> On Jun 28, 2017, at 8:11 AM, Nico Williams <n...@cryptonector.com> wrote: > > On Wed, Jun 28, 2017 at 07:28:59AM +0200, Lars-Johan Liman wrote: >> All (pun intended!), >> >> On Mon, Jun 26, 2017 at 11:18:28AM +0200, Andreas Haupt wrote: >>>> Heimdal 7.3 seems to suffer from a bug in privilege checking. A prinicipal >>>> having all rights on the database is unable to extract keytabs: >> >> n...@cryptonector.com: >>> This is on purpose. >> >>> We decided that it was never a good idea for "all" to have meant >>> "extract keys", because in general that's not desirable. >> >> I very seldom raise my voice on this mailing list, but here I must, on >> sheer principal grounds. > > I hope you feel welcomed. Please speak up more often! > >> Chosen names must have obvious meanings. To have a status called "all" >> which isn't *ALL* is confusing at best. It will confuse the h-ll out of >> sysadmins over the globe for years to come, wasting time and money for >> no good purpose at all. I would have spent hours upon hours not >> understanding what the problem was, had I run into this trap. >> >> The "keep it simple" principle and the principle of least surprise are >> two fundamental principles for successful system management. >> >> Please fix this, either by changing the name "all" to "most" (or >> preferrably to somthing better), or by changing the behaviour to be >> *ALL*. Either is fine, but having "all" not mean *ALL* is not a good way >> forward.
I absolutely agree with everything you said. I think it very to make unwise to make a backwards-incompatible change in order to turn the permissions description from plain english into a Heimdal-unique code. I’ll repeat my recommendation that we do what Love suggested. > Renaming "all" to "most" would have been a backwards-incompatible change > too. We chose a different backwards-incompatible change. > > Changing the meaning of "all" was a backwards-incompatible change that > we WANTED to make because the previous situation was... not good! By > making this change we're confronting sites with the underlying problem > that allowing extraction keys from the HDB is NOT a good thing, and > we're letting them choose how to move past this (they have two options). > > Since there is a trivial way to get "all" + "get-keys", this change, > though backwards-incompatible, is of rather limited impact. Disagree. I think you are looking at the wrong impact when you make that judgement. Making “all” mean something other than “all” creates a *permanent* source of confusion. > It is true > that switching to "ext_keytab -r", which is what we want sites to do, is > more difficult and requires careful consideration by them, but again, > you can get the old "all" by granting your admins "all" + "get-keys", so > you're not forced to use "ext_keytab -r”. How about putting the recommended change (to ...-r) in the permissions error for ext_keytab? > In general, backwards-compatibility is a high priority. But security is > a higher priority. In general, we might remove interfaces and important > behaviors, but we won't break them. If some interface in Heimdal is > insecure, and no backwards-compatible change can make it secure, then > we're just going to make a backwards-incompatible change. > > All in all, we considered this carefully. It's been discussed > extensively now, and we will make no further changes in this area other > than to improve the error message that users get. I'm not being > flippant here, and I'm not ignoring your input. We appreciate that this > change was surprising and caused some pain and we appreciate your input, > and if we reject your proposal in this case, please understand that it's > only after careful consideration. The point is that you have made the pain, or at least the confusion, permanent by making the language mean something other than what it says. That does not seem well considered, and I think the amount of flack you’re getting reflects that. If you’re really going to be that hard-nosed about it, then you had better put a hell of a lot of warning messages and explanations all over the place. Better yet, just do away with any ability to extract (current) keys altogether. Make the -r option the only thing possible unless you build with some configure option like --enable-insecure-key-extraction. Then you can phase it out over a couple of revisions, like we did with single-DES. Honestly, is it really that big a deal to change “all” to e.g. “admin”? If you’re going to make a fundamental semantic change, you shouldn’t hide the fact. You should make it as obvious as you possibly can. > Nico > -- Personal email. hbh...@oxy.edu
