Jump to: navigation, search

Difference between revisions of "Fenix"

(Undo revision 172651 by Tomi Juvonen (talk))
(Train priorities)
(10 intermediate revisions by the same user not shown)
Line 1: Line 1:
 
[[File:Fenix-birdi 192x192.png|thumb|left]]
 
[[File:Fenix-birdi 192x192.png|thumb|left]]
 
==What is Fenix?==
 
==What is Fenix?==
Fenix implements rolling infrastructure maintenance, upgrade and scaling. It can do this also
+
Optimized rolling infrastructure pluggable one-click maintenance, upgrade, scaling and life-cycle
in interaction with the application on top of it if the application supports it. In Telco world we talk
+
workflow with VNFM interaction guaranteeing zero impact on VNF service.  
about VNFM, but one can implement own simple manager for any application.
 
  
Infrastructure admin can call Fenix API to start a maintenance workflow session. This
+
Infrastructure admin can call Fenix API to start a pluggable workflow session. This
 
session will make needed maintenance, upgrade and scaling operations to infrastructure
 
session will make needed maintenance, upgrade and scaling operations to infrastructure
optionally in interaction with the application manager to guarantee zero downtime for its service.
+
optionally in interaction with the application manager (VNFM) to guarantee zero downtime for its
Interaction gives the ability for application manager to know about new capabilities coming over
+
service. Interaction gives the ability for the application manager to know about new capabilities
maintenance to make his own upgrade. The application can have a time window to finish what
+
coming over maintenance to make his own upgrade. The application can have a time window to
he is doing, make own action to re-instantiate his instance or have Fenix to make the migration.
+
finish what he is doing, make own action to re-instantiate his instance or have Fenix to make the
Also, seamless application scaling or retirement will be possible.
+
migration. Also, seamless application scaling or retirement will be possible.
  
As Fenix will have project-specific messaging with information about instances affected towards
+
As Fenix has above described project-specific messaging with information about instances
application manager, it will also have admin level messaging. This messaging can tell what host
+
affected towards application manager, it also has admin-level messaging. This messaging can
is down for maintenance, so any infrastructure components can have this information. Special
+
tell what host is down for maintenance, so any infrastructure components can have this information.
case for this would also be telling about adding or removing a host.
+
Special case for this messaging would also be telling about adding or removing a host.
  
The architecture is pluggable to manage different use cases, clouds and payloads. There will
+
The architecture is pluggable to manage different use cases, clouds and payloads. There is
be a plugin for workflow used and for the pre-, host- and post-actions.
+
plugin for workflow used and for the pre-, host- and post-actions.
 +
 
 +
The latest addition is also the ETSI FEAT03 defined constraints for instances and instance groups
 +
that can be used to optimize any rolling workflow to be done in the shortest possible time window,
 +
while at the same time keeping zero impact on VNF service.
  
 
Presentation in OIS Denver 2019:<br />
 
Presentation in OIS Denver 2019:<br />
Line 26: Line 29:
 
[https://drive.google.com/open?id=1k6lcKF6n132qN-LMfxfcCxYk52nscfb8 Resiliency Closed Loop - Telco Self-healing and Maintenance]<br />
 
[https://drive.google.com/open?id=1k6lcKF6n132qN-LMfxfcCxYk52nscfb8 Resiliency Closed Loop - Telco Self-healing and Maintenance]<br />
 
Live demo in ONS Europe 2019:<br />
 
Live demo in ONS Europe 2019:<br />
[https://drive.google.com/open?id=1gV36EkSNehdeaprCZojRqbQP7ntb6r1j Resiliency Closed Loop - Telco Self-healing and Maintenance]<br />
+
[https://drive.google.com/file/d/1BkYciG7-ItpoZjMan1mPDsX7kWjGvBSl/view?usp=sharing Resiliency Closed Loop - Telco Self-healing and Maintenance]<br />
 
Unconference in ONS Europe 2019:<br />
 
Unconference in ONS Europe 2019:<br />
 
[https://etherpad.opnfv.org/p/ONS-Ant2019-Fenix Host maintenance with Fenix]
 
[https://etherpad.opnfv.org/p/ONS-Ant2019-Fenix Host maintenance with Fenix]
Line 44: Line 47:
 
* 2003846 Support for action plug-in
 
* 2003846 Support for action plug-in
 
* 2003830 Fenix workflow error handling and use of database
 
* 2003830 Fenix workflow error handling and use of database
<br><br/>
 
With these, Fenix is already usable for trialing. Anyhow it will be in Train priorities to have it more mature.
 
