Bug 215526
Summary: | noresvport + nfsv4 no longer working as expected | ||
---|---|---|---|
Product: | File System | Reporter: | Daire Byrne (daire) |
Component: | NFS | Assignee: | Trond Myklebust (trondmy) |
Status: | NEW --- | ||
Severity: | normal | CC: | bfields, pmenzel+bugzilla.kernel.org |
Priority: | P1 | ||
Hardware: | All | ||
OS: | Linux | ||
Kernel Version: | 5.12+ | Subsystem: | |
Regression: | No | Bisected commit-id: |
Description
Daire Byrne
2022-01-24 12:30:46 UTC
Also in that range and look potentially relevant on a quick skim: 87871d990a2c NFSv4: Initialise connection to the server in nfs4_alloc_client() aebe64667664 SUNRPC: prevent port reuse on transports which don't request it. Yea, I just applied: 87871d990a2c NFSv4: Initialise connection to the server in nfs4_alloc_client() on top of v5.11 and that was enough to break noresvport for NFSv4.2. Daire So that is commit dd99e9f98fbf (NFSv4: Initialise connection to the server in nfs4_alloc_client()), which went into v5.14-rc1. Trond’s commit 468d126dab45 (NFS: Fix initialisation of nfs_client cl_flags field) has the problematic commit in its fixes tag [1]. It was just pulled by Linus, and is going to be in 5.17-rc4. Can you please test, if that fixes your issue? [1]: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=468d126dab45718feeb728319be20bd869a5eaa7 Apologies, I forgot to update this ticket last week. Yes, I can confirm that Trond's recent patch does indeed fix this issue. Cheers, Daire Unfortunately, the issues is not referenced in the commit, but as the commit is referenced here now, I guess it can be closed as fixed. |