Replies: 5 comments 6 replies
-
It's not just you. At least at WUR it is not the ideal solution, but they also see that webdav is impossible to add mfa to. They have pointed to possibilities of the usage of VPN as well. Having said that, I know that the Yoda devs have been working on this for a long time now (and maybe even looking at other protocol if memory serves me right) also due to security demands at UU. |
Beta Was this translation helpful? Give feedback.
-
We are considering S3 as replacement for webDAV. iRODS is working on a S3 client API. We have considered putting WebDAV behind a VPN so the users have to log in on the VPN first. But this does not work for users outside our institution, but could work when used in combination with SRAM. |
Beta Was this translation helpful? Give feedback.
-
Now that SRAM is upon us, would it be easy to make SRAM 2fa for webdav connections available, or would this still be rather complicated (if complicated, probably better to wait for S3). |
Beta Was this translation helpful? Give feedback.
-
Hello all, |
Beta Was this translation helpful? Give feedback.
-
Is that still up for consideration? Or would by the time work on this was done, S3 already be available? |
Beta Was this translation helpful? Give feedback.
-
Hi all,
I'm hoping you can help us with a thorny issue that came up at the VU:
In discussions with our security officers a major issue was MFA. Unfortunately SRAM will only partly solve this issue, the lack of MFA on all interfaces (meaning WebDAV and iRODS) is also seen as a blocking issue to use Yoda for sensitive data. The use of Data Access Passwords is not seen as a sufficient mitigating measure by our ISOs.
As far as I know work has been done to be able to use an OAuth/OIDC/SAML login on the iRODS CLI (https://github.com/irods-contrib/irods_working_group_authentication), that could potentially provide a solution.
That still leaves WebDAV. Their are of course no easy solutions for this, since the protocol simply doesn't support it. Possible workarounds I can think of:
None of these solutions are simple or elegant.
Any other ideas? Maybe the UU has considered other protocols in the past?
Do the security officers of other consortium members also see Data Access Passwords as an insufficient mitigation for the lack of MFA on WebDAV? If it's just us, it would be more fruitful to convince ours ISOs to accept the measures we take together as a consortium, even if they don't entirely agree.
Beta Was this translation helpful? Give feedback.
All reactions