"nfs4_discover_server_trunking unhandled error" on Ctrl-C

DISCUSSION Mkrtchyan, Tigran 4 participants 5 messages Jun 13, 2017 – Jun 14, 2017

Subsystems (2)

NFS client NFS server

lore

Share & tools

lore

Mkrtchyan, Tigran

author

2017-06-13 13:32 UTC #0 lore raw

If I hit Ctrl-C during mount, then I see following message in the log file:

NFS: nfs4_discover_server_trunking unhandled error -512. Exiting with error EIO

produced at nfs4state.c:2533. Is this expected behavior?

Tigran.

Olga Kornievskaia

2017-06-13 16:21 UTC #1 lore raw

Quoted 7 lines · Tigran

If I hit Ctrl-C during mount, then I see following message in the log file:
NFS: nfs4_discover_server_trunking unhandled error -512. Exiting with error EIO
produced at nfs4state.c:2533. Is this expected behavior?

What kernel version (or upstream commit) is that? Is this reproducible? Anything on the network trace that server returns that could explain an error it doesn't expect. Also I'm thinking that nfs_wait_client_init_complete() could be interrupted by ctrl-c and return ERESTARTSYS and that could translate to "unhanded error".

Mkrtchyan, Tigran

author

2017-06-13 19:22 UTC #2 lore raw

Hi Olga,

this is 4.12.-rc4. I will check packet capture tomorrow, however, I am pretty sure that this is client internal interrupt processing.

Tigran.

----- Original Message -----

Quoted 25 lines

From: "Olga Kornievskaia" <[email protected]>
To: "Tigran Mkrtchyan" <[email protected]>
Cc: "linux-nfs" <[email protected]>, "Andy Adamson" <[email protected]>
Sent: Tuesday, June 13, 2017 6:21:01 PM
Subject: Re: "nfs4_discover_server_trunking unhandled error" on Ctrl-C
Quoted 7 lines · Tigran
If I hit Ctrl-C during mount, then I see following message in the log file:
NFS: nfs4_discover_server_trunking unhandled error -512. Exiting with error EIO
produced at nfs4state.c:2533. Is this expected behavior?
What kernel version (or upstream commit) is that? Is this reproducible? Anything on the network trace that server returns that could explain an error it doesn't expect. Also I'm thinking that nfs_wait_client_init_complete() could be interrupted by ctrl-c and return ERESTARTSYS and that could translate to "unhanded error". --
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Jeff Layton

2017-06-14 12:29 UTC #3 lore raw

That warning has been around for a long time. We're running the server trunking check and caught a signal in the middle of it. It's harmless, AFAICT.

Should we merge a patch like this (untested) one, just to silence it? It really is an expected situation that we needn't warn anyone about:

fs/nfs/nfs4state.c +3 -0

index b34de036501b..ae9a9ad2894d 100644
@@ -2182,6 +2182,9 @@ int nfs4_discover_server_trunking(struct nfs_client *clp,
                                 * in nfs4_exchange_id */
                status = -EKEYEXPIRED;
                break;
+       case -ERESTARTSYS:
+               status = -EINTR;
+               break;
        default:
                pr_warn("NFS: %s unhandled error %d. Exiting with error EIO\n",
                                __func__, status);

Quoted 37 lines · Tigran

Hi Olga,
this is 4.12.-rc4. I will check packet capture tomorrow, however, I am pretty sure that this is client internal interrupt processing.
Tigran.
----- Original Message -----
Quoted 15 lines
From: "Olga Kornievskaia" <[email protected]>
To: "Tigran Mkrtchyan" <[email protected]>
Cc: "linux-nfs" <[email protected]>, "Andy Adamson" <[email protected]>
Sent: Tuesday, June 13, 2017 6:21:01 PM
Subject: Re: "nfs4_discover_server_trunking unhandled error" on Ctrl-C
Quoted 7 lines · Tigran
If I hit Ctrl-C during mount, then I see following message in the log file:
NFS: nfs4_discover_server_trunking unhandled error -512. Exiting with error EIO
produced at nfs4state.c:2533. Is this expected behavior?
Quoted 9 lines
What kernel version (or upstream commit) is that? Is this reproducible? Anything on the network trace that server returns that could explain an error it doesn't expect. Also I'm thinking that nfs_wait_client_init_complete() could be interrupted by ctrl-c and return ERESTARTSYS and that could translate to "unhanded error". --
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Signature · 1 line

Jeff Layton <[email protected]>

Chuck Lever

2017-06-14 14:10 UTC #4 lore raw

Quoted 15 lines · Jeff Layton

That warning has been around for a long time. We're running the server trunking check and caught a signal in the middle of it. It's harmless, AFAICT.
Should we merge a patch like this (untested) one, just to silence it? It really is an expected situation that we needn't warn anyone about:
fs/nfs/nfs4state.c +0 -0
index b34de036501b..ae9a9ad2894d 100644

fs/nfs/nfs4state.c int nfs4_discover_server_trunking(struct nfs_client *clp, · line 2190

Quoted 9 lines

fs/nfs/nfs4state.c +3 -0

@@ -2182,6 +2182,9 @@ int nfs4_discover_server_trunking(struct nfs_client *clp,
                                * in nfs4_exchange_id */
               status = -EKEYEXPIRED;
               break;
+       case -ERESTARTSYS:
+               status = -EINTR;
+               break;
       default:
               pr_warn("NFS: %s unhandled error %d. Exiting with error EIO\n",
                               __func__, status);

I agree that this message is noise in this case, and the patch looks sane.

Quoted 34 lines · Tigran

Quoted 33 lines
Hi Olga,
this is 4.12.-rc4. I will check packet capture tomorrow, however, I am pretty sure that this is client internal interrupt processing.
Tigran.
----- Original Message -----
Quoted 24 lines
From: "Olga Kornievskaia" <[email protected]>
To: "Tigran Mkrtchyan" <[email protected]>
Cc: "linux-nfs" <[email protected]>, "Andy Adamson" <[email protected]>
Sent: Tuesday, June 13, 2017 6:21:01 PM
Subject: Re: "nfs4_discover_server_trunking unhandled error" on Ctrl-C
Quoted 7 lines · Tigran
If I hit Ctrl-C during mount, then I see following message in the log file:
NFS: nfs4_discover_server_trunking unhandled error -512. Exiting with error EIO
produced at nfs4state.c:2533. Is this expected behavior?
What kernel version (or upstream commit) is that? Is this reproducible? Anything on the network trace that server returns that could explain an error it doesn't expect. Also I'm thinking that nfs_wait_client_init_complete() could be interrupted by ctrl-c and return ERESTARTSYS and that could translate to "unhanded error". --
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Quoted 11 lines

Quoted 4 lines
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Signature · 5 lines
Jeff Layton <[email protected]>
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

--
Chuck Lever