Disabling the veth rename when sandboxing
Nobody has claimed this yet.
- Dominant language
- Go
- Stars
- 2.2k
- Forks
- 875
- PR merge metrics
- No merged PRs in 30d
Description
When creating a veth pair, and sandboxing the host part inside a network namespace on linux, is there a reason why the interface is renamed from e.g. vethabcdef0 to veth0 (or above if already taken), except for ordering them inside the network namespace?
We are encountering an issue because we are interfacing with the kernel at the time of interface deletion and trying to correlate that with data collected at runtime. But the mapping is known to libnetwork alone and not even logged, so we can only know about "veth0" at runtime, while the deletion tells us that "vethabcdef0" has been deleted.
Would it be possible to disable the rename, or log the operation with relevant metadata so it can be acted upon?
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No files or tests are named. Start by tracing Linux veth-pair creation and the host-interface rename in libnetwork, then examine how interface deletion metadata is exposed. Done means either the rename can be disabled during sandboxing or the original-to-renamed mapping is logged with enough metadata for correlation.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- go, linux
- Domain
- networking, operating-systems
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100