linux-nfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Chuck Lever III <chuck.lever@oracle.com>
To: Bruce Fields <bfields@fieldses.org>
Cc: Linux NFS Mailing List <linux-nfs@vger.kernel.org>,
	David Wysochanski <dwysocha@redhat.com>
Subject: Re: [PATCH RFC 21/21] NFSD: Add tracepoints to observe clientID activity
Date: Tue, 11 May 2021 15:59:00 +0000	[thread overview]
Message-ID: <7922E4B6-BB7F-4D9E-B85D-D1A97835AF3F@oracle.com> (raw)
In-Reply-To: <162066202717.94415.8666073108309704792.stgit@klimt.1015granger.net>

As Dave reported yesterday, this patch is unfinished and is probably
junk. But any thoughts on how the tracepoints should be organized
in this code would help.

So I was thinking we probably want a tracepoint to fire for each
case that is handled in this code (and in nfsd4_exchangeid).
However, this comment in nfsd4_setclientid:

   /* Cases below refer to rfc 3530 section 14.2.33: */

Is confusing.

- RFC 3530 is superceded by RFC 7530, and the section numbers have changed.

- The cases in this section in both RFCs aren't numbered, they are
bullet points.


> On May 10, 2021, at 11:53 AM, Chuck Lever <chuck.lever@oracle.com> wrote:
> 
> We are especially interested in capturing clientID conflicts.
> 
> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
> ---
> fs/nfsd/nfs4state.c |    9 +++++++--
> fs/nfsd/trace.h     |   37 +++++++++++++++++++++++++++++++++++++
> 2 files changed, 44 insertions(+), 2 deletions(-)
> 
> diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
> index a61601fe422a..528cabffa1e9 100644
> --- a/fs/nfsd/nfs4state.c
> +++ b/fs/nfsd/nfs4state.c
> @@ -3180,6 +3180,7 @@ nfsd4_exchange_id(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
> 			}
> 			/* case 6 */
> 			exid->flags |= EXCHGID4_FLAG_CONFIRMED_R;
> +			trace_nfsd_clid_existing(conf);
> 			goto out_copy;
> 		}
> 		if (!creds_match) { /* case 3 */
> @@ -3188,15 +3189,18 @@ nfsd4_exchange_id(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
> 				trace_nfsd_clid_cred_mismatch(conf, rqstp);
> 				goto out;
> 			}
> +			trace_nfsd_clid_new(new);
> 			goto out_new;
> 		}
> 		if (verfs_match) { /* case 2 */
> 			conf->cl_exchange_flags |= EXCHGID4_FLAG_CONFIRMED_R;
> +			trace_nfsd_clid_existing(conf);
> 			goto out_copy;
> 		}
> 		/* case 5, client reboot */
> 		trace_nfsd_clid_verf_mismatch(conf, rqstp, &verf);
> 		conf = NULL;
> +		trace_nfsd_clid_new(new);
> 		goto out_new;
> 	}
> 
> @@ -3996,10 +4000,12 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
> 		if (same_verf(&conf->cl_verifier, &clverifier)) {
> 			copy_clid(new, conf);
> 			gen_confirm(new, nn);
> +			trace_nfsd_clid_existing(new);
> 		} else
> 			trace_nfsd_clid_verf_mismatch(conf, rqstp,
> 						      &clverifier);
> -	}
> +	} else
> +		trace_nfsd_clid_new(new);
> 	new->cl_minorversion = 0;
> 	gen_callback(new, setclid, rqstp);
> 	add_to_unconfirmed(new);
> @@ -4017,7 +4023,6 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
> 	return status;
> }
> 
> -
> __be32
> nfsd4_setclientid_confirm(struct svc_rqst *rqstp,
> 			struct nfsd4_compound_state *cstate,
> diff --git a/fs/nfsd/trace.h b/fs/nfsd/trace.h
> index 523045c37749..6ddff13e3181 100644
> --- a/fs/nfsd/trace.h
> +++ b/fs/nfsd/trace.h
> @@ -626,6 +626,43 @@ TRACE_EVENT(nfsd_clid_verf_mismatch,
> 	)
> );
> 
> +DECLARE_EVENT_CLASS(nfsd_clid_class,
> +	TP_PROTO(const struct nfs4_client *clp),
> +	TP_ARGS(clp),
> +	TP_STRUCT__entry(
> +		__field(u32, cl_boot)
> +		__field(u32, cl_id)
> +		__array(unsigned char, addr, sizeof(struct sockaddr_in6))
> +		__field(unsigned long, flavor)
> +		__array(unsigned char, verifier, NFS4_VERIFIER_SIZE)
> +		__field(unsigned int, namelen)
> +		__dynamic_array(unsigned char, name, clp->cl_name.len)
> +	),
> +	TP_fast_assign(
> +		memcpy(__entry->addr, &clp->cl_addr,
> +			sizeof(struct sockaddr_in6));
> +		__entry->flavor = clp->cl_cred.cr_flavor;
> +		memcpy(__entry->verifier, (void *)&clp->cl_verifier,
> +		       NFS4_VERIFIER_SIZE);
> +		__entry->namelen = clp->cl_name.len;
> +		memcpy(__get_dynamic_array(name), clp->cl_name.data,
> +			clp->cl_name.len);
> +	),
> +	TP_printk("addr=%pISpc name='%.*s' verifier=0x%s flavor=%s client=%08x:%08x\n",
> +		__entry->addr, __entry->namelen, __get_str(name),
> +		__print_hex_str(__entry->verifier, NFS4_VERIFIER_SIZE),
> +		show_nfsd_authflavor(__entry->flavor),
> +		__entry->cl_boot, __entry->cl_id)
> +);
> +
> +#define DEFINE_CLID_EVENT(name) \
> +DEFINE_EVENT(nfsd_clid_class, nfsd_clid_##name, \
> +	TP_PROTO(const struct nfs4_client *clp), \
> +	TP_ARGS(clp))
> +
> +DEFINE_CLID_EVENT(new);
> +DEFINE_CLID_EVENT(existing);
> +
> /*
>  * from fs/nfsd/filecache.h
>  */
> 
> 

