diff mbox

[4/5] ocfs2/dlm: solve a BUG when deref failed in dlm_drop_lockres_ref

Message ID 579a73bd.aWzQ0Rkjfu9OwqME%akpm@linux-foundation.org (mailing list archive)
State New, archived
Headers show

Commit Message

Andrew Morton July 28, 2016, 9:06 p.m. UTC
From: piaojun <piaojun@huawei.com>
Subject: ocfs2/dlm: solve a BUG when deref failed in dlm_drop_lockres_ref

We found a BUG situation that lockres is migrated during deref described
below.  To solve the BUG, we could purge lockres directly when other node
says I did not have a ref.  Additionally, we'd better purge lockres if
master goes down, as no one will response deref done.

Node 1                  Node 2(old master)             Node3(new master)
dlm_purge_lockres
send deref to N2

                        leave domain
                        migrate lockres to N3
                                                       finish migration
                                                       send do assert
                                                       master to N1

receive do assert msg
form N3, but can not
find lockres because
DROPPING_REF is set,
so the owner is still
N2.

                        receive deref from N1
                        and response -EINVAL
                        because lockres is migrated

BUG when receive -EINVAL
in dlm_drop_lockres_ref

Fixes: 842b90b62461d ("ocfs2/dlm: return in progress if master can not clear the refmap bit right now")

Link: http://lkml.kernel.org/r/57845103.3070406@huawei.com
Signed-off-by: Jun Piao <piaojun@huawei.com>
Reviewed-by: Joseph Qi <joseph.qi@huawei.com>
Reviewed-by: Jiufei Xue <xuejiufei@huawei.com>
Cc: Mark Fasheh <mfasheh@suse.de>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 fs/ocfs2/dlm/dlmmaster.c |    9 ++++++---
 fs/ocfs2/dlm/dlmthread.c |   13 +++++++++++--
 2 files changed, 17 insertions(+), 5 deletions(-)

Comments

Mark Fasheh July 28, 2016, 10:12 p.m. UTC | #1
On Thu, Jul 28, 2016 at 02:06:05PM -0700, Andrew Morton wrote:
> From: piaojun <piaojun@huawei.com>
> Subject: ocfs2/dlm: solve a BUG when deref failed in dlm_drop_lockres_ref
> 
> We found a BUG situation that lockres is migrated during deref described
> below.  To solve the BUG, we could purge lockres directly when other node
> says I did not have a ref.  Additionally, we'd better purge lockres if
> master goes down, as no one will response deref done.
> 
> Node 1                  Node 2(old master)             Node3(new master)
> dlm_purge_lockres
> send deref to N2
> 
>                         leave domain
>                         migrate lockres to N3
>                                                        finish migration
>                                                        send do assert
>                                                        master to N1
> 
> receive do assert msg
> form N3, but can not
> find lockres because
> DROPPING_REF is set,
> so the owner is still
> N2.
> 
>                         receive deref from N1
>                         and response -EINVAL
>                         because lockres is migrated
> 
> BUG when receive -EINVAL
> in dlm_drop_lockres_ref
> 
> Fixes: 842b90b62461d ("ocfs2/dlm: return in progress if master can not clear the refmap bit right now")
> 
> Link: http://lkml.kernel.org/r/57845103.3070406@huawei.com
> Signed-off-by: Jun Piao <piaojun@huawei.com>
> Reviewed-by: Joseph Qi <joseph.qi@huawei.com>
> Reviewed-by: Jiufei Xue <xuejiufei@huawei.com>

Reviewed-by: Mark Fasheh <mfasheh@suse.de>

The only thing is I wonder if those ML_NOTICE messages in this patch and
the previous one will cause unnecessary end-user concern.

