wipro.com
Sent: Tuesday, May 12, 2015 11:29 AM
To: openstack@lists.openstack.org
Subject: [Potential Spoofed] [Openstack] Issue with receiving response with 
large payload from a node js application running on openstack

Hello,

I have got a nodejs application which is deployed on openstack provisioned 
resources. Somehow whenever my response has a large payload I am unable to 
receive the response to my web browser however I am able to print the response 
at the nodejs server end which is running in openstack. I am using LBAAS from 
neutron and use the IP of LBAAS to hit nodejs.

Scenario1: I send a request for an id which exist in DB. DB returns json data 
which is delivered to nodejs. However unable to send it to browser running on 
my host. I get a 504 error

Scenario2: I send a request for an id which DOES NOT exist in DB. This time I 
get back the json response all the way till browser.

To summarize all the communications are working fine, only difference between 
the 2 scenarios is payload size. I have been playing around with MTU on network 
as well as the compute node for br-tun and eth1 but till now have not succeeded 
. I am also trying out few things with tcpdump but nothing substantial I have 
found till now.

Need some help on how to debug this out.

Regards
Ashish

The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] [Potential Spoofed] Re: heat: Using IP address of one VM with-in another VM

Just to add another point I am using docker container here as the image. Can 
that be a issue.

From: Ashish Jain (WT01 - BAS)
Sent: Thursday, May 07, 2015 12:56 PM
To: chris.frie...@windriver.com; openstack@lists.openstack.org
Subject: Re: [Potential Spoofed] Re: [Openstack] heat: Using IP address of 
one VM with-in another VM

Hello,

Somehow I see that user data section is not working for me. Rest all works fine 
 Here is a simple template.

heat_template_version: 2013-05-23
description: A simple server.
resources:
  server:
type: OS::Nova::Server
properties:
  flavor: m1.tiny
  networks:
- network: demo-net
  key_name: demo-key-new
  image: httpd
  user_data:
#!/bin/bash
mkdir -p /opt/test
touch /usr/bin/test/1.txt
echo hello  /usr/bin/test/1.txt

outputs:
  HttpIP:
description: IP of HTTP server
value:
  str_replace:
template: http://host
  params:
host: { get_attr: [server, first_address] }

Regards
Ashish

From: ashish.jai...@wipro.com ashish.jai...@wipro.com
Sent: Wednesday, May 06, 2015 9:12 PM
To: chris.frie...@windriver.com; openstack@lists.openstack.org
Subject: [Potential Spoofed] Re: [Openstack] heat: Using IP address of one VM 
with-in another VM

Thanks a lot Chris. That is helpful.

Regards
Ashish

From: Chris Friesen chris.frie...@windriver.com
Sent: Wednesday, May 06, 2015 7:16 PM
To: openstack@lists.openstack.org
Subject: Re: [Openstack] heat: Using IP address of one VM with-in another VM

On 05/06/2015 06:20 AM, ashish.jai...@wipro.com wrote:
 Hi,

 Is it possible to somehow pass IP of a VM to another VM during heat 
 deployment.
 For example IP of  a DB server in a VM to a client in another VM.

Yes.  As an example take a look at
https://github.com/openstack/heat-templates/blob/master/hot/F20/WordPress_2_Instances.yaml;.

The WebServer is configured with the IP address of the DatabaseServer:

db_ipaddr: { get_attr: [DatabaseServer, networks, private, 0] }

Chris

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] where are the logs for user data section of heat

Hello,


Where can I find the logs for user_data section of heat template? I have enable 
debug mode logging in heat.conf but I still do not see any logging for 
user_data section of heat template.


Regards

Ashish

The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] [Potential Spoofed] Re: heat: Using IP address of one VM with-in another VM

Hello,

Somehow I see that user data section is not working for me. Rest all works fine 
 Here is a simple template.

heat_template_version: 2013-05-23
description: A simple server.
resources:
  server:
type: OS::Nova::Server
properties:
  flavor: m1.tiny
  networks:
- network: demo-net
  key_name: demo-key-new
  image: httpd
  user_data:
#!/bin/bash
mkdir -p /opt/test
touch /usr/bin/test/1.txt
echo hello  /usr/bin/test/1.txt

outputs:
  HttpIP:
description: IP of HTTP server
value:
  str_replace:
template: http://host
  params:
host: { get_attr: [server, first_address] }


Regards
Ashish




From: ashish.jai...@wipro.com ashish.jai...@wipro.com
Sent: Wednesday, May 06, 2015 9:12 PM
To: chris.frie...@windriver.com; openstack@lists.openstack.org
Subject: [Potential Spoofed] Re: [Openstack] heat: Using IP address of one VM 
with-in another VM

Thanks a lot Chris. That is helpful.