--
Chuck Lever




  parent reply	other threads:[~2021-05-11 15:59 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-10 15:51 [PATCH RFC 00/21] NFSD callback and lease management observability Chuck Lever
2021-05-10 15:51 ` [PATCH RFC 01/21] NFSD: Constify @fh argument of knfsd_fh_hash() Chuck Lever
2021-05-10 15:51 ` [PATCH RFC 02/21] NFSD: Capture every CB state transition Chuck Lever
2021-05-10 15:51 ` [PATCH RFC 03/21] NFSD: Drop TRACE_DEFINE_ENUM for NFSD4_CB_<state> macros Chuck Lever
2021-05-10 15:52 ` [PATCH RFC 04/21] NFSD: Add cb_lost tracepoint Chuck Lever
2021-05-10 15:52 ` [PATCH RFC 05/21] NFSD: Adjust cb_shutdown tracepoint Chuck Lever
2021-05-10 15:52 ` [PATCH RFC 06/21] NFSD: Remove spurious cb_setup_err tracepoint Chuck Lever
2021-05-10 20:27   ` J. Bruce Fields
2021-05-10 20:29     ` Chuck Lever III
2021-05-11 17:44       ` Bruce Fields
2021-05-10 15:52 ` [PATCH RFC 07/21] NFSD: Enhance the nfsd_cb_setup tracepoint Chuck Lever
2021-05-10 15:52 ` [PATCH RFC 08/21] NFSD: Add an nfsd_cb_lm_notify tracepoint Chuck Lever
2021-05-10 15:52 ` [PATCH RFC 09/21] NFSD: Add an nfsd_cb_offload tracepoint Chuck Lever
2021-05-14 18:22   ` Olga Kornievskaia
2021-05-14 19:11     ` Chuck Lever III
2021-05-10 15:52 ` [PATCH RFC 10/21] NFSD: Replace the nfsd_deleg_break tracepoint Chuck Lever
2021-05-10 15:52 ` [PATCH RFC 11/21] NFSD: Add an nfsd_cb_probe tracepoint Chuck Lever
2021-05-10 15:52 ` [PATCH RFC 12/21] NFSD: Remove the nfsd_cb_work and nfsd_cb_done tracepoints Chuck Lever
2021-05-10 15:52 ` [PATCH RFC 13/21] NFSD: Update nfsd_cb_args tracepoint Chuck Lever
2021-05-10 15:53 ` [PATCH RFC 14/21] NFSD: Add nfsd_clid_cred_mismatch tracepoint Chuck Lever
2021-05-10 15:53 ` [PATCH RFC 15/21] NFSD: Add nfsd_clid_verf_mismatch tracepoint Chuck Lever
2021-05-11 15:49   ` J. Bruce Fields
2021-05-10 15:53 ` [PATCH RFC 16/21] NFSD: Remove nfsd_clid_inuse_err Chuck Lever
2021-05-10 15:53 ` [PATCH RFC 17/21] NFSD: Add nfsd_clid_confirmed tracepoint Chuck Lever
2021-05-10 15:53 ` [PATCH RFC 18/21] NFSD: Add nfsd_clid_destroyed tracepoint Chuck Lever
2021-05-10 15:53 ` [PATCH RFC 19/21] NFSD: Add a couple more nfsd_clid_expired call sites Chuck Lever
2021-05-10 15:53 ` [PATCH RFC 20/21] NFSD: Rename nfsd_clid_class Chuck Lever
2021-05-10 15:53 ` [PATCH RFC 21/21] NFSD: Add tracepoints to observe clientID activity Chuck Lever
2021-05-10 19:56   ` David Wysochanski
2021-05-10 20:00     ` Chuck Lever III
2021-05-11 15:59   ` Chuck Lever III [this message]
2021-05-11 17:38     ` Bruce Fields
2021-05-11 17:40       ` Chuck Lever III
2021-05-11 17:43         ` Bruce Fields
2021-05-10 17:24 ` [PATCH RFC 00/21] NFSD callback and lease management observability David Wysochanski

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=7922E4B6-BB7F-4D9E-B85D-D1A97835AF3F@oracle.com \
    --to=chuck.lever@oracle.com \
    --cc=bfields@fieldses.org \
    --cc=dwysocha@redhat.com \
    --cc=linux-nfs@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).