Skip to contentSkip to conversation

NFSv4: try lease recovery on NFS4ERR_EXPIRED

PATCH Superseded Robert Milkowski 2 participants 6 messages Jan 8, 2020 – Jan 27, 2020

Subsystems (2)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 NFSv4: try lease recovery on NFS4ERR_EXPIRED

From: Robert Milkowski <[email protected]>

Currently, if an nfs server returns NFS4ERR_EXPIRED to open(), etc. we return EIO to applications without even trying to recover.

Fixes 272289a3df72 ("NFSv4: nfs4_do_handle_exception() handle revoke/expiry

of a single stateid")

---
 fs/nfs/nfs4proc.c | 4 ++++
 1 file changed, 4 insertions(+)
fs/nfs/nfs4proc.c +0 -0
index 76d3716..2478405 100644
@@ -481,6 +481,10 @@ static int nfs4_do_handle_exception(struct nfs_server

*server,
stateid);
goto wait_on_recovery;
}
+ if (state == NULL) {
+ nfs4_schedule_lease_recovery(clp);
+ goto wait_on_recovery;
+ }
/* Fall through */
case -NFS4ERR_OPENMODE:
if (inode) {

Signature · 1 line
1.8.3.1
#1 lore raw

Anyone please?

-----Original Message-----
From: Robert Milkowski <[email protected]>
Sent: 08 January 2020 21:48
To: [email protected]
Cc: 'Trond Myklebust' <[email protected]>; 'Chuck Lever' <[email protected]>; 'Anna Schumaker' <[email protected]>; [email protected]
Subject: [PATCH v2] NFSv4: try lease recovery on NFS4ERR_EXPIRED

From: Robert Milkowski <[email protected]>

Currently, if an nfs server returns NFS4ERR_EXPIRED to open(), etc. we return EIO to applications without even trying to recover.

Fixes 272289a3df72 ("NFSv4: nfs4_do_handle_exception() handle revoke/expiry

of a single stateid")

---
 fs/nfs/nfs4proc.c | 4 ++++
 1 file changed, 4 insertions(+)
fs/nfs/nfs4proc.c index 76d3716..2478405 renamed from fs/nfs/nfs4proc.c +0 -0

100644

fs/nfs/nfs4proc.c +0 -0
@@ -481,6 +481,10 @@ static int nfs4_do_handle_exception(struct nfs_server

*server,
stateid);
goto wait_on_recovery;
}
+ if (state == NULL) {
+ nfs4_schedule_lease_recovery(clp);
+ goto wait_on_recovery;
+ }
/* Fall through */
case -NFS4ERR_OPENMODE:
if (inode) {
--
1.8.3.1

#2 lore raw
Quoted 45 lines · Robert Milkowski

Anyone please?

-----Original Message-----
From: Robert Milkowski <[email protected]>
Sent: 08 January 2020 21:48
To: [email protected]
Cc: 'Trond Myklebust' <[email protected]>; 'Chuck Lever' <[email protected]>; 'Anna Schumaker' <[email protected]

Quoted 1 line

;

[email protected]
Subject: [PATCH v2] NFSv4: try lease recovery on NFS4ERR_EXPIRED

From: Robert Milkowski <[email protected]>

Currently, if an nfs server returns NFS4ERR_EXPIRED to open(), etc. we return EIO to applications without even trying to recover.

Fixes 272289a3df72 ("NFSv4: nfs4_do_handle_exception() handle

revoke/expiry
of a single stateid")

---
 fs/nfs/nfs4proc.c | 4 ++++
 1 file changed, 4 insertions(+)
fs/nfs/nfs4proc.c index renamed from fs/nfs/nfs4proc.c +0 -0

76d3716..2478405
100644

fs/nfs/nfs4proc.c +0 -0
@@ -481,6 +481,10 @@ static int nfs4_do_handle_exception(struct

nfs_server
*server,

fs/nfs/nfs4proc.c +4 -1
 						stateid);
 				goto wait_on_recovery;
 			}
+			if (state == NULL) {
+				nfs4_schedule_lease_recovery(clp);
+				goto wait_on_recovery;
+			}
 			/* Fall through */
 		case -NFS4ERR_OPENMODE:
 			if (inode) {
--
Quoted 3 lines

1.8.3.1

Does this apply to any case other than NFS4ERR_EXPIRED in the specific case of nfs4_do_open()? I can't see that it does. It looks to me as if the open recovery routines already have their own handling of this case.