Regards
Ashish

From: Chris Friesen chris.frie...@windriver.com
Sent: Wednesday, May 06, 2015 7:16 PM
To: openstack@lists.openstack.org
Subject: Re: [Openstack] heat: Using IP address of one VM with-in another VM

On 05/06/2015 06:20 AM, ashish.jai...@wipro.com wrote:
 Hi,


 Is it possible to somehow pass IP of a VM to another VM during heat 
 deployment.
 For example IP of  a DB server in a VM to a client in another VM.

Yes.  As an example take a look at
https://github.com/openstack/heat-templates/blob/master/hot/F20/WordPress_2_Instances.yaml;.

The WebServer is configured with the IP address of the DatabaseServer:

db_ipaddr: { get_attr: [DatabaseServer, networks, private, 0] }


Chris


___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] [Potential Spoofed] Re: [Potential Spoofed] Re: heat: Using IP address of one VM with-in another VM

Does the images need to have cloud-init pre-installed for user_data section to 
work?

From: ashish.jai...@wipro.com ashish.jai...@wipro.com
Sent: Thursday, May 07, 2015 1:21 PM
To: chris.frie...@windriver.com; openstack@lists.openstack.org
Subject: [Potential Spoofed] Re: [Openstack] [Potential Spoofed] Re: heat: 
Using IP address of one VM with-in another VM

Just to add another point I am using docker container here as the image. Can 
that be a issue.

From: Ashish Jain (WT01 - BAS)
Sent: Thursday, May 07, 2015 12:56 PM
To: chris.frie...@windriver.com; openstack@lists.openstack.org
Subject: Re: [Potential Spoofed] Re: [Openstack] heat: Using IP address of 
one VM with-in another VM

Hello,

Somehow I see that user data section is not working for me. Rest all works fine 
 Here is a simple template.

heat_template_version: 2013-05-23
description: A simple server.
resources:
  server:
type: OS::Nova::Server
properties:
  flavor: m1.tiny
  networks:
- network: demo-net
  key_name: demo-key-new
  image: httpd
  user_data:
#!/bin/bash
mkdir -p /opt/test
touch /usr/bin/test/1.txt
echo hello  /usr/bin/test/1.txt

outputs:
  HttpIP:
description: IP of HTTP server
value:
  str_replace:
template: http://host
  params:
host: { get_attr: [server, first_address] }

Regards
Ashish

From: ashish.jai...@wipro.com ashish.jai...@wipro.com
Sent: Wednesday, May 06, 2015 9:12 PM
To: chris.frie...@windriver.com; openstack@lists.openstack.org
Subject: [Potential Spoofed] Re: [Openstack] heat: Using IP address of one VM 
with-in another VM

Thanks a lot Chris. That is helpful.

Regards
Ashish

From: Chris Friesen chris.frie...@windriver.com
Sent: Wednesday, May 06, 2015 7:16 PM
To: openstack@lists.openstack.org
Subject: Re: [Openstack] heat: Using IP address of one VM with-in another VM

On 05/06/2015 06:20 AM, ashish.jai...@wipro.com wrote:
 Hi,

 Is it possible to somehow pass IP of a VM to another VM during heat 
 deployment.
 For example IP of  a DB server in a VM to a client in another VM.

Yes.  As an example take a look at
https://github.com/openstack/heat-templates/blob/master/hot/F20/WordPress_2_Instances.yaml;.

The WebServer is configured with the IP address of the DatabaseServer:

db_ipaddr: { get_attr: [DatabaseServer, networks, private, 0] }

Chris

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies

Re: [Openstack] [“Potential Spoofed”] Neutron Load Balancer behavior while autoscaling


Hello,


I just tested with manually attaching an instance to a neutron load balancer 
pool. But somehow I see the same behavior where the instance attached while 
Jmeter is pumping messages does not serve any request. However same instance 
servers when I restart message pumping from Jmeter.


PS: I am using ha proxy.