The fixes though look good, thanks for those.
	--Mark


> Cc: Mark Fasheh <mfasheh@suse.de>
> Cc: Joel Becker <jlbec@evilplan.org>
> Cc: Junxiao Bi <junxiao.bi@oracle.com>
> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
> ---
> 
>  fs/ocfs2/dlm/dlmmaster.c |    9 ++++++---
>  fs/ocfs2/dlm/dlmthread.c |   13 +++++++++++--
>  2 files changed, 17 insertions(+), 5 deletions(-)
> 
> diff -puN fs/ocfs2/dlm/dlmmaster.c~ocfs2-dlm-solve-a-bug-when-deref-failed-in-dlm_drop_lockres_ref fs/ocfs2/dlm/dlmmaster.c
> --- a/fs/ocfs2/dlm/dlmmaster.c~ocfs2-dlm-solve-a-bug-when-deref-failed-in-dlm_drop_lockres_ref
> +++ a/fs/ocfs2/dlm/dlmmaster.c
> @@ -2276,9 +2276,12 @@ int dlm_drop_lockres_ref(struct dlm_ctxt
>  		mlog(ML_ERROR, "%s: res %.*s, DEREF to node %u got %d\n",
>  		     dlm->name, namelen, lockname, res->owner, r);
>  		dlm_print_one_lock_resource(res);
> -		BUG();
> -	}
> -	return ret ? ret : r;
> +		if (r == -ENOMEM)
> +			BUG();
> +	} else
> +		ret = r;
> +
> +	return ret;
>  }
>  
>  int dlm_deref_lockres_handler(struct o2net_msg *msg, u32 len, void *data,
> diff -puN fs/ocfs2/dlm/dlmthread.c~ocfs2-dlm-solve-a-bug-when-deref-failed-in-dlm_drop_lockres_ref fs/ocfs2/dlm/dlmthread.c
> --- a/fs/ocfs2/dlm/dlmthread.c~ocfs2-dlm-solve-a-bug-when-deref-failed-in-dlm_drop_lockres_ref
> +++ a/fs/ocfs2/dlm/dlmthread.c
> @@ -175,6 +175,15 @@ static void dlm_purge_lockres(struct dlm
>  	     res->lockname.len, res->lockname.name, master);
>  
>  	if (!master) {
> +		if (res->state & DLM_LOCK_RES_DROPPING_REF) {
> +			mlog(ML_NOTICE, "%s: res %.*s already in "
> +				"DLM_LOCK_RES_DROPPING_REF state\n",
> +				dlm->name, res->lockname.len,
> +				res->lockname.name);
> +			spin_unlock(&res->spinlock);
> +			return;
> +		}
> +
>  		res->state |= DLM_LOCK_RES_DROPPING_REF;
>  		/* drop spinlock...  retake below */
>  		spin_unlock(&res->spinlock);
> @@ -203,8 +212,8 @@ static void dlm_purge_lockres(struct dlm
>  		dlm->purge_count--;
>  	}
>  
> -	if (!master && ret != 0) {
> -		mlog(0, "%s: deref %.*s in progress or master goes down\n",
> +	if (!master && ret == DLM_DEREF_RESPONSE_INPROG) {
> +		mlog(0, "%s: deref %.*s in progress\n",
>  			dlm->name, res->lockname.len, res->lockname.name);
>  		spin_unlock(&res->spinlock);
>  		return;
> _
--
Mark Fasheh
piaojun July 29, 2016, 3:08 a.m. UTC | #2
Hello Mark,

On 2016-7-29 6:12, Mark Fasheh wrote:
> On Thu, Jul 28, 2016 at 02:06:05PM -0700, Andrew Morton wrote:
>> From: piaojun <piaojun@huawei.com>
>> Subject: ocfs2/dlm: solve a BUG when deref failed in dlm_drop_lockres_ref
>>
>> We found a BUG situation that lockres is migrated during deref described
>> below.  To solve the BUG, we could purge lockres directly when other node
>> says I did not have a ref.  Additionally, we'd better purge lockres if
>> master goes down, as no one will response deref done.
>>
>> Node 1                  Node 2(old master)             Node3(new master)
>> dlm_purge_lockres
>> send deref to N2
>>
>>                         leave domain
>>                         migrate lockres to N3
>>                                                        finish migration
>>                                                        send do assert
>>                                                        master to N1
>>
>> receive do assert msg
>> form N3, but can not
>> find lockres because
>> DROPPING_REF is set,
>> so the owner is still
>> N2.
>>
>>                         receive deref from N1
>>                         and response -EINVAL
>>                         because lockres is migrated
>>
>> BUG when receive -EINVAL
>> in dlm_drop_lockres_ref
>>
>> Fixes: 842b90b62461d ("ocfs2/dlm: return in progress if master can not clear the refmap bit right now")
>>
>> Link: http://lkml.kernel.org/r/57845103.3070406@huawei.com
>> Signed-off-by: Jun Piao <piaojun@huawei.com>
>> Reviewed-by: Joseph Qi <joseph.qi@huawei.com>
>> Reviewed-by: Jiufei Xue <xuejiufei@huawei.com>
> 
> Reviewed-by: Mark Fasheh <mfasheh@suse.de>
> 
> The only thing is I wonder if those ML_NOTICE messages in this patch and
> the previous one will cause unnecessary end-user concern.
> 
> The fixes though look good, thanks for those.
> 	--Mark
> 
> 
Those ML_NOTICE log just server as reminders for developer, I think
end-user usually care about ML_NOTICE log.