  
 
===Train priorities===
 
===Train priorities===
Line 51: Line 52:
 
<br><br/>
 
<br><br/>
 
As there were two non-Telco users wanting to start using Fenix, it was the most important thing driving the priorities:
 
As there were two non-Telco users wanting to start using Fenix, it was the most important thing driving the priorities:
<br><br/>
+
<br/>
 
https://etherpad.openstack.org/p/DEN2019-fenix-PTG
 
https://etherpad.openstack.org/p/DEN2019-fenix-PTG
 
<br><br/>
 
<br><br/>
 
Also, there was a discussion about ETSI NFV harmonization:
 
Also, there was a discussion about ETSI NFV harmonization:
<br><br/>
+
<br />
 
https://etherpad.openstack.org/p/DEN2019-fenix-ETSI-NFV-PTG
 
https://etherpad.openstack.org/p/DEN2019-fenix-ETSI-NFV-PTG
<br><br/>
 
This will still need more discussion and thinking. It is a very important topic, but not the first priority
 
  
 
===Ussuri===
 
===Ussuri===
 
Open Infrastructure Summit - Shanghai
 
Open Infrastructure Summit - Shanghai
<br />
 
Forum: https://etherpad.openstack.org/p/Shanghai2019-Fenix-forum
 
 
<br />
 
<br />
 
PTG:https://etherpad.openstack.org/p/Shanghai2019-Fenix-PTG
 
PTG:https://etherpad.openstack.org/p/Shanghai2019-Fenix-PTG
Line 70: Line 67:
 
<br />
 
<br />
 
Specification to start implementing ETSI FEAT03: NFVI software modification.
 
Specification to start implementing ETSI FEAT03: NFVI software modification.
https://review.opendev.org/#/c/682788
+
https://fenix.readthedocs.io/en/latest/specifications/ussuri-etsi-feat03.html
 +
<br />
 +
* Security
 +
* Update/continue failed session
 +
* Parallel execution
 +
* Kubernetes
  
 
==Documentation==
 
==Documentation==
Line 78: Line 80:
 
===Meetings===
 
===Meetings===
 
bi-weekly meeting on Monday 5 AM UTC ("odd" weeks on #openstack-Fenix)
 
bi-weekly meeting on Monday 5 AM UTC ("odd" weeks on #openstack-Fenix)
 
+
<br />
<br /><br />
 
 
Topics for next meeting
 
Topics for next meeting
 
Topics (feel free to add one):
 
Topics (feel free to add one):

Revision as of 07:05, 17 February 2020

Fenix-birdi 192x192.png

What is Fenix?

Optimized rolling infrastructure pluggable one-click maintenance, upgrade, scaling and life-cycle workflow with VNFM interaction guaranteeing zero impact on VNF service.

Infrastructure admin can call Fenix API to start a pluggable workflow session. This session will make needed maintenance, upgrade and scaling operations to infrastructure optionally in interaction with the application manager (VNFM) to guarantee zero downtime for its service. Interaction gives the ability for the application manager to know about new capabilities coming over maintenance to make his own upgrade. The application can have a time window to finish what he is doing, make own action to re-instantiate his instance or have Fenix to make the migration. Also, seamless application scaling or retirement will be possible.

As Fenix has above described project-specific messaging with information about instances affected towards application manager, it also has admin-level messaging. This messaging can tell what host is down for maintenance, so any infrastructure components can have this information. Special case for this messaging would also be telling about adding or removing a host.

The architecture is pluggable to manage different use cases, clouds and payloads. There is plugin for workflow used and for the pre-, host- and post-actions.

The latest addition is also the ETSI FEAT03 defined constraints for instances and instance groups that can be used to optimize any rolling workflow to be done in the shortest possible time window, while at the same time keeping zero impact on VNF service.

Presentation in OIS Denver 2019:
Fenix Denver - forum presentation
Presentation in ONS Europe 2019:
Resiliency Closed Loop - Telco Self-healing and Maintenance
Live demo in ONS Europe 2019:
Resiliency Closed Loop - Telco Self-healing and Maintenance
Unconference in ONS Europe 2019:
Host maintenance with Fenix

Development

Stein release notes

During Stein release, Fenix most important new features were the stories below:

  • 2004147 Ability to download plug-ins and SW changes
  • 2004146 Support for one-click upgrade
  • 2003846 Support for action plug-in
  • 2003830 Fenix workflow error handling and use of database

Train priorities

There was a session in Denver PTG to define the priorities.

As there were two non-Telco users wanting to start using Fenix, it was the most important thing driving the priorities:
https://etherpad.openstack.org/p/DEN2019-fenix-PTG

Also, there was a discussion about ETSI NFV harmonization:
https://etherpad.openstack.org/p/DEN2019-fenix-ETSI-NFV-PTG

Ussuri

Open Infrastructure Summit - Shanghai
PTG:https://etherpad.openstack.org/p/Shanghai2019-Fenix-PTG

The main priority:
Specification to start implementing ETSI FEAT03: NFVI software modification. https://fenix.readthedocs.io/en/latest/specifications/ussuri-etsi-feat03.html

  • Security
  • Update/continue failed session
  • Parallel execution
  • Kubernetes

Documentation

Communication and Meetings

Meetings

bi-weekly meeting on Monday 5 AM UTC ("odd" weeks on #openstack-Fenix)
Topics for next meeting Topics (feel free to add one):

  • Status
  • AoB

Contact Us

  • IRC channel for regular daily discussions: #openstack-fenix
  • Use [fenix] tag for Fenix emails on OpenStack Mailing Lists