Regards
Ashish



From: ashish.jai...@wipro.com ashish.jai...@wipro.com
Sent: Wednesday, May 06, 2015 11:47 AM
To: openstack@lists.openstack.org
Subject: [“Potential Spoofed”] [Openstack] Neutron Load Balancer behavior while 
autoscaling


Hello,


I am using neutron load balancer along with heat. I auto-scale instances and 
all the new instances are automatically  attached to neutron load balancer. I 
am using apache jmeter to create CPU load.


I instantiate apache jmeter for 5 minutes to create load and I am able to auto 
spawn instances and able to attach them to neutron load balancer, however the 
load is not distributed to all the instances ( I am using round robin policy).


Once Jmeter run in complete I start jmeter again and this time the load is 
distributed evenly to all the instances


Anyone seen this behavior, any clues what could be wrong.


Regards
Ashish

  The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should  not disseminate, distribute or copy this 
e-mail. Please notify the sender immediately and destroy all copies of this 
message and any attachments. WARNING: Computer viruses can be transmitted via 
email. The recipient should check this email and any attachments  for the 
presence of viruses. The company accepts no liability for any damage caused by 
any virus transmitted by this email. www.wipro.com
The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Neutron Load Balancer behavior while autoscaling

Hello,


I am using neutron load balancer along with heat. I auto-scale instances and 
all the new instances are automatically  attached to neutron load balancer. I 
am using apache jmeter to create CPU load.


I instantiate apache jmeter for 5 minutes to create load and I am able to auto 
spawn instances and able to attach them to neutron load balancer, however the 
load is not distributed to all the instances ( I am using round robin policy).


Once Jmeter run in complete I start jmeter again and this time the load is 
distributed evenly to all the instances


Anyone seen this behavior, any clues what could be wrong.


Regards

Ashish

The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] [“Potential Spoofed”] Re: [“Potential Spoofed”] Neutron Load Balancer behavior while autoscaling

Looks like the issue is with Jmeter and Lbaas compatibility, I wrote a simple 
shell script to simulate a similar load situation and somehow I do not see the 
issue any more.

Regards
Ashish

From: ashish.jai...@wipro.com ashish.jai...@wipro.com
Sent: Wednesday, May 06, 2015 3:15 PM
To: openstack@lists.openstack.org
Subject: [“Potential Spoofed”] Re: [Openstack] [“Potential Spoofed”]  Neutron 
Load Balancer behavior while autoscaling

Hello,

I just tested with manually attaching an instance to a neutron load balancer 
pool. But somehow I see the same behavior where the instance attached while 
Jmeter is pumping messages does not serve any request. However same instance 
servers when I restart message pumping from Jmeter.

PS: I am using ha proxy.

Regards
Ashish

From: ashish.jai...@wipro.com ashish.jai...@wipro.com
Sent: Wednesday, May 06, 2015 11:47 AM
To: openstack@lists.openstack.org
Subject: [“Potential Spoofed”] [Openstack] Neutron Load Balancer behavior while 
autoscaling

Hello,

I am using neutron load balancer along with heat. I auto-scale instances and 
all the new instances are automatically  attached to neutron load balancer. I 
am using apache jmeter to create CPU load.

I instantiate apache jmeter for 5 minutes to create load and I am able to auto 
spawn instances and able to attach them to neutron load balancer, however the 
load is not distributed to all the instances ( I am using round robin policy).

Once Jmeter run in complete I start jmeter again and this time the load is 
distributed evenly to all the instances

Anyone seen this behavior, any clues what could be wrong.

Regards
Ashish

  The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should  not disseminate, distribute or copy this 
e-mail. Please notify the sender immediately and destroy all copies of this 
message and any attachments. WARNING: Computer viruses can be transmitted via 
email. The recipient should check this email and any attachments  for the 
presence of viruses. The company accepts no liability for any damage caused by 
any virus transmitted by this email. www.wipro.com
The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] heat: Using IP address of one VM with-in another VM

Hi,


Is it possible to somehow pass IP of a VM to another VM during heat deployment. 
For example IP of  a DB server in a VM to a client in another VM.


Regards

Ashish

The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] heat: Using IP address of one VM with-in another VM

Thanks a lot Chris. That is helpful.