Thanks
Jun
>> Cc: Mark Fasheh <mfasheh@suse.de>
>> Cc: Joel Becker <jlbec@evilplan.org>
>> Cc: Junxiao Bi <junxiao.bi@oracle.com>
>> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
>> ---
>>
>>  fs/ocfs2/dlm/dlmmaster.c |    9 ++++++---
>>  fs/ocfs2/dlm/dlmthread.c |   13 +++++++++++--
>>  2 files changed, 17 insertions(+), 5 deletions(-)
>>
>> diff -puN fs/ocfs2/dlm/dlmmaster.c~ocfs2-dlm-solve-a-bug-when-deref-failed-in-dlm_drop_lockres_ref fs/ocfs2/dlm/dlmmaster.c
>> --- a/fs/ocfs2/dlm/dlmmaster.c~ocfs2-dlm-solve-a-bug-when-deref-failed-in-dlm_drop_lockres_ref
>> +++ a/fs/ocfs2/dlm/dlmmaster.c
>> @@ -2276,9 +2276,12 @@ int dlm_drop_lockres_ref(struct dlm_ctxt
>>  		mlog(ML_ERROR, "%s: res %.*s, DEREF to node %u got %d\n",
>>  		     dlm->name, namelen, lockname, res->owner, r);
>>  		dlm_print_one_lock_resource(res);
>> -		BUG();
>> -	}
>> -	return ret ? ret : r;
>> +		if (r == -ENOMEM)
>> +			BUG();
>> +	} else
>> +		ret = r;
>> +
>> +	return ret;
>>  }
>>  
>>  int dlm_deref_lockres_handler(struct o2net_msg *msg, u32 len, void *data,
>> diff -puN fs/ocfs2/dlm/dlmthread.c~ocfs2-dlm-solve-a-bug-when-deref-failed-in-dlm_drop_lockres_ref fs/ocfs2/dlm/dlmthread.c
>> --- a/fs/ocfs2/dlm/dlmthread.c~ocfs2-dlm-solve-a-bug-when-deref-failed-in-dlm_drop_lockres_ref
>> +++ a/fs/ocfs2/dlm/dlmthread.c
>> @@ -175,6 +175,15 @@ static void dlm_purge_lockres(struct dlm
>>  	     res->lockname.len, res->lockname.name, master);
>>  
>>  	if (!master) {
>> +		if (res->state & DLM_LOCK_RES_DROPPING_REF) {
>> +			mlog(ML_NOTICE, "%s: res %.*s already in "
>> +				"DLM_LOCK_RES_DROPPING_REF state\n",
>> +				dlm->name, res->lockname.len,
>> +				res->lockname.name);
>> +			spin_unlock(&res->spinlock);
>> +			return;
>> +		}
>> +
>>  		res->state |= DLM_LOCK_RES_DROPPING_REF;
>>  		/* drop spinlock...  retake below */
>>  		spin_unlock(&res->spinlock);
>> @@ -203,8 +212,8 @@ static void dlm_purge_lockres(struct dlm
>>  		dlm->purge_count--;
>>  	}
>>  
>> -	if (!master && ret != 0) {
>> -		mlog(0, "%s: deref %.*s in progress or master goes down\n",
>> +	if (!master && ret == DLM_DEREF_RESPONSE_INPROG) {
>> +		mlog(0, "%s: deref %.*s in progress\n",
>>  			dlm->name, res->lockname.len, res->lockname.name);
>>  		spin_unlock(&res->spinlock);
>>  		return;
>> _
> --
> Mark Fasheh
> 
> .
>
Mark Fasheh July 29, 2016, 5:47 p.m. UTC | #3
On Fri, Jul 29, 2016 at 11:08:50AM +0800, piaojun wrote:
> Hello Mark,
> 
> On 2016-7-29 6:12, Mark Fasheh wrote:
> > On Thu, Jul 28, 2016 at 02:06:05PM -0700, Andrew Morton wrote:
> >> From: piaojun <piaojun@huawei.com>
> >> Subject: ocfs2/dlm: solve a BUG when deref failed in dlm_drop_lockres_ref
> >>
> >> We found a BUG situation that lockres is migrated during deref described
> >> below.  To solve the BUG, we could purge lockres directly when other node
> >> says I did not have a ref.  Additionally, we'd better purge lockres if
> >> master goes down, as no one will response deref done.
> >>
> >> Node 1                  Node 2(old master)             Node3(new master)
> >> dlm_purge_lockres
> >> send deref to N2
> >>
> >>                         leave domain
> >>                         migrate lockres to N3
> >>                                                        finish migration
> >>                                                        send do assert
> >>                                                        master to N1
> >>
> >> receive do assert msg
> >> form N3, but can not
> >> find lockres because
> >> DROPPING_REF is set,
> >> so the owner is still
> >> N2.
> >>
> >>                         receive deref from N1
> >>                         and response -EINVAL
> >>                         because lockres is migrated
> >>
> >> BUG when receive -EINVAL
> >> in dlm_drop_lockres_ref
> >>
> >> Fixes: 842b90b62461d ("ocfs2/dlm: return in progress if master can not clear the refmap bit right now")
> >>
> >> Link: http://lkml.kernel.org/r/57845103.3070406@huawei.com
> >> Signed-off-by: Jun Piao <piaojun@huawei.com>
> >> Reviewed-by: Joseph Qi <joseph.qi@huawei.com>
> >> Reviewed-by: Jiufei Xue <xuejiufei@huawei.com>
> > 
> > Reviewed-by: Mark Fasheh <mfasheh@suse.de>
> > 
> > The only thing is I wonder if those ML_NOTICE messages in this patch and
> > the previous one will cause unnecessary end-user concern.
> > 
> > The fixes though look good, thanks for those.
> > 	--Mark
> > 
> > 
> Those ML_NOTICE log just server as reminders for developer, I think
> end-user usually care about ML_NOTICE log.

