Nexus 9000 Series Boot mode LXC vs normal

[ggnogame1]

Hello, I would like to check the functional/performance difference between LXC mode (booting mode for using enhanced ISSU) and normal mode in Nexus 9000 Series.

For normal mode, NXOS works directly with hardware, but for LXC mode, I understood that Supervisor/Line card works as Linux Container respectively.
However, is there a functional/performance difference between these booting modes?

ex) For LXC mode, there are restrictions on certain protocol and service-related traffic, etc..
     Or if it's LXC mode, you can't use certain features, etc..

I wonder if there is a difference in function/performance as above.

Thank you.

[shambhu.kumar]

There are three methods to upgrade N9K

Disruptive NX-OS software upgrade
- switch will undergo warm reload
-management plane (SNMP, SSH will disrupt),
-Control plan (EIGRP, OSPF , BGP etc.) disrupt
-Data plan (Layer2 and layer 3 forwarding ) disrupt. All interfaces will go down.

install all nxos bootflash:nxos.x.x.x.bin

===================================================
Standard Non-Disruptive (ISSU) NX-OS software Upgrade
-Require specific upgrade path, from one version to another
-Switch will not reload
-management plan (SNMP, SSH will disrupt),
-Control plan (EIGRP, OSPF, BGP etc.) not disrupt
Data plan (Layer2 and layer 3 forwarding) not disrupt. All interfaces will not go down

install all nxos bootflash:nxos.x.x.x.bin non-disruptive
(use if switchover upgradation path is very sure).

=============================================================
Enhance Non-Disruptive (ISSU) NX-OS Software Upgrade
Require LXC boot mode (boot mode lxc)
Runs NX-OS as an LXC container
Take effect after a switch reload once configured.

switch(config)# boot mode lxc
show boot mode :
switch(config)# reload
This command will reboot the system. (y/n)? [n] Y
loader>
Require specific upgrade path, from one version to another
Switch will not reload
management plan (SNMP, SSH will disrupt),
Control plan(EIGRP, OSPF , BGP etc) not disrupt
Data plan(Layer2 and layer 3 forwarding ) not disrupt. All interface will not go down.

However, configuring enhanced or container-based ISSU on single supervisor the software runs inside a separate Linux container (LXC) for the supervisors and the line cards. A third container is created as part of the ISSU procedure, and it is brought up as a standby supervisor.
The virtual instances (or the Linux containers) communicate with each other using an emulated Ethernet connection. In the normal state, only two Linux containers are instantiated: vSup1 (a virtual SUP container in an active role) and vLC (a virtual linecard container).

install all nxos bootflash:nxos.x.x.x.bin non-disruptive

Regards

[ggnogame1]

Hello Shambhu.kumar, Thank you for your reply.

However, my question was about the functional difference between when NXOS operates on LXC Container and when it operates in normal mode.
I've heard from somewhere that operating on the LXC Container reduces the overall capacity of the equipment.
I don't think there are any documents related to it..
When NXOS is operating on the LXC Container, are there any issues such as service degradation related to specific features or protocols?

Thank you.