Regards
Ashish

From: Chris Friesen chris.frie...@windriver.com
Sent: Wednesday, May 06, 2015 7:16 PM
To: openstack@lists.openstack.org
Subject: Re: [Openstack] heat: Using IP address of one VM with-in another VM

On 05/06/2015 06:20 AM, ashish.jai...@wipro.com wrote:
 Hi,

 Is it possible to somehow pass IP of a VM to another VM during heat 
 deployment.
 For example IP of  a DB server in a VM to a client in another VM.

Yes.  As an example take a look at
https://github.com/openstack/heat-templates/blob/master/hot/F20/WordPress_2_Instances.yaml;.

The WebServer is configured with the IP address of the DatabaseServer:

db_ipaddr: { get_attr: [DatabaseServer, networks, private, 0] }

Chris

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [openstack-dev] [nova-docker] Where is the docker cgroups directory?

2015-04-21 Thread ashish.jain14

Hi,


Somehow restarting compute node helped.


Regards

Ashish


From: Ashish Jain (WT01 - BAS)
Sent: Tuesday, April 21, 2015 9:06 PM
To: openstack-dev@lists.openstack.org
Subject: [nova-docker] Where is the docker cgroups directory?


Hello,


I am using nova-docker and unable to figure out where is the cgroups docker 
directory /sys/fs/cgroup/cpuacct/docker when using nova-docker hypervisor?

While running a standalone docker installation the directory is located as 
pointed above. This is the directory used by docker to store cpu stats. Looking 
for something similar when using nova-docker.


Regards

Ashish

The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.com
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

[openstack-dev] [nova-docker][ceilometer][heat] Autoscaling docker in openstack

Hi,

I have been working on running docker on openstack. I had a discussion on
multiple IRC and IIUC there are 5 different ways of running docker on
openstack. IIUC currently there is no way to autoscale docker on openstack.
Please correct me if I am wrong

1) Using nova-docker driver - Running docker as a Nova::Server using
nova-docker hypervisor
2) Using nova-plugin for heat - Running docker using
DockerInc::Docker::Container
3) Using magnum - IIUC no automation as of now, manually it is possible. Not
enough documentation available
4) heat compose - Saw some samples available
@https://github.com/openstack/heat-templates/tree/master/hot/software-config/elements/heat-config-docker-compose
5) Swarm support - Still in development

Issues with each on the above approaches

1) Using nova-docker driver - IIUC there is no way for ceilometer to collect
and emit statistics for docker hypervisor. So that mean ceilometer does not
have any stats available once you switch to docker driver.
This link
(https://github.com/openstack/ceilometer/tree/master/ceilometer/compute/virt)
currently does not have anything for docker hypervisor.

2) Using nova-plugin for heat - Using this approach docker containers run on a
Nova VM. However I do not see any illustration which suggests that you can
autoscale using this approach.

3) Using magnum - Currently only possible by manually invoking it.

4) heat compose - Sample available at the above link just talks about deploying
it up but nothing about auto scaling

5) Swarm Support - Still in dev

While I understand some of these options may enable us during the future
release to autoscale docker on openstack. But looking currently I feel option
#1 is most mature(probably) and by plugging in a ceilometer inspector for
docker hypervisor it may be possible. Another approach could be to using
cfn-push-stats to probably push some stats from docker container.

Please advice through your valued suggestions that time being what is the best
way to achieve auto scaling for docker on openstack. I am ready to contribute
to it in the best possible way.

Regards
Ashish

The information contained in this electronic message and any attachments to
this message are intended for the exclusive use of the addressee(s) and may
contain proprietary, confidential or privileged information. If you are not the
intended recipient, you should not disseminate, distribute or copy this e-mail.
Please notify the sender immediately and destroy all copies of this message and
any attachments. WARNING: Computer viruses can be transmitted via email. The
recipient should check this email and any attachments for the presence of
viruses. The company accepts no liability for any damage caused by any virus
transmitted by this email. www.wipro.com

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] [nova-docker][ceilometer][heat] Autoscaling docker in openstack

Hi Segey,


So IIUC approach #2 may still help to autoscale docker on openstack. I will try 
that out and post questions on heat irc thanks.


Regards

Ashish


From: Sergey Kraynev skray...@mirantis.com
Sent: Friday, April 17, 2015 7:01 PM
To: OpenStack Development Mailing List (not for usage questions)
Subject: Re: [openstack-dev] [nova-docker][ceilometer][heat] Autoscaling docker 
in openstack

Hi, Ashish.

Honestly I am not familiar with most part of these ways, but can add more 
information from Heat side (item 2).

