Uh oh!
There was an error while loading. Please reload this page.
HBASE-28317: Expose client TLS certificate on RpcCallContext - #5644
Merged
bbeaudreault merged 3 commits intoJan 29, 2024
Conversation
Apache-HBase
commented
Jan 21, 2024
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Jan 22, 2024
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Jan 22, 2024
💔 -1 overall
This message was automatically generated. |
| sslHandler.setWrapDataSize( | ||
| conf.getInt(HBASE_SERVER_NETTY_TLS_WRAP_SIZE, DEFAULT_HBASE_SERVER_NETTY_TLS_WRAP_SIZE)); | ||
| sslHandler.handshakeFuture().addListener(future -> { |
Contributor
There was a problem hiding this comment.
@Apache9 do you think a listener is the right approach here? I'm wondering if we should instead pass the SSLHandler into NettyServerRpcConnection, and directly call handler.handshakeFuture().get() in setupHandlers(). That way it's more explicit that we have the certificate on the connection prior to handling any requests. With the listener approach, I'm not 100% sure we can guarantee that the listener is executed in the eventLoop prior to the channelRead0 in NettyRpcServerPreambleHandler.
Contributor
There was a problem hiding this comment.
We should never do future.get in event loop thread, as it has serious performance impact, or even dead lock...
Contributor
There was a problem hiding this comment.
That makes sense. Do you think we should leave it as is?
Contributor
There was a problem hiding this comment.
https://stackoverflow.com/questions/70654915/order-of-invocation-of-listeners
Norman Maurer answered that the invocation order of listeners is the same with adding order. And I checked the code, if there are multiple listeners netty will call them sequentially, with the adding order. So here, since we are the first listener(at least first in the listeners added by our code), so I think we can guarantee that NettyRpcServerPreambleHandler.channelRead can see the certificates.
| Certificate[] certificates = sslHandler.engine().getSession().getPeerCertificates(); | ||
| if (certificates.length > 0) { | ||
| conn.clientCertificate = (X509Certificate) certificates[0]; | ||
| } else { |
Contributor
There was a problem hiding this comment.
We should inspect the sslHandler.engine().getNeedClientAuth() and throw an exception if no certificate is found here and/or in the exception below. See https://github.com/apache/zookeeper/blob/master/zookeeper-server/src/main/java/org/apache/zookeeper/server/NettyServerCnxnFactory.java#L418-L463 for how zookeeper handles hits.
Another thing I notice in zookeeper is that they set the entire Certificate[] onto the connection, so that someone can inspect the whole certificate chain if they want. I think we should probably do that in order to avoid having to break this interface in the future if someone wants more than the head of the chain.
ContributorAuthor
There was a problem hiding this comment.
I've changed the interface to use the certificate chain. I'm also logging when getNeedClientAuth() is true but no certificate is found. I don't think throwing an exception in this listener is any more useful than logging, it wouldn't block a connection from forming. I also don't necessarily think that it's the job of this code to verify that connections that require mutual auth are doing it. Netty is already enforcing that.
Apache-HBase
commented
Jan 24, 2024
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Jan 24, 2024
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Jan 25, 2024
💔 -1 overall
This message was automatically generated. |
charlesconnell
commented
Jan 25, 2024
ContributorAuthor
Test failures appear unrelated to this PR |
Apache9
approved these changes
Jan 26, 2024
bbeaudreault
approved these changes
Jan 26, 2024
Uh oh!
There was an error while loading. Please reload this page.
Apache-HBase
commented
Jan 26, 2024
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Jan 26, 2024
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Jan 26, 2024
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Jan 28, 2024
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Jan 28, 2024
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Jan 28, 2024
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Jan 28, 2024
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Jan 29, 2024
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Jan 29, 2024
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Jan 29, 2024
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Jan 29, 2024
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Jan 29, 2024
🎊 +1 overall
This message was automatically generated. |
bbeaudreault pushed a commit
that referenced
this pull request
Jan 29, 2024
Signed-off-by: Duo Zhang <zhangduo@apache.org> Signed-off-by: Bryan Beaudreault <bbeaudreault@apache.org>
bbeaudreault pushed a commit
that referenced
this pull request
Jan 29, 2024
Signed-off-by: Duo Zhang <zhangduo@apache.org> Signed-off-by: Bryan Beaudreault <bbeaudreault@apache.org>
bbeaudreault pushed a commit
that referenced
this pull request
Jan 29, 2024
Signed-off-by: Duo Zhang <zhangduo@apache.org> Signed-off-by: Bryan Beaudreault <bbeaudreault@apache.org>
charlesconnell added a commit
to HubSpot/hbase
that referenced
this pull request
Jan 29, 2024
) (#69) Signed-off-by: Duo Zhang <zhangduo@apache.org> Signed-off-by: Bryan Beaudreault <bbeaudreault@apache.org>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for freeto join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
At my employer we plan on using a coprocessor to log information about some requests to HBase. For this to be useful to us, we need to know who each request is coming from. We use HBase's TLS support with mutual authentication to authenticate clients. I'd like a way to expose the client certificate used on a request to coprocessors. For setups using Kerberos authentication, RpcCall exposes the Kerberos principal shortname via
getRequestUser(), so this would be the TLS equivalent to that. I've tested this patch applied to 2.5.x in my company's infrastructure.I found that in order for this patch to be useful to me, I also need netty to fix a related bug which I reported as netty/netty#13796. This doesn't need to block this PR.