If so, why not just add it as a special case in the nfs4_do_open() error handling? Otherwise this patch will end up overriding other generic cases where we have an inode, but no open state.

Note that _nfs4_do_open() already waits for lease recovery, so we only need the call to nfs_schedule_lease_recovery().

Signature · 3 lines
Trond Myklebust
Linux NFS client maintainer, Hammerspace
[email protected]
#3 lore raw
Quoted 51 lines

-----Original Message-----
From: Trond Myklebust <[email protected]>
Sent: 17 January 2020 17:24
To: [email protected]; [email protected]
Cc: [email protected]; [email protected]; [email protected]
Subject: Re: [PATCH v2] NFSv4: try lease recovery on NFS4ERR_EXPIRED

Quoted 42 lines · Robert Milkowski

Anyone please?

-----Original Message-----
From: Robert Milkowski <[email protected]>
Sent: 08 January 2020 21:48
To: [email protected]
Cc: 'Trond Myklebust' <[email protected]>; 'Chuck Lever' <[email protected]>; 'Anna Schumaker' <[email protected]

Quoted 1 line

;

[email protected]
Subject: [PATCH v2] NFSv4: try lease recovery on NFS4ERR_EXPIRED

From: Robert Milkowski <[email protected]>

Currently, if an nfs server returns NFS4ERR_EXPIRED to open(), etc. we return EIO to applications without even trying to recover.

Fixes 272289a3df72 ("NFSv4: nfs4_do_handle_exception() handle

revoke/expiry of a single stateid")

---
 fs/nfs/nfs4proc.c | 4 ++++
 1 file changed, 4 insertions(+)
fs/nfs/nfs4proc.c index renamed from fs/nfs/nfs4proc.c +0 -0

76d3716..2478405
100644

fs/nfs/nfs4proc.c +4 -0
@@ -481,6 +481,10 @@ static int nfs4_do_handle_exception(struct
 nfs_server *server,
 						stateid);
 				goto wait_on_recovery;
 			}
+			if (state == NULL) {
+				nfs4_schedule_lease_recovery(clp);
+				goto wait_on_recovery;
+			}
 			/* Fall through */
 		case -NFS4ERR_OPENMODE:

if (inode) {

Quoted 8 lines
Quoted 4 lines

--
1.8.3.1

Does this apply to any case other than NFS4ERR_EXPIRED in the specific case of nfs4_do_open()? I can't see that it does. It looks to me as if the open recovery routines already have their own handling of this case.

I only observed the issue with open(). After further review I think you are right and it only applies to nfs4_do_open().

Quoted 5 lines

If so, why not just add it as a special case in the nfs4_do_open() error handling? Otherwise this patch will end up overriding other generic cases where we have an inode, but no open state.

Fair point.
So perhaps, few lines further instead of:

			if (inode) {
...
			if (state == NULL) {
					break;
			}

There should be:

			if (inode) {
...
			if (state == NULL) {
				nfs4_schedule_lease_recovery(clp);
				goto wait_on_recovery;
			}

This way we know that inode cannot be null at this point, and it's a case where both inode and state are NULL. This would be a little bit more general in case we reach this point.

But if you think it is better to move it to nfs4_do_open() then I've just tested the following patch:

fs/nfs/nfs4proc.c +5 -0
index 76d3716..b7c4044 100644
@@ -3187,6 +3187,11 @@ static struct nfs4_state *nfs4_do_open(struct inode *dir,
                        exception.retry = 1;
                        continue;
                }
+               if (status == -NFS4ERR_EXPIRED) {
+                       nfs4_schedule_lease_recovery(server->nfs_client);
+                       exception.retry = 1;
+                       continue;
+               }
                if (status == -EAGAIN) {
                        /* We must have found a delegation */
                        exception.retry = 1;

Please let me know which way you want to proceed and I will submit an updated patch.