I am surprised, that you have missed Heat autoscaling mechanism (You should 
look it :) ). It's one of the important part of Heat project.
It allows to scale vms/stacks by using Ceilometer alarms. There are couple 
examples of autoscale templates:

https://github.com/openstack/heat-templates/blob/master/hot/autoscaling.yaml
 (with LoadBalancer)
https://github.com/openstack/heat-templates/blob/master/hot/asg_of_servers.yaml
https://github.com/openstack/heat-templates/blob/master/hot/asg_of_stacks.yaml
https://github.com/openstack/heat-templates/blob/master/hot/asg_of_servers.yaml

https://github.com/openstack/heat-templates/blob/master/hot/asg_of_servers.yaml
It's true, that Docker plugin for Heat create docker server on Nova::Server 
resource. So you may write template Docker resource + Server resource (similar 
on third template) and scale by using Ceilometer alarms.
If you have any questions how to use it, please got to #heat irc channel and 
ask us :)
Also another way (AFAIK) is to use SoftwareDeployment/Config and deploy Server 
with docker inside (without docker plugin). In this way, I suppose, Steve Baker 
can help with advise :)


On 17 April 2015 at 16:06, 
ashish.jai...@wipro.commailto:ashish.jai...@wipro.com wrote:

Hi,

I have been working on running docker on openstack. I had a discussion on 
multiple IRC and IIUC there are 5 different ways of running docker on 
openstack. IIUC currently there is no way to autoscale docker on openstack. 
Please correct me if I am wrong


1) Using nova-docker driver - Running docker as a Nova::Server using 
nova-docker hypervisor
2) Using nova-plugin for heat - Running docker using 
DockerInc::Docker::Container
3) Using magnum - IIUC no automation as of now, manually it is possible. Not 
enough documentation available
4) heat compose - Saw some samples available 
@https://github.com/openstack/heat-templates/tree/master/hot/software-config/elements/heat-config-docker-compose
5) Swarm support - Still in development

Issues with each on the above approaches

1) Using nova-docker driver - IIUC there is no way for ceilometer to collect 
and emit statistics for docker hypervisor. So that mean ceilometer does not 
have any stats available once you switch to docker driver.
This link 
(https://github.com/openstack/ceilometer/tree/master/ceilometer/compute/virt) 
currently does not have anything for docker hypervisor.

2) Using nova-plugin for heat - Using this approach docker containers run on a 
Nova VM. However I do not see any illustration which suggests that you can 
autoscale using this approach.

3) Using magnum - Currently only possible by manually invoking it.

4) heat compose - Sample available at the above link just talks about deploying 
it up but nothing about auto scaling

5) Swarm Support - Still in dev

While I understand some of these options may enable us during the future 
release to autoscale docker on openstack. But looking currently I feel option 
#1 is most mature(probably) and by plugging in a ceilometer inspector for 
docker hypervisor it may be possible. Another approach could be to using 
cfn-push-stats to probably push some stats from docker container.

Please advice through your valued suggestions that time being what is the best 
way to achieve auto scaling for docker on openstack. I am ready to contribute 
to it in the best possible way.

Regards
Ashish






The information contained in this electronic message and any attachments to 
this message are intended for the exclusive use of the addressee(s) and may 
contain proprietary, confidential or privileged information. If you are not the 
intended recipient, you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately and destroy all copies of this message and 
any attachments. WARNING: Computer viruses can be transmitted via email. The 
recipient should check this email and any attachments for the presence of 
viruses. The company accepts no liability for any damage caused by any virus 
transmitted by this email. www.wipro.comhttp://www.wipro.com

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: 
openstack-dev-requ...@lists.openstack.org?subject:unsubscribehttp://openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

The

Re: [openstack-dev] [nova-docker][ceilometer][heat] Autoscaling docker in openstack

Yes limitation #1 indeed exists, have got confirmation from few of the 
developers. Here is one blueprint which talks about this


https://blueprints.launchpad.net/ceilometer/+spec/container-monitoring



From: VACHNIS, AVI (AVI) avi.vach...@alcatel-lucent.com
Sent: Friday, April 17, 2015 7:41 PM
To: Ashish Jain (WT01 - BAS); openstack-dev@lists.openstack.org
Subject: Re: [openstack-dev] [nova-docker][ceilometer][heat] Autoscaling docker 
in openstack


Hi,

@Ashish, if the limitation you've mentioned for #1 still exists, I join your 
question how heat auto-scale-group may work w/o ceilometer being able to 
collect docker metrics?



