[PATCH 0/3] mdadm: re-add journal to array with bad journal

[PATCH 0/3] mdadm: re-add journal to array with bad journal

[Song Liu]

These patches try to recreate missing/faulty journal in mdadm. Since
the array is readonly with journal missing, it is necessary to
restart the array after recreating journal.

Song Liu (3):
  add sysfs_array_state to struct mdinfo
  [mdadm] recreate journal in mdadm
  make .spec file work

 Assemble.c                  |  8 +++++++-
 Manage.c                    | 32 +++++++++++++++++++++++++++++++-
 ReadMe.c                    |  3 ++-
 mdadm.c                     |  8 ++++++++
 mdadm.h                     |  5 +++++
 mdadm.spec                  |  8 ++++----
 super1.c                    |  3 ++-
 sysfs.c                     |  7 +++++++
 udev-md-raid-assembly.rules |  2 +-
 9 files changed, 67 insertions(+), 9 deletions(-)

--
2.4.6

[PATCH 1/3] add sysfs_array_state to struct mdinfo

[Song Liu]

Add sysfs_array_state to struct mdinfo, and add GET_ARRAY_STATE to
options of sysfs_read.

Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Shaohua Li <shli@fb.com>
---
 mdadm.h | 4 ++++
 sysfs.c | 7 +++++++
 2 files changed, 11 insertions(+)

diff --git a/mdadm.h b/mdadm.h
index 5d5e97f..1b027bb 100644
--- a/mdadm.h
+++ b/mdadm.h
@@ -280,6 +280,9 @@ struct mdinfo {
 	#define	DS_REMOVE	1024
 	#define	DS_UNBLOCK	2048
 	int prev_state, curr_state, next_state;
+
+	/* info read from sysfs */
+	char		sysfs_array_state[20];
 };
 
 struct createinfo {
@@ -563,6 +566,7 @@ enum sysfs_read_flags {
 	GET_SIZE	= (1 << 22),
 	GET_STATE	= (1 << 23),
 	GET_ERROR	= (1 << 24),
+	GET_ARRAY_STATE = (1 << 25),
 };
 
 /* If fd >= 0, get the array it is open on,
diff --git a/sysfs.c b/sysfs.c
index f1fd610..2600343 100644
--- a/sysfs.c
+++ b/sysfs.c
@@ -230,6 +230,13 @@ struct mdinfo *sysfs_read(int fd, char *devnm, unsigned long options)
 			goto abort;
 	}
 
+	if (options & GET_ARRAY_STATE) {
+		strcpy(base, "array_state");
+		if (load_sys(fname, sra->sysfs_array_state))
+			goto abort;
+	} else
+		sra->sysfs_array_state[0] = 0;
+
 	if (! (options & GET_DEVS))
 		return sra;
 
-- 
2.4.6

[PATCH 2/3] [mdadm] recreate journal in mdadm

[Song Liu]

Example work flow:

./mdadm --fail /dev/md1 /dev/sdb2
mdadm: set /dev/sdb2 faulty in /dev/md1

./mdadm --stop /dev/md1
mdadm: stopped /dev/md1

./mdadm -A --scan --force
mdadm: Journal is missing or stale, starting array read only.
mdadm: /dev/md/1 has been started with 15 drives.

./mdadm --add-journal /dev/md1 /dev/sdb2
mdadm: Journal created on /dev/sdb2.
mdadm: Please restart /dev/md1.

./mdadm --stop /dev/md1
mdadm: stopped /dev/md1

./mdadm -A --scan
mdadm: clearing FAULTY flag for new journal in /dev/md/1 for /dev/sdb2
mdadm: Marking array /dev/md/1 as 'clean'
mdadm: /dev/md/1 has been started with 15 drives and 1 journal.

Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Shaohua Li <shli@fb.com>
---
 Assemble.c |  8 +++++++-
 Manage.c   | 32 +++++++++++++++++++++++++++++++-
 ReadMe.c   |  1 +
 mdadm.c    |  8 ++++++++
 mdadm.h    |  1 +
 super1.c   |  3 ++-
 6 files changed, 50 insertions(+), 3 deletions(-)

diff --git a/Assemble.c b/Assemble.c
index 90c9918..6b9a6da 100644
--- a/Assemble.c
+++ b/Assemble.c
@@ -1688,7 +1688,13 @@ try_again:

 		if (st->ss->update_super(st, &devices[j].i, "assemble", NULL,
 					 c->verbose, 0, NULL)) {
-			if (c->force) {
+			if (devices[j].i.disk.state & (1<<MD_DISK_JOURNAL)) {
+				if (c->verbose >= 0)
+					pr_err("clearing FAULTY flag for new journal in %s for %s\n",
+					       mddev, devices[j].devname);
+				change = 1;
+				content->journal_clean = 1;
+			} else if (c->force) {
 				if (c->verbose >= 0)
 					pr_err("clearing FAULTY flag for device %d in %s for %s\n",
 					       j, mddev, devices[j].devname);
diff --git a/Manage.c b/Manage.c
index 2df303d..930865a 100644
--- a/Manage.c
+++ b/Manage.c
@@ -825,7 +825,8 @@ int Manage_add(int fd, int tfd, struct mddev_dev *dv,
 		}

