Toolkit for troubleshooting BottlerocketOS nodes

[ThiyaguE]

What I'd like:

We lack a ready-to-use troubleshooting toolkit for BottlerocketOS nodes. standard tools like 'iotop' and other utilities are unavailable due to the lightweight nature of the OS, requiring custom images with baked-in tools. We request AWS to build a comprehensive toolkit with essential tools, leveraging their expertise and experience with other Bottlerocket users/customers across the industry.

Any alternatives you've considered:

I've to create a custom container with necessary tools installed.

[vigh-m]

Hi, Bottlerocket host containers should have the right privileges to support this level of debugging. I tested this out in the included admin-container with top:

top - 21:21:13 up  3:02,  0 users,  load average: 0.00, 0.01, 0.00
Tasks: 491 total,   1 running, 250 sleeping,   0 stopped,   0 zombie
%Cpu(s):  0.0 us,  0.0 sy,  0.0 ni, 99.9 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
KiB Mem :  1.8/97092152 [||                                                                                                  ]
KiB Swap:  0.0/0        [                                                                                                    ]

which is exactly what I see on the instance itself:

bash-5.1# cat /proc/meminfo
MemTotal:       97092152 kB
MemFree:        91944020 kB
MemAvailable:   95338316 kB
Buffers:            6236 kB
Cached:          4085584 kB
SwapCached:            0 kB

For my case, I had to set the TERM variable

# echo $TERM
xterm-256color

You should be able to use the admin-container or bring your own host container and enable superpowered mode for it. However, you should be aware that this is a highly privileged container and should only be used for administrative tasks. I would not recommend leaving such a container enabled in a production environment.