-
Notifications
You must be signed in to change notification settings - Fork 23
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
RFC: chaos engineering as a service #14
base: main
Are you sure you want to change the base?
Conversation
two seperate programs that serve similar purpose. | ||
**Goal is to unify the two.** | ||
2. Poor observability of experiment results from within the dashboard | ||
**Goal is to collect the metrics by Prometheus and show in dashboard.** |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
maybe collect the metrics by Prometheus and show in dashboard
is just one way of improve the observability. 😃
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@WangXiangUSTC What do you suggest we could do besides Prometheus integration for this?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
maybe some data of the chaos experiment itself. In fact, I don’t have a definite idea yet
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Good examples of metrics from Litmus as mentioned in this blog.
I think we must have metrics around pass/fail/awaited for all chaos operations which client is going to perform. We already have one pass metric but we certainly can have more metric around this (like per namespace) to make data more visible and easy to consume.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yeah, I agree with you
Signed-off-by: Shivansh Saini <[email protected]>
Signed-off-by: Shivansh Saini <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
In addition, do we need to describe how to control some role's privileges in this RFC?
Chaosd runs on physic nodes outside kubernetes cluster, so it is vulnerable to attack | ||
from internet. To prevent misuse of chaosd, it needs to allow only authenticated | ||
requests. The easiest and secure setup is to use SSL certificates to both encrypt | ||
the request data and for authentication. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
should unit them into one line
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
have some problems below
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
umm.. basically combine to one sentence??
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
to one line
In this setup, private key of the certificate will be generated and kept with the | ||
dashboard and public key would be stored on chaosd nodes. On any request, | ||
chaosd would first verify the digital signatures presented by the client to | ||
authenticate the request. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I have a question, the private key is kept by client(dashboard),is it looks strange?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
For authentication using certificates, the requesting entity must have the private key. Since dashboard will be calling http endpoints of chaosd, it'll have private key
@WangXiangUSTC that's a discussion in itself since it corresponds to what actions can be taken on the dashboard. I don't have the full clarity on everything so we could probably discuss it in a team meeting? |
OK |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
Related issue: chaos-mesh/chaos-mesh#1462