When I copy some older private_key files I generated with Shallot time ago to a new tor process, then give it a kill -1, it parses the private_key and produces a different hostname to match it. It's odd because the private key is the one that should have the nice name, instead it now gets an ugly new one. Any guess on what I could have done wrong? Apparently nobody else ran into this problem...
Trac: Username: vynX
To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information
Child items
0
Show closed items
No child items are currently assigned. Use child items to break down this issue into smaller parts.
Linked items
0
Link issues together to show that they're related.
Learn more.
Stopping tor, copying the onion directory in place (the two files), then restarting tor... same effect. What could cause the same key file to produce a different hash? Or is it impossible and thus the mistake somewhere in my file handling...?
What do you mean by all the files? I don't have the state of last year's tor. In fact this is a shallot key I hadn't previously used in deployment. I just generated it a year earlier. Can shallot produce bogus key pairs? I tried two and they both receive this treatment.
I got the same hostname in def/hostname as I saw in abc/hostname.
So I wasn't able to reproduce this bug. What we're going to need to figure this out is step-by-step instructions to try to reproduce the problem you're seeing.
Err. Unless I'm misreading this, the old private key never was actually used to tor prior to producing the wrong result. The bug seems to be that the host name given by tor for a given private key doesn't match the expected one from Shallot.
I don't think this is reproducible without the modulus and public exponent (the public key) in question, and it's not immediately clear to me how that sort of thing fails....