Applying, thanks, sorry for the delay.--b. On Wed, Jun 15, 2022 at 12:48:17PM -0700, dai.ngo@xxxxxxxxxx wrote: > > On 6/15/22 12:34 PM, J. Bruce Fields wrote: > >THere are tests that want to explicitly test for DELAY returns. (Grep > >for ERR_DELAY. Look at the delegation tests especially.) Does this > >work for them? > > Those tests expect NFS4_OK but also handle NFS4ERR_DELAY themselves > if the OPEN causes recall. With this patch, the NFS4ERR_DELAY is handled > internally by open_file so the ERR_DELAY never get to those tests. > All tests passed with this patch. > > -Dai > > > I assumed we'd want an optional parameter that allowed > >to caller to circument the DELAY handling. > > > >--b. > > > >On Wed, Jun 15, 2022 at 10:34:54AM -0700, Dai Ngo wrote: > >>Enhance open_file to handle NFS4ERR_DELAY returned by the server > >>in case of share/access/delegation conflict. > >> > >>Signed-off-by: Dai Ngo <dai.ngo@xxxxxxxxxx> > >>--- > >> nfs4.0/nfs4lib.py | 7 ++++++- > >> 1 file changed, 6 insertions(+), 1 deletion(-) > >> > >>diff --git a/nfs4.0/nfs4lib.py b/nfs4.0/nfs4lib.py > >>index 934def3b7333..e0299e8d6676 100644 > >>--- a/nfs4.0/nfs4lib.py > >>+++ b/nfs4.0/nfs4lib.py > >>@@ -677,7 +677,12 @@ class NFS4Client(rpc.RPCClient): > >> claim_type=claim_type, deleg_type=deleg_type, > >> deleg_cur_info=deleg_cur_info)] > >> ops += [op4.getfh()] > >>- res = self.compound(ops) > >>+ while 1: > >>+ res = self.compound(ops) > >>+ if res.status == NFS4ERR_DELAY: > >>+ time.sleep(2) > >>+ else: > >>+ break > >> self.advance_seqid(owner, res) > >> if set_recall and (res.status != NFS4_OK or \ > >> res.resarray[-2].switch.switch.delegation == OPEN_DELEGATE_NONE): > >>-- > >>2.27.0