 		/* Make sure device is large enough */
-		if (tst->sb &&
+		if (dv->disposition != 'j' &&  /* skip size check for Journal */
+		    tst->sb &&
 		    tst->ss->avail_size(tst, ldsize/512, INVALID_SECTORS) <
 		    array_size) {
 			if (dv->disposition == 'M')
@@ -929,8 +930,30 @@ int Manage_add(int fd, int tfd, struct mddev_dev *dv,
 	else
 		disc.number = raid_slot;
 	disc.state = 0;
+
+	/* only add journal to array that supports journaling */
+	if (dv->disposition == 'j') {
+		struct mdinfo mdi;
+		struct mdinfo *mdp;
+
+		mdp = sysfs_read(fd, NULL, GET_ARRAY_STATE);
+
+		if (strncmp(mdp->sysfs_array_state, "readonly", 8) != 0) {
+			pr_err("%s is not readonly, cannot add journal.\n", devname);
+			return -1;
+		}
+
+		tst->ss->getinfo_super(tst, &mdi, NULL);
+		if (mdi.journal_device_required == 0) {
+			pr_err("%s does not support journal device.\n", devname);
+			return -1;
+		}
+	}
+
 	if (array->not_persistent==0) {
 		int dfd;
+		if (dv->disposition == 'j')
+			disc.state |= 1 << MD_DISK_JOURNAL;
 		if (dv->writemostly == 1)
 			disc.state |= 1 << MD_DISK_WRITEMOSTLY;
 		dfd = dev_open(dv->devname, O_RDWR | O_EXCL|O_DIRECT);
@@ -1038,6 +1061,11 @@ int Manage_add(int fd, int tfd, struct mddev_dev *dv,
 		ping_monitor(devnm);
 		sysfs_free(sra);
 		close(container_fd);
+	} else if (dv->disposition == 'j') {
+		/* do not call ioctl(ADD_NEW_DISK) for journal */
+		pr_err("Journal created on %s.\n", dv->devname);
+		pr_err("Please restart %s.\n", devname);
+		return 1;
 	} else {
 		tst->ss->free_super(tst);
 		if (ioctl(fd, ADD_NEW_DISK, &disc)) {
@@ -1277,6 +1305,7 @@ int Manage_subdevs(char *devname, int fd,
 	 *	   try HOT_ADD_DISK
 	 *         If that fails EINVAL, try ADD_NEW_DISK
 	 *  'S' - add the device as a spare - don't try re-add
+	 *  'j' - add the device as a journal device
 	 *  'A' - re-add the device
 	 *  'r' - remove the device: HOT_REMOVE_DISK
 	 *        device can be 'faulty' or 'detached' in which case all
@@ -1509,6 +1538,7 @@ int Manage_subdevs(char *devname, int fd,
 			goto abort;
 		case 'a':
 		case 'S': /* --add-spare */
+		case 'j': /* --add-journal */
 		case 'A':
 		case 'M': /* --re-add missing */
 		case 'F': /* --re-add faulty  */
diff --git a/ReadMe.c b/ReadMe.c
index fb5a671..566fc8d 100644
--- a/ReadMe.c
+++ b/ReadMe.c
@@ -157,6 +157,7 @@ struct option long_options[] = {
     /* Management */
     {"add",       0, 0, Add},
     {"add-spare", 0, 0, AddSpare},
+    {"add-journal", 0, 0, AddJournal},
     {"remove",    0, 0, Remove},
     {"fail",      0, 0, Fail},
     {"set-faulty",0, 0, Fail},
diff --git a/mdadm.c b/mdadm.c
index f56a8cf..feec3b7 100644
--- a/mdadm.c
+++ b/mdadm.c
@@ -190,6 +190,7 @@ int main(int argc, char *argv[])
 		case 'a':
 		case Add:
 		case AddSpare:
+		case AddJournal:
 		case 'r':
 		case Remove:
 		case Replace:
@@ -925,6 +926,13 @@ int main(int argc, char *argv[])
 		case O(MANAGE,AddSpare): /* add drive - never re-add */
 			devmode = 'S';
 			continue;
+		case O(MANAGE,AddJournal): /* add journal */
+			if (s.journaldisks && (s.level < 4 || s.level > 6)) {
+				pr_err("--add-journal is only supported for RAID level 4/5/6.\n");
+				exit(2);
+			}
+			devmode = 'j';
+			continue;
 		case O(MANAGE,ReAdd):
 			devmode = 'A';
 			continue;
diff --git a/mdadm.h b/mdadm.h
index 1b027bb..c16bcd7 100644
--- a/mdadm.h
+++ b/mdadm.h
@@ -345,6 +345,7 @@ enum special_options {
 	ManageOpt,
 	Add,
 	AddSpare,
+	AddJournal,
 	Remove,
 	Fail,
 	Replace,
diff --git a/super1.c b/super1.c
index 1735c2d..893f9bf 100644
--- a/super1.c
+++ b/super1.c
@@ -1713,7 +1713,8 @@ static int write_init_super1(struct supertype *st)
 		if (rfd >= 0)
 			close(rfd);

-		sb->events = 0;
+		if (!(di->disk.state & (1<<MD_DISK_JOURNAL)))
+			sb->events = 0;

 		refst = dup_super(st);
 		if (load_super1(refst, di->fd, NULL)==0) {
--
2.4.6

Re: [PATCH 0/3] mdadm: re-add journal to array with bad journal

[NeilBrown]

[-- Attachment #1: Type: text/plain, Size: 1292 bytes --]

On Wed, Dec 02 2015, Song Liu <songliubraving@fb.com> wrote:

> These patches try to recreate missing