On Apr 23, 2012, at 5:13 PM, Myklebust, Trond wrote:
> On Mon, 2012-04-23 at 16:54 -0400, Chuck Lever wrote:
>> A SETCLIENTID boot verifier is nothing more than a boot timestamp.
>> An NFSv4 server is obligated to wipe all NFSv4 state for an NFS client
>> when the client presents an updated SETCLIENTID boot verifier. This
>> is how servers detect client reboots.
>>
>> nfs4_reset_all_state() forces a boot verifier refresh to cause a
>> server to wipe state as part of recovering from a server reporting
>> that it has revoked some or all of a client's NFSv4 state. This wipes
>> the slate for full state recovery.
>>
>> Soon we want to get rid of the per-nfs_client cl_boot_time field,
>> however. Without cl_boot_time, the NFS client will need to find a
>> different way to force the server to purge the client's NFSv4 state.
>>
>> Because these verifiers are opaque (ie, the server doesn't know or
>> care that they are timestamps), we can do this by using the same
>> trick we use now, but then afterwards establish a fresh client ID
>> using the old boot verifier again.
>>
>> Hopefully there are no extra paranoid server implementations that keep
>> track of the client's boot verifiers and prevent clients from reusing
>> a previous one.
>>
>> Signed-off-by: Chuck Lever <chuck.lever@xxxxxxxxxx>
>> ---
>>
>> fs/nfs/nfs4_fs.h | 1 +
>> fs/nfs/nfs4proc.c | 9 +++++++--
>> fs/nfs/nfs4state.c | 7 ++++++-
>> 3 files changed, 14 insertions(+), 3 deletions(-)
>>
>> diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
>> index d8c2d39..2953f2c 100644
>> --- a/fs/nfs/nfs4_fs.h
>> +++ b/fs/nfs/nfs4_fs.h
>> @@ -24,6 +24,7 @@ enum nfs4_client_state {
>> NFS4CLNT_RECALL_SLOT,
>> NFS4CLNT_LEASE_CONFIRM,
>> NFS4CLNT_SERVER_SCOPE_MISMATCH,
>> + NFS4CLNT_PURGE_STATE,
>> };
>>
>> enum nfs4_session_state {
>> diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
>> index 84a26d9..b19cf81 100644
>> --- a/fs/nfs/nfs4proc.c
>> +++ b/fs/nfs/nfs4proc.c
>> @@ -3878,8 +3878,13 @@ static void nfs4_construct_boot_verifier(struct nfs_client *clp,
>> {
>> __be32 verf[2];
>>
>> - verf[0] = (__be32)clp->cl_boot_time.tv_sec;
>> - verf[1] = (__be32)clp->cl_boot_time.tv_nsec;
>> + if (test_bit(NFS4CLNT_PURGE_STATE, &clp->cl_state)) {
>> + verf[0] = (__be32)CURRENT_TIME.tv_sec;
>> + verf[1] = (__be32)CURRENT_TIME.tv_nsec;
>> + } else {
>> + verf[0] = (__be32)clp->cl_boot_time.tv_sec;
>> + verf[1] = (__be32)clp->cl_boot_time.tv_nsec;
>> + }
>> memcpy(bootverf->data, verf, sizeof(bootverf->data));
>> }
>>
>> diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
>> index cbef366..7f56502 100644
>> --- a/fs/nfs/nfs4state.c
>> +++ b/fs/nfs/nfs4state.c
>> @@ -1612,7 +1612,7 @@ void nfs41_handle_recall_slot(struct nfs_client *clp)
>> static void nfs4_reset_all_state(struct nfs_client *clp)
>> {
>> if (test_and_set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) == 0) {
>> - clp->cl_boot_time = CURRENT_TIME;
>> + set_bit(NFS4CLNT_PURGE_STATE, &clp->cl_state);
>> nfs4_state_start_reclaim_nograce(clp);
>> nfs4_schedule_state_manager(clp);
>> }
>> @@ -1759,6 +1759,11 @@ static void nfs4_state_manager(struct nfs_client *clp)
>>
>> /* Ensure exclusive access to NFSv4 state */
>> do {
>> + if (test_bit(NFS4CLNT_PURGE_STATE, &clp->cl_state)) {
>> + nfs4_reclaim_lease(clp);
>> + clear_bit(NFS4CLNT_PURGE_STATE, &clp->cl_state);
>
> This needs to set NFS4CLNT_LEASE_EXPIRED, so that we do a second
> SETCLIENTID after the state has been cleared. Otherwise we end up with a
> lease with the wrong verifier.
Agreed, brain fart. I had thought part of the reclaim lease path set it, but it doesn't.
Obviously these are not thoroughly tested yet. I forgot to set the "-E" flag on "stg mail" when sending these so I didn't have a cover letter to mention that.
>
>> + }
>> +
>> if (test_and_clear_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state)) {
>> /* We're going to have to re-establish a clientid */
>> status = nfs4_reclaim_lease(clp);
>>
>
> --
> Trond Myklebust
> Linux NFS client maintainer
>
> NetApp
> Trond.Myklebust@xxxxxxxxxx
> www.netapp.com
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at http://vger.kernel.org/majordomo-info.html
--
Chuck Lever
chuck[dot]lever[at]oracle[dot]com
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
