From: Jeff Layton <jlayton@redhat.com>
To: bfields@fieldses.org
Cc: linux-nfs@vger.kernel.org
Subject: [PATCH 02/11] nfsd: change heuristic for selecting the client_tracking_ops
Date: Mon, 12 Nov 2012 15:00:49 -0500 [thread overview]
Message-ID: <1352750458-16857-3-git-send-email-jlayton@redhat.com> (raw)
In-Reply-To: <1352750458-16857-1-git-send-email-jlayton@redhat.com>
First, try to use the new usermodehelper upcall. It should succeed or
fail quickly, so there's little cost to doing so.
If it fails, and the legacy tracking dir exists, use that. If it
doesn't exist then fall back to using nfsdcld.
Signed-off-by: Jeff Layton <jlayton@redhat.com>
---
fs/nfsd/nfs4recover.c | 36 +++++++++++++++++++++++++++---------
1 file changed, 27 insertions(+), 9 deletions(-)
diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c
index 2fc2f6c..e71f713 100644
--- a/fs/nfsd/nfs4recover.c
+++ b/fs/nfsd/nfs4recover.c
@@ -1064,17 +1064,35 @@ nfsd4_client_tracking_init(struct net *net)
int status;
struct path path;
- if (!client_tracking_ops) {
- client_tracking_ops = &nfsd4_cld_tracking_ops;
- status = kern_path(nfs4_recoverydir(), LOOKUP_FOLLOW, &path);
- if (!status) {
- if (S_ISDIR(path.dentry->d_inode->i_mode))
- client_tracking_ops =
- &nfsd4_legacy_tracking_ops;
- path_put(&path);
- }
+ /* just run the init if it the method is already decided */
+ if (client_tracking_ops)
+ goto do_init;
+
+ /*
+ * First, try a UMH upcall. It should succeed or fail quickly, so
+ * there's little harm in trying that first.
+ */
+ client_tracking_ops = &nfsd4_umh_tracking_ops;
+ status = client_tracking_ops->init(net);
+ if (!status)
+ return status;
+
+ /*
+ * See if the recoverydir exists and is a directory. If it is,
+ * then use the legacy ops.
+ */
+ client_tracking_ops = &nfsd4_legacy_tracking_ops;
+ status = kern_path(nfs4_recoverydir(), LOOKUP_FOLLOW, &path);
+ if (!status) {
+ status = S_ISDIR(path.dentry->d_inode->i_mode);
+ path_put(&path);
+ if (status)
+ goto do_init;
}
+ /* Finally, try to use nfsdcld */
+ client_tracking_ops = &nfsd4_cld_tracking_ops;
+do_init:
status = client_tracking_ops->init(net);
if (status) {
printk(KERN_WARNING "NFSD: Unable to initialize client "
--
1.7.11.7
next prev parent reply other threads:[~2012-11-12 20:01 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-11-12 20:00 [PATCH 00/11] nfsd: add nfsdcltrack support and limit the use of md5 hashes in nfsdv4 code Jeff Layton
2012-11-12 20:00 ` [PATCH 01/11] nfsd: add a usermodehelper upcall for NFSv4 client ID tracking Jeff Layton
2012-11-12 20:00 ` Jeff Layton [this message]
2012-11-12 20:00 ` [PATCH 03/11] nfsd: pass info about the legacy recoverydir in environment variables Jeff Layton
2012-11-12 20:00 ` [PATCH 04/11] nfsd: warn about impending removal of nfsdcld upcall Jeff Layton
2012-11-12 20:00 ` [PATCH 05/11] nfsd: have nfsd4_find_reclaim_client take a char * argument Jeff Layton
2012-11-12 20:00 ` [PATCH 06/11] nfsd: break out reclaim record removal into separate function Jeff Layton
2012-11-12 20:00 ` [PATCH 07/11] nfsd: make nfs4_client_to_reclaim return a pointer to the reclaim record Jeff Layton
2012-11-12 20:00 ` [PATCH 08/11] nfsd: don't search for client by hash on legacy reboot recovery gracedone Jeff Layton
2012-11-12 20:00 ` [PATCH 09/11] nfsd: move the confirmed and unconfirmed hlists to a rbtree Jeff Layton
2012-11-12 20:00 ` [PATCH 10/11] nfsd: get rid of cl_recdir field Jeff Layton
2012-11-12 20:00 ` [PATCH 11/11] nfsd: release the legacy reclaimable clients list in grace_done Jeff Layton
2012-11-12 23:57 ` [PATCH 00/11] nfsd: add nfsdcltrack support and limit the use of md5 hashes in nfsdv4 code J. Bruce Fields
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=1352750458-16857-3-git-send-email-jlayton@redhat.com \
--to=jlayton@redhat.com \
--cc=bfields@fieldses.org \
--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).