You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Source Code Releases
All the releases are signed by the public PGP key A0E98066 of Yichun Zhang.
That's 32 bits to identify a key — and that's a bit few bits; others can generate fake keys with the same last 32 bits.
In fact, someone did:
$ gpg --keyserver keyserver.ubuntu.com --keyserver-options timeout=10 --recv-key A0E98066
gpg: key A84A5A40A0E98066: public key "Totally Legit Signing Key <[email protected]>" imported
gpg: Total number processed: 1
gpg: imported: 1
As you can see, someone has generated a fake key that ends with A0E98066.
(That's Ubuntu's keyserver, which I tried, when MIT didn't work; they're supposed to be in sync with each other I think?)
What do you think about specifying the full key ID on the website, or the last 64 bits?
There's an old issue and merged PR related to that,
issue: #30 "Add information to the GPG key"
PR: #32 "Add information about public key to verify release files"
but somehow the changes in the PR seems to have gotten lost during the years. Anyway, in the PR, I see that the last 64 bits of the key are: 0xB550E09EA0E98066
The text was updated successfully, but these errors were encountered:
Maybe might as well include the full key? Don't know if 64 bit is that much nowadays?
I think this is the complete public key? 25451EB088460026195BD62CB550E09EA0E98066
gpg --keyserver pgpkeys.mit.edu --keyserver-options timeout=10 --recv-key 25451EB088460026195BD62CB550E09EA0E98066
gpg: key B550E09EA0E98066: 1 signature not checked due to a missing key
gpg: key B550E09EA0E98066: "Yichun Zhang (agentzh) <agentzh at ... googlemail ...>" not changed
gpg: Total number processed: 1
gpg: unchanged: 1
On the website, https://openresty.org/en/download.html, there's this:
That's 32 bits to identify a key — and that's a bit few bits; others can generate fake keys with the same last 32 bits.
In fact, someone did:
As you can see, someone has generated a fake key that ends with A0E98066.
(That's Ubuntu's keyserver, which I tried, when MIT didn't work; they're supposed to be in sync with each other I think?)
What do you think about specifying the full key ID on the website, or the last 64 bits?
There's an old issue and merged PR related to that,
issue: #30 "Add information to the GPG key"
PR: #32 "Add information about public key to verify release files"
but somehow the changes in the PR seems to have gotten lost during the years. Anyway, in the PR, I see that the last 64 bits of the key are: 0xB550E09EA0E98066
The text was updated successfully, but these errors were encountered: