Skip to contentSkip to conversation

Question on nfs-utils patch for 'keyctl_invalidate'.

DISCUSSION Ben Greear 3 participants 4 messages Apr 15, 2014 – Apr 16, 2014

Subsystems (2)
NFS client NFS server
lore
Share & tools
lore

Ben Greear

author
#0 lore raw

It seems that key_invalidate is only available in quite recent keyutils packages.

Anyone know if there is a relatively simple way to implement the same behaviour using older keyutils packages?

The reason I ask is that I recently updated to latest nfs-utils package and how build fails on many of my older systems...

Thanks,
Ben

Signature · 2 lines
Ben Greear <[email protected]>
Candela Technologies Inc  http://www.candelatech.com
#1 lore raw
Quoted 5 lines · Ben Greear

It seems that key_invalidate is only available in quite recent keyutils packages.

Anyone know if there is a relatively simple way to implement the same behaviour using older keyutils packages?

CC-ing the keyutils maintainer... David, when did that kernel change go in that cause keyctl_revoke() to no longer work?

The nfs-utils commit is:

commit 2ae0763a618d30037ebb2520f6292f80d838a440

Author: Steve Dickson <[email protected]>
Date:   Tue Mar 25 10:56:58 2014 -0400

nfsidmap: Keys need to be invalidated instead of revoked

It sound like we need to do some type of #ifdef kernel version....

steved.

Quoted 7 lines

The reason I ask is that I recently updated to latest nfs-utils package and how build fails on many of my older systems...

Thanks,
Ben

author
#2 lore raw
Quoted 13 lines · Steve Dickson
Quoted 5 lines · Ben Greear

It seems that key_invalidate is only available in quite recent keyutils packages.

Anyone know if there is a relatively simple way to implement the same behaviour using older keyutils packages?

CC-ing the keyutils maintainer... David, when did that kernel change go in that cause keyctl_revoke() to no longer work?

The nfs-utils commit is:

Quoted 7 lines

commit 2ae0763a618d30037ebb2520f6292f80d838a440

Author: Steve Dickson <[email protected]>
Date:   Tue Mar 25 10:56:58 2014 -0400

nfsidmap: Keys need to be invalidated instead of revoked

It sound like we need to do some type of #ifdef kernel version....

We might need an #ifdef based on keyutils version (which is what I implemented to at least get it compiling), but if you need to key on kernel version then you need run-time probing of the kernel as build system may run different kernel than installed system.

Thanks,
Ben

Quoted 10 lines

steved.

Quoted 7 lines

The reason I ask is that I recently updated to latest nfs-utils package and how build fails on many of my older systems...

Thanks,
Ben

Signature · 2 lines
Ben Greear <[email protected]>
Candela Technologies Inc  http://www.candelatech.com
#3 lore raw
In reply to Steve Dickson
Quoted 2 lines · Steve Dickson

CC-ing the keyutils maintainer... David, when did that kernel change go in that cause keyctl_revoke() to no longer work?

I think this is what you are referring to:

Committer: David Howells <[email protected]> 2012-10-02 19:24:56
Parent: 3a50597de8635cd05133bd12c95681c82fe7b878 (KEYS: Make the session and process keyrings per-thread)
Child: f8aa23a55f813c9bddec2a6176e0e67274e6e7c1 (KEYS: Use keyring_alloc() to create special keyrings)
Branches: cred-fixes, master and many more (159)
Follows: v3.6-rc7
Precedes: fscache-fixes-20121220

Quoted 1 line

It sound like we need to do some type of #ifdef kernel version....

You could try keyctl_invalidate(), then if that doesn't work, call keyctl_revoke().

David