On 26 Sep 2011, at 00:02, Andrew Deason <[email protected]> wrote: > > You can also go the route that Derrick suggested where we use the > existing rmtsys mechanism (which is usually used to forward pioctls to > another machine).
Strategically, I think it would be nice not to add a dependency on the rmtsys code. At the moment, the only user of that functionality is the NFS translator, which is becoming unsupportable on today's operating systems. Having support for rmtsys on the client side adds a lot of complexity to our code. It would be nice to consider a world in which we can build OpenAFS without it. > > Also, to get authentication working, just get the pioctls working. aklog > just sets credentials via a pioctl, so once it can work with them, it > shouldn't need anything else special. I think libuafs (or another > library) really needs an API that allows setting credentials more easily > (so you don't need to issue the pioctl manually in whatever > libuafs-using application you're using), but for the purposes of the > fuse client it's not necessary. One thing that I am currently working on is turning aklog and friends into a library. With this, and the new token code, it should be possible for the libuafs API to simply expose a "set a token" function. > S._______________________________________________ OpenAFS-devel mailing list [email protected] https://lists.openafs.org/mailman/listinfo/openafs-devel