Ok, I had different experiences but it's not a big deal one way or the
other. If it helps you guys track what's going on then it's probably worth
it :)
	--Mark

--
Mark Fasheh
diff mbox

Patch

diff -puN fs/ocfs2/dlm/dlmmaster.c~ocfs2-dlm-solve-a-bug-when-deref-failed-in-dlm_drop_lockres_ref fs/ocfs2/dlm/dlmmaster.c
--- a/fs/ocfs2/dlm/dlmmaster.c~ocfs2-dlm-solve-a-bug-when-deref-failed-in-dlm_drop_lockres_ref
+++ a/fs/ocfs2/dlm/dlmmaster.c
@@ -2276,9 +2276,12 @@  int dlm_drop_lockres_ref(struct dlm_ctxt
 		mlog(ML_ERROR, "%s: res %.*s, DEREF to node %u got %d\n",
 		     dlm->name, namelen, lockname, res->owner, r);
 		dlm_print_one_lock_resource(res);
-		BUG();
-	}
-	return ret ? ret : r;
+		if (r == -ENOMEM)
+			BUG();
+	} else
+		ret = r;
+
+	return ret;
 }
 
 int dlm_deref_lockres_handler(struct o2net_msg *msg, u32 len, void *data,
diff -puN fs/ocfs2/dlm/dlmthread.c~ocfs2-dlm-solve-a-bug-when-deref-failed-in-dlm_drop_lockres_ref fs/ocfs2/dlm/dlmthread.c
--- a/fs/ocfs2/dlm/dlmthread.c~ocfs2-dlm-solve-a-bug-when-deref-failed-in-dlm_drop_lockres_ref
+++ a/fs/ocfs2/dlm/dlmthread.c
@@ -175,6 +175,15 @@  static void dlm_purge_lockres(struct dlm
 	     res->lockname.len, res->lockname.name, master);
 
 	if (!master) {
+		if (res->state & DLM_LOCK_RES_DROPPING_REF) {
+			mlog(ML_NOTICE, "%s: res %.*s already in "
+				"DLM_LOCK_RES_DROPPING_REF state\n",
+				dlm->name, res->lockname.len,
+				res->lockname.name);
+			spin_unlock(&res->spinlock);
+			return;
+		}
+
 		res->state |= DLM_LOCK_RES_DROPPING_REF;
 		/* drop spinlock...  retake below */
 		spin_unlock(&res->spinlock);
@@ -203,8 +212,8 @@  static void dlm_purge_lockres(struct dlm
 		dlm->purge_count--;
 	}
 
-	if (!master && ret != 0) {
-		mlog(0, "%s: deref %.*s in progress or master goes down\n",
+	if (!master && ret == DLM_DEREF_RESPONSE_INPROG) {
+		mlog(0, "%s: deref %.*s in progress\n",
 			dlm->name, res->lockname.len, res->lockname.name);
 		spin_unlock(&res->spinlock);
 		return;