[GH-ISSUE #50] Upgrade OpenSSL to 0.8.x #32

Closed
opened 2026-03-07 22:18:10 +03:00 by kerem · 9 comments
Owner

Originally created by @chaoticsmol on GitHub (Oct 9, 2016).
Original GitHub issue: https://github.com/hickory-dns/hickory-dns/issues/50

Originally assigned to: @bluejekyll on GitHub.

Before the upgrade to ring, I'd like to request that trust-dns adopt rust-openssl version 0.8. I found in my use of rust-openssl in rust-ftp that some usage of rust-openssl version 0.7 lead to a memory error that was resolved immediately by the update to 0.8.

For a tiny bit more information, see the issue I opened on hyperium/hyper.

Originally created by @chaoticsmol on GitHub (Oct 9, 2016). Original GitHub issue: https://github.com/hickory-dns/hickory-dns/issues/50 Originally assigned to: @bluejekyll on GitHub. Before the upgrade to _ring_, I'd like to request that trust-dns adopt rust-openssl version 0.8. I found in my use of rust-openssl in rust-ftp that some usage of rust-openssl version 0.7 lead to a memory error that was resolved immediately by the update to 0.8. For a tiny bit more information, see the [issue I opened on hyperium/hyper](https://github.com/hyperium/hyper/issues/907).
kerem closed this issue 2026-03-07 22:18:10 +03:00
Author
Owner

@bluejekyll commented on GitHub (Oct 10, 2016):

If you want to submit a PR for that, I'd happily accept.

<!-- gh-comment-id:252762398 --> @bluejekyll commented on GitHub (Oct 10, 2016): If you want to submit a PR for that, I'd happily accept.
Author
Owner

@bluejekyll commented on GitHub (Oct 18, 2016):

I started work on this to hopefully fix Travis OSX builds, but man, openssl 0.8.0 is quite different from 0.7 :(

<!-- gh-comment-id:254405169 --> @bluejekyll commented on GitHub (Oct 18, 2016): I started work on this to hopefully fix Travis OSX builds, but man, openssl 0.8.0 is quite different from 0.7 :(
Author
Owner

@chaoticsmol commented on GitHub (Oct 18, 2016):

It is pretty different. I'd have liked to get to this myself but have been over-loaded with other things.

<!-- gh-comment-id:254508795 --> @chaoticsmol commented on GitHub (Oct 18, 2016): It is pretty different. I'd have liked to get to this myself but have been over-loaded with other things.
Author
Owner

@bluejekyll commented on GitHub (Oct 19, 2016):

2b14581 is the conversion to openssl 0.8.

While that's basically complete, of course my integration tests for the proof of a signed example.com are now failing :(

looking into it.

<!-- gh-comment-id:254721040 --> @bluejekyll commented on GitHub (Oct 19, 2016): 2b14581 is the conversion to openssl 0.8. While that's basically complete, of course my integration tests for the proof of a signed example.com are now failing :( looking into it.
Author
Owner

@bluejekyll commented on GitHub (Oct 19, 2016):

oh, and I have all the warnings related to the futures stuff cleaned up in a different branch. I'll be getting rid of those in a little bit.

<!-- gh-comment-id:254721553 --> @bluejekyll commented on GitHub (Oct 19, 2016): oh, and I have all the warnings related to the futures stuff cleaned up in a different branch. I'll be getting rid of those in a little bit.
Author
Owner

@bluejekyll commented on GitHub (Oct 19, 2016):

Travis is passing again with this patch at least :)

<!-- gh-comment-id:254722510 --> @bluejekyll commented on GitHub (Oct 19, 2016): Travis is passing again with this patch at least :)
Author
Owner

@bluejekyll commented on GitHub (Oct 19, 2016):

FYI, looking more deeply, not all secure queries are failing, and some parts of the chain are validating properly. I'm not entirely able to say right now that there isn't a bug here. rollernet tests are all passing right now, which tells me that it's only example.com that is having issues.

It validates the www.example.com against example.com's DNSKEY, but then fails to validate the DNSKEY for example.com. But since rollertnet tests are functioning this might not be an issue with the upgrade.

<!-- gh-comment-id:254729806 --> @bluejekyll commented on GitHub (Oct 19, 2016): FYI, looking more deeply, not all secure queries are failing, and some parts of the chain are validating properly. I'm not entirely able to say right now that there isn't a bug here. rollernet tests are all passing right now, which tells me that it's only example.com that is having issues. It validates the www.example.com against example.com's DNSKEY, but then fails to validate the DNSKEY for example.com. But since rollertnet tests are functioning this might not be an issue with the upgrade.
Author
Owner

@bluejekyll commented on GitHub (Oct 19, 2016):

cargo test -- --ignored will run all of the integration tests for anyone interested in looking at this.

<!-- gh-comment-id:254730063 --> @bluejekyll commented on GitHub (Oct 19, 2016): `cargo test -- --ignored` will run all of the integration tests for anyone interested in looking at this.
Author
Owner

@bluejekyll commented on GitHub (Oct 20, 2016):

Ok, got to the bottom of it. In changing a bunch of interfaces to return Result (due to some of the changes in OpenSSL), I had made overzealous use of try!() and underthought placement in the validation loops for DNSSec. Everything checks out now.

<!-- gh-comment-id:255010095 --> @bluejekyll commented on GitHub (Oct 20, 2016): Ok, got to the bottom of it. In changing a bunch of interfaces to return Result (due to some of the changes in OpenSSL), I had made overzealous use of try!() and underthought placement in the validation loops for DNSSec. Everything checks out now.
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
starred/hickory-dns#32
No description provided.