@Sergey, hey. Are you saying that ceilometer do collects metrics on docker 
underlying nova::server resource?



-Avi



-- Original message--

From: ashish.jai...@wipro.com

Date: Fri, Apr 17, 2015 4:56 PM

To: openstack-dev@lists.openstack.org;

Subject:Re: [openstack-dev] [nova-docker][ceilometer][heat] Autoscaling docker 
in openstack



Hi Segey,


So IIUC approach #2 may still help to autoscale docker on openstack. I will try 
that out and post questions on heat irc thanks.


Regards

Ashish


From: Sergey Kraynev skray...@mirantis.com
Sent: Friday, April 17, 2015 7:01 PM
To: OpenStack Development Mailing List (not for usage questions)
Subject: Re: [openstack-dev] [nova-docker][ceilometer][heat] Autoscaling docker 
in openstack

Hi, Ashish.

Honestly I am not familiar with most part of these ways, but can add more 
information from Heat side (item 2).

I am surprised, that you have missed Heat autoscaling mechanism (You should 
look it :) ). It's one of the important part of Heat project.
It allows to scale vms/stacks by using Ceilometer alarms. There are couple 
examples of autoscale templates:

https://github.com/openstack/heat-templates/blob/master/hot/autoscaling.yaml
 (with LoadBalancer)
https://github.com/openstack/heat-templates/blob/master/hot/asg_of_servers.yaml
https://github.com/openstack/heat-templates/blob/master/hot/asg_of_stacks.yaml
https://github.com/openstack/heat-templates/blob/master/hot/asg_of_servers.yaml

https://github.com/openstack/heat-templates/blob/master/hot/asg_of_servers.yaml
It's true, that Docker plugin for Heat create docker server on Nova::Server 
resource. So you may write template Docker resource + Server resource (similar 
on third template) and scale by using Ceilometer alarms.
If you have any questions how to use it, please got to #heat irc channel and 
ask us :)
Also another way (AFAIK) is to use SoftwareDeployment/Config and deploy Server 
with docker inside (without docker plugin). In this way, I suppose, Steve Baker 
can help with advise :)


On 17 April 2015 at 16:06, 
ashish.jai...@wipro.commailto:ashish.jai...@wipro.com wrote:

Hi,

I have been working on running docker on openstack. I had a discussion on 
multiple IRC and IIUC there are 5 different ways of running docker on 
openstack. IIUC currently there is no way to autoscale docker on openstack. 
Please correct me if I am wrong


1) Using nova-docker driver - Running docker as a Nova::Server using 
nova-docker hypervisor
2) Using nova-plugin for heat - Running docker using 
DockerInc::Docker::Container
3) Using magnum - IIUC no automation as of now, manually it is possible. Not 
enough documentation available
4) heat compose - Saw some samples available 
@https://github.com/openstack/heat-templates/tree/master/hot/software-config/elements/heat-config-docker-compose
5) Swarm support - Still in development

Issues with each on the above approaches

1) Using nova-docker driver - IIUC there is no way for ceilometer to collect 
and emit statistics for docker hypervisor. So that mean ceilometer does not 
have any stats available once you switch to docker driver.
This link 
(https://github.com/openstack/ceilometer/tree/master/ceilometer/compute/virt) 
currently does not have anything for docker hypervisor.

2) Using nova-plugin for heat - Using this approach docker containers run on a 
Nova VM. However I do not see any illustration which suggests that you can 
autoscale using this approach.

3) Using magnum - Currently only possible by manually invoking it.

4) heat compose - Sample available at the above link just talks about deploying 
it up but nothing about auto scaling

5) Swarm Support - Still in dev

While I understand some of these options may enable us during the future 
release to autoscale docker on openstack. But looking currently I feel option 
#1 is most mature(probably) and by plugging in a ceilometer inspector for 
docker hypervisor it may be possible. Another approach could be to using 
cfn-push-stats to probably push some stats from docker container.

Please advice through your valued suggestions that time being what is the best 
way to achieve auto scaling for docker on openstack. I am ready to contribute 
to it in the best possible way.

Regards
Ashish






The information

Re: [openstack-dev] [nova-docker][ceilometer][heat] Autoscaling docker in openstack