How to avoid Denial-of-service attacks when using RGW facing public internet?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi,
I'm new to Ceph and trying to understand how things work.

Hypothetical situation: An organisation would like to provide a storage
solution (S3 API) to its members. The members need to access the
S3 API from the public internet. (There is no VPN).

Are there any recommendations regarding having RGW facing the public internet?
I'm thinking about the risk of DoS attacks.
(see https://en.wikipedia.org/wiki/Denial-of-service_attack)

I would assume that anonymous access to the RGW API needs to be disallowed.
All access should be either authenticated or in the form of a pre-signed URL.

I guess that object ACLs that give anonymous read access could be allowed but
to be able to actually access such an object, a time-limited presigned
URL needs to be used (that was created by an authenticated user). A
download could therefore happen only after an active decision was
made.

The members of the organisation want to set up their own websites.
Such a website could generate time-limited presigned URLs that are
sent to the web browser. The visitors of the website could therefore
download/upload directly from/to the RGW.

My main point with the discussion above is that allowing direct
anonymous read access to objects (without presigned URLs) will make it
difficult to dynamically adjust the amount of legitimate read
requests. When using presigned URLs there will be an authenticated
user that signed it. The downloads originating from that presigned URL
could therefore consume the quota of that authenticated user. If an
authenticated user would run out of their download quota, no more read
traffic would originate from that user.

I'm doing a lot of guessing here. If you know more about this, I would
be glad if you could share some advice. Do you know of any reading
material about it (such as blog posts)?

Thank you,
Erik Sjölund
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx




[Index of Archives]     [Information on CEPH]     [Linux Filesystem Development]     [Ceph Development]     [Ceph Large]     [Ceph Dev]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]


  Powered by Linux