chain: read the hook device attributes back - #366
Open
MemorManeo wants to merge 1 commit into
Open
Conversation
AddChain marshals NFTA_HOOK_DEV, but hookFromMsg reads only the hook
number and the priority, so a chain read back from the kernel never
carries the device its hook is attached to. That makes an ingress or
egress chain indistinguishable from one attached to every device.
Read NFTA_HOOK_DEV and the NFTA_HOOK_DEVS list, and add Chain.Devices for
the multi-device case that a single string cannot represent
("devices = { eth0, eth1 }"). AddChain marshals the list when one is set,
so such a chain now round-trips. Setting Device alone is unchanged.
Found while writing a tool that reads rulesets over netlink: without the
device, an ingress chain has to be treated as seeing every packet, which
makes every verdict about such a host unusable.
|
Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA). View this failed invocation of the CLA check for more information. For the most up to date status, view the checks section at the bottom of the pull request. |
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 free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
AddChainmarshalsNFTA_HOOK_DEV, buthookFromMsgreads only the hook number and the priority, so a chain read back from the kernel never carries the device its hook is attached to. An ingress or egress chain is then indistinguishable from one attached to every device.This reads
NFTA_HOOK_DEVand theNFTA_HOOK_DEVSlist, and addsChain.Devicesfor the multi-device case a single string cannot represent (devices = { eth0, eth1 }).AddChainmarshals the list when one is set, so such a chain now round-trips. SettingDevicealone behaves exactly as before, so this is additive for existing callers.NFTA_HOOK_DEVSandNFTA_DEVICE_NAMEare not exported bygolang.org/x/sys/unix, so their values are declared locally with the kernel enum they come from named alongside.Testing
Three tests cover the single-device attribute, the nested list, and a chain with no device at all. The existing suite passes unchanged.
Why
Found while writing a tool that audits which ports a host exposes by reading its ruleset over netlink. Without the device, an ingress chain has to be treated as seeing every packet, which makes every verdict about such a host unusable. The workaround was a second
NFT_MSG_GETCHAINdump alongside this library's own, purely to recover an attribute the library had already read and discarded, which is the kind of thing that belongs upstream instead.