Quoted 3 lines

Note that _nfs4_do_open() already waits for lease recovery, so we only need the call to nfs_schedule_lease_recovery().

Yep

Signature · 1 line
Robert Milkowski
#4 lore raw
Quoted 119 lines · Robert Milkowski
Quoted 57 lines

-----Original Message-----
From: Trond Myklebust <[email protected]>
Sent: 17 January 2020 17:24
To: [email protected]; [email protected]
Cc: [email protected]; [email protected]; [email protected]
Subject: Re: [PATCH v2] NFSv4: try lease recovery on NFS4ERR_EXPIRED

Quoted 47 lines · Robert Milkowski

Anyone please?

-----Original Message-----
From: Robert Milkowski <[email protected]>
Sent: 08 January 2020 21:48
To: [email protected]
Cc: 'Trond Myklebust' <[email protected]>; 'Chuck Lever' <[email protected]>; 'Anna Schumaker' < [email protected]

Quoted 1 line

;

[email protected]
Subject: [PATCH v2] NFSv4: try lease recovery on NFS4ERR_EXPIRED

From: Robert Milkowski <[email protected]>

Currently, if an nfs server returns NFS4ERR_EXPIRED to open(), etc.
we return EIO to applications without even trying to recover.

Fixes 272289a3df72 ("NFSv4: nfs4_do_handle_exception() handle

revoke/expiry of a single stateid")

---
 fs/nfs/nfs4proc.c | 4 ++++
 1 file changed, 4 insertions(+)
fs/nfs/nfs4proc.c index renamed from fs/nfs/nfs4proc.c +0 -0

76d3716..2478405
100644

fs/nfs/nfs4proc.c +4 -0
@@ -481,6 +481,10 @@ static int nfs4_do_handle_exception(struct
 nfs_server *server,
 						stateid);
 				goto wait_on_recovery;
 			}
+			if (state == NULL) {
+				nfs4_schedule_lease_recovery(clp);
+				goto wait_on_recovery;
+			}
 			/* Fall through */
 		case -NFS4ERR_OPENMODE:

if (inode) {
--
1.8.3.1

Quoted 8 lines
Quoted 1 line

Does this apply to any case other than NFS4ERR_EXPIRED in the specific
case of nfs4_do_open()? I can't see that it does. It looks to me as if
the open recovery routines already have their own handling of this case.

I only observed the issue with open(). After further review I think you are right and it only applies to nfs4_do_open().

Quoted 5 lines

If so, why not just add it as a special case in the nfs4_do_open() error
handling? Otherwise this patch will end up overriding other generic cases where we have an inode, but no open state.

Fair point.
So perhaps, few lines further instead of:

			if (inode) {
...
			if (state == NULL) {
					break;
			}

There should be:

			if (inode) {
...
			if (state == NULL) {
				nfs4_schedule_lease_recovery(clp);
				goto wait_on_recovery;
			}

This way we know that inode cannot be null at this point, and it's a case where both inode and state are NULL.
This would be a little bit more general in case we reach this point.

But if you think it is better to move it to nfs4_do_open() then I've just tested the following patch:

fs/nfs/nfs4proc.c +5 -0
index 76d3716..b7c4044 100644
@@ -3187,6 +3187,11 @@ static struct nfs4_state *nfs4_do_open(struct
 inode *dir,
                        exception.retry = 1;
                        continue;
                }
+               if (status == -NFS4ERR_EXPIRED) {
+                       nfs4_schedule_lease_recovery(server-
 >nfs_client);
+                       exception.retry = 1;
+                       continue;
+               }
                if (status == -EAGAIN) {
Quoted 3 lines
                        /* We must have found a delegation */
                        exception.retry = 1;

This looks like what I'm asking for, yes. That seems like the minimal patch that addresses the problem you're describing.

Signature · 3 lines
Trond Myklebust
Linux NFS client maintainer, Hammerspace
[email protected]
#5 lore raw
Quoted 125 lines · Trond Myklebust
Quoted 119 lines · Robert Milkowski
Quoted 57 lines

-----Original Message-----
From: Trond Myklebust <[email protected]>
Sent: 17 January 2020 17:24
To: [email protected]; [email protected]
Cc: [email protected]; [email protected]; [email protected]
Subject: Re: [PATCH v2] NFSv4: try lease recovery on NFS4ERR_EXPIRED

Quoted 47 lines · Robert Milkowski

Anyone please?

-----Original Message-----
From: Robert Milkowski <[email protected]>
Sent: 08 January 2020 21:48
To: [email protected]
Cc: 'Trond Myklebust' <[email protected]>; 'Chuck Lever' <[email protected]>; 'Anna Schumaker' < [email protected]

Quoted 1 line

;

[email protected]
Subject: [PATCH v2] NFSv4: try lease recovery on NFS4ERR_EXPIRED

From: Robert Milkowski <[email protected]>

Currently, if an nfs server returns NFS4ERR_EXPIRED to open(), etc.
we return EIO to applications without even trying to recover.

Fixes 272289a3df72 ("NFSv4: nfs4_do_handle_exception() handle

revoke/expiry of a single stateid")

---
 fs/nfs/nfs4proc.c | 4 ++++
 1 file changed, 4 insertions(+)
fs/nfs/nfs4proc.c index renamed from fs/nfs/nfs4proc.c +0 -0

76d3716..2478405
100644

fs/nfs/nfs4proc.c +4 -0
@@ -481,6 +481,10 @@ static int nfs4_do_handle_exception(struct
 nfs_server *server,
                                          stateid);
                          goto wait_on_recovery;
                  }
+                 if (state == NULL) {
+                         nfs4_schedule_lease_recovery(clp);
+                         goto wait_on_recovery;
+                 }
                  /* Fall through */
          case -NFS4ERR_OPENMODE:
                  if (inode) {

--
1.8.3.1

Quoted 8 lines
Quoted 1 line

Does this apply to any case other than NFS4ERR_EXPIRED in the specific
case of nfs4_do_open()? I can't see that it does. It looks to me as if
the open recovery routines already have their own handling of this case.

I only observed the issue with open(). After further review I think you are right and it only applies to nfs4_do_open().

Quoted 5 lines

If so, why not just add it as a special case in the nfs4_do_open() error
handling? Otherwise this patch will end up overriding other generic cases where we have an inode, but no open state.

Fair point.
So perhaps, few lines further instead of:

                      if (inode) {
...
                      if (state == NULL) {
                                      break;
                      }

There should be:

                      if (inode) {
...
                      if (state == NULL) {
                              nfs4_schedule_lease_recovery(clp);
                              goto wait_on_recovery;
                      }

This way we know that inode cannot be null at this point, and it's a case where both inode and state are NULL.
This would be a little bit more general in case we reach this point.

But if you think it is better to move it to nfs4_do_open() then I've just tested the following patch:

fs/nfs/nfs4proc.c +5 -0
index 76d3716..b7c4044 100644
@@ -3187,6 +3187,11 @@ static struct nfs4_state *nfs4_do_open(struct
 inode *dir,
                        exception.retry = 1;
                        continue;
                }
+               if (status == -NFS4ERR_EXPIRED) {
+                       nfs4_schedule_lease_recovery(server-
 >nfs_client);
+                       exception.retry = 1;
+                       continue;
+               }
                if (status == -EAGAIN) {
Quoted 3 lines
                        /* We must have found a delegation */
                        exception.retry = 1;
Quoted 3 lines

This looks like what I'm asking for, yes. That seems like the minimal patch that addresses the problem you're describing.

Ok, will submit later today or tomorrow.
Thanks.

Signature · 1 line
Robert Milkowski