From: andros@netapp.com
To: trond.myklebust@netapp.com
Cc: linux-nfs@vger.kernel.org, Andy Adamson <andros@netapp.com>
Subject: [PATCH Version 2 05/12] NFSv4.1: mark deviceid invalid on filelayout DS connection errors
Date: Wed, 21 Mar 2012 15:46:17 -0400 [thread overview]
Message-ID: <1332359184-1887-6-git-send-email-andros@netapp.com> (raw)
In-Reply-To: <1332359184-1887-1-git-send-email-andros@netapp.com>
From: Andy Adamson <andros@netapp.com>
This prevents the use of any layout for i/o that references the deviceid.
I/O is redirected through the MDS.
Redirect the unhandled failed I/O to the MDS without marking either the
layout or the deviceid invalid.
Signed-off-by: Andy Adamson <andros@netapp.com>
---
fs/nfs/nfs4filelayout.c | 65 ++++++++++++++++++++++++++++++++++------------
fs/nfs/nfs4filelayout.h | 6 ++++
2 files changed, 54 insertions(+), 17 deletions(-)
diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/nfs4filelayout.c
index 3802937..1f1be26 100644
--- a/fs/nfs/nfs4filelayout.c
+++ b/fs/nfs/nfs4filelayout.c
@@ -116,7 +116,7 @@ void filelayout_reset_read(struct rpc_task *task, struct nfs_read_data *data)
static int filelayout_async_handle_error(struct rpc_task *task,
struct nfs4_state *state,
struct nfs_client *clp,
- int *reset)
+ unsigned long *reset)
{
struct nfs_server *mds_server = NFS_SERVER(state->inode);
struct nfs_client *mds_client = mds_server->nfs_client;
@@ -158,10 +158,23 @@ static int filelayout_async_handle_error(struct rpc_task *task,
break;
case -NFS4ERR_RETRY_UNCACHED_REP:
break;
+ /* RPC connection errors */
+ case -ECONNREFUSED:
+ case -EHOSTDOWN:
+ case -EHOSTUNREACH:
+ case -ENETUNREACH:
+ case -EIO:
+ case -ETIMEDOUT:
+ case -EPIPE:
+ dprintk("%s DS connection error. Retry through MDS %d\n",
+ __func__, task->tk_status);
+ set_bit(NFS4_RESET_DEVICEID, reset);
+ set_bit(NFS4_RESET_TO_MDS, reset);
+ break;
default:
- dprintk("%s DS error. Retry through MDS %d\n", __func__,
- task->tk_status);
- *reset = 1;
+ dprintk("%s Unhandled DS error. Retry through MDS %d\n",
+ __func__, task->tk_status);
+ set_bit(NFS4_RESET_TO_MDS, reset);
break;
}
out:
@@ -179,16 +192,22 @@ wait_on_recovery:
static int filelayout_read_done_cb(struct rpc_task *task,
struct nfs_read_data *data)
{
- int reset = 0;
+ struct nfs4_deviceid_node *devid = FILELAYOUT_DEVID_NODE(data->lseg);
+ unsigned long reset = 0;
dprintk("%s DS read\n", __func__);
if (filelayout_async_handle_error(task, data->args.context->state,
data->ds_clp, &reset) == -EAGAIN) {
- dprintk("%s calling restart ds_clp %p ds_clp->cl_session %p\n",
- __func__, data->ds_clp, data->ds_clp->cl_session);
- if (reset)
+
+ dprintk("%s reset 0x%lx ds_clp %p session %p\n", __func__,
+ reset, data->ds_clp, data->ds_clp->cl_session);
+
+ if (test_bit(NFS4_RESET_TO_MDS, &reset)) {
filelayout_reset_read(task, data);
+ if (test_bit(NFS4_RESET_DEVICEID, &reset))
+ filelayout_mark_devid_invalid(devid);
+ }
rpc_restart_call_prepare(task);
return -EAGAIN;
}
@@ -260,14 +279,20 @@ static void filelayout_read_release(void *data)
static int filelayout_write_done_cb(struct rpc_task *task,
struct nfs_write_data *data)
{
- int reset = 0;
+ struct nfs4_deviceid_node *devid = FILELAYOUT_DEVID_NODE(data->lseg);
+ unsigned long reset = 0;
if (filelayout_async_handle_error(task, data->args.context->state,
data->ds_clp, &reset) == -EAGAIN) {
- dprintk("%s calling restart ds_clp %p ds_clp->cl_session %p\n",
- __func__, data->ds_clp, data->ds_clp->cl_session);
- if (reset)
+
+ dprintk("%s reset 0x%lx ds_clp %p session %p\n", __func__,
+ reset, data->ds_clp, data->ds_clp->cl_session);
+
+ if (test_bit(NFS4_RESET_TO_MDS, &reset)) {
filelayout_reset_write(task, data);
+ if (test_bit(NFS4_RESET_DEVICEID, &reset))
+ filelayout_mark_devid_invalid(devid);
+ }
rpc_restart_call_prepare(task);
return -EAGAIN;
}
@@ -290,16 +315,22 @@ static void prepare_to_resend_writes(struct nfs_write_data *data)
static int filelayout_commit_done_cb(struct rpc_task *task,
struct nfs_write_data *data)
{
- int reset = 0;
+ struct nfs4_deviceid_node *devid = FILELAYOUT_DEVID_NODE(data->lseg);
+ unsigned long reset = 0;
if (filelayout_async_handle_error(task, data->args.context->state,
data->ds_clp, &reset) == -EAGAIN) {
- dprintk("%s calling restart ds_clp %p ds_clp->cl_session %p\n",
- __func__, data->ds_clp, data->ds_clp->cl_session);
- if (reset)
+
+ dprintk("%s reset 0x%lx ds_clp %p session %p\n", __func__,
+ reset, data->ds_clp, data->ds_clp->cl_session);
+
+ if (test_bit(NFS4_RESET_TO_MDS, &reset)) {
prepare_to_resend_writes(data);
- else
+ if (test_bit(NFS4_RESET_DEVICEID, &reset))
+ filelayout_mark_devid_invalid(devid);
+ } else {
rpc_restart_call_prepare(task);
+ }
return -EAGAIN;
}
diff --git a/fs/nfs/nfs4filelayout.h b/fs/nfs/nfs4filelayout.h
index b54b389..08b667a 100644
--- a/fs/nfs/nfs4filelayout.h
+++ b/fs/nfs/nfs4filelayout.h
@@ -41,6 +41,12 @@
#define NFS4_PNFS_MAX_STRIPE_CNT 4096
#define NFS4_PNFS_MAX_MULTI_CNT 256 /* 256 fit into a u8 stripe_index */
+/* internal use */
+enum nfs4_fl_reset_state {
+ NFS4_RESET_TO_MDS = 0,
+ NFS4_RESET_DEVICEID,
+};
+
enum stripetype4 {
STRIPE_SPARSE = 1,
STRIPE_DENSE = 2
--
1.7.6.4
next prev parent reply other threads:[~2012-03-21 19:46 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-03-21 19:46 [PATCH Version 2 00/12] NFSv4.1 file layout data server quick failover andros
2012-03-21 19:46 ` [PATCH Version 2 01/12] NFSv4.1 move nfs4_reset_read and nfs_reset_write andros
2012-03-21 19:46 ` [PATCH Version 2 02/12] NFSv4.1: cleanup filelayout invalid deviceid handling andros
2012-03-21 19:46 ` [PATCH Version 2 03/12] NFSv4.1 cleanup filelayout invalid layout handling andros
2012-03-21 19:46 ` [PATCH Version 2 04/12] NFSv4.1 set RPC_TASK_SOFTCONN for filelayout DS RPC calls andros
2012-03-21 19:46 ` andros [this message]
2012-03-21 20:39 ` [PATCH Version 2 05/12] NFSv4.1: mark deviceid invalid on filelayout DS connection errors Myklebust, Trond
2012-03-22 13:23 ` Adamson, Andy
2012-03-22 13:44 ` Myklebust, Trond
2012-03-22 13:53 ` Andy Adamson
2012-03-21 19:46 ` [PATCH Version 2 06/12] NFSv4.1: send filelayout DS commits to the MDS on invalid deviceid andros
2012-03-21 19:46 ` [PATCH Version 2 07/12] NFSv4.1 Check invalid deviceid upon slot table waitq wakeup andros
2012-03-21 19:46 ` [PATCH Version 2 08/12] NFSv4.1 wake up all tasks on un-connected DS slot table waitq andros
2012-03-21 19:46 ` [PATCH Version 2 09/12] NFSv4.1 ref count nfs_client across filelayout data server io andros
2012-03-21 19:46 ` [PATCH Version 2 10/12] NFSv4.1 de reference a disconnected data server client record andros
2012-03-21 19:46 ` [PATCH Version 2 11/12] NFSv4.1 check for NULL pnfs_layout_hdr in pnfs scan commit lists andros
2012-03-21 19:46 ` [PATCH Version 2 12/12] NFSv4.1 have filelayout_initiate_commit return void andros
2012-03-21 20:47 ` Myklebust, Trond
2012-03-25 14:59 ` Fred Isaman
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=1332359184-1887-6-git-send-email-andros@netapp.com \
--to=andros@netapp.com \
--cc=linux-nfs@vger.kernel.org \
--cc=trond.myklebust@netapp.com \
/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