| Message ID | 1738263687-28256-1-git-send-email-dai.ngo@oracle.com (mailing list archive) |
|---|---|
| State | Under Review |
| Delegated to: | Chuck Lever |
| Headers | show |
| Series | [v2,1/1] NFSD: fix hang in nfsd4_shutdown_callback | expand |
On Thu, 2025-01-30 at 11:01 -0800, Dai Ngo wrote: > If nfs4_client is in courtesy state then there is no point to retry > the callback. This causes nfsd4_shutdown_callback to hang since > cl_cb_inflight is not 0. This hang lasts about 15 minutes until TCP > notifies NFSD that the connection was dropped. > > This patch modifies nnfsd4_run_cb_work to skip the RPC call if > nfs4_client is in courtesy state. > > Signed-off-by: Dai Ngo <dai.ngo@oracle.com> > --- > fs/nfsd/nfs4callback.c | 7 +++++-- > 1 file changed, 5 insertions(+), 2 deletions(-) > > diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c > index 50e468bdb8d4..cf6d29828f4e 100644 > --- a/fs/nfsd/nfs4callback.c > +++ b/fs/nfsd/nfs4callback.c > @@ -1583,8 +1583,11 @@ nfsd4_run_cb_work(struct work_struct *work) > nfsd4_process_cb_update(cb); > > clnt = clp->cl_cb_client; > - if (!clnt) { > - /* Callback channel broken, or client killed; give up: */ > + if (!clnt || clp->cl_state == NFSD4_COURTESY) { > + /* > + * Callback channel broken, client killed or > + * nfs4_client in courtesy state; give up. > + */ > nfsd41_destroy_cb(cb); > return; > } Reviewed-by: Jeff Layton <jlayton@kernel.org>
For subsequent postings, please send to all NFSD reviewers listed in the kernel's MAINTAINERS file. Thanks! On 1/30/25 2:01 PM, Dai Ngo wrote: > If nfs4_client is in courtesy state then there is no point to retry > the callback. Perhaps this should be amended "there is no point to send the callback." because this revision of the patch affects first transmissions as well as retries. > This causes nfsd4_shutdown_callback to hang since > cl_cb_inflight is not 0. This hang lasts about 15 minutes until TCP > notifies NFSD that the connection was dropped. > > This patch modifies nnfsd4_run_cb_work to skip the RPC call if > nfs4_client is in courtesy state. > Fixes: 66af25799940 ("NFSD: add courteous server support for thread with only delegation") Cc: stable@vger.kernel.org > Signed-off-by: Dai Ngo <dai.ngo@oracle.com> > --- > fs/nfsd/nfs4callback.c | 7 +++++-- > 1 file changed, 5 insertions(+), 2 deletions(-) > > diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c > index 50e468bdb8d4..cf6d29828f4e 100644 > --- a/fs/nfsd/nfs4callback.c > +++ b/fs/nfsd/nfs4callback.c > @@ -1583,8 +1583,11 @@ nfsd4_run_cb_work(struct work_struct *work) > nfsd4_process_cb_update(cb); > > clnt = clp->cl_cb_client; > - if (!clnt) { > - /* Callback channel broken, or client killed; give up: */ > + if (!clnt || clp->cl_state == NFSD4_COURTESY) { > + /* > + * Callback channel broken, client killed or > + * nfs4_client in courtesy state; give up. > + */ > nfsd41_destroy_cb(cb); > return; > }
From: Chuck Lever <chuck.lever@oracle.com> On Thu, 30 Jan 2025 11:01:27 -0800, Dai Ngo wrote: > If nfs4_client is in courtesy state then there is no point to retry > the callback. This causes nfsd4_shutdown_callback to hang since > cl_cb_inflight is not 0. This hang lasts about 15 minutes until TCP > notifies NFSD that the connection was dropped. > > This patch modifies nnfsd4_run_cb_work to skip the RPC call if > nfs4_client is in courtesy state. > > [...] Applied to nfsd-testing, thanks! [1/1] NFSD: fix hang in nfsd4_shutdown_callback commit: c1d827093999bd9c19d81b0af4b3034a55b49d5a -- Chuck Lever
diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index 50e468bdb8d4..cf6d29828f4e 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -1583,8 +1583,11 @@ nfsd4_run_cb_work(struct work_struct *work) nfsd4_process_cb_update(cb); clnt = clp->cl_cb_client; - if (!clnt) { - /* Callback channel broken, or client killed; give up: */ + if (!clnt || clp->cl_state == NFSD4_COURTESY) { + /* + * Callback channel broken, client killed or + * nfs4_client in courtesy state; give up. + */ nfsd41_destroy_cb(cb); return; }
If nfs4_client is in courtesy state then there is no point to retry the callback. This causes nfsd4_shutdown_callback to hang since cl_cb_inflight is not 0. This hang lasts about 15 minutes until TCP notifies NFSD that the connection was dropped. This patch modifies nnfsd4_run_cb_work to skip the RPC call if nfs4_client is in courtesy state. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> --- fs/nfsd/nfs4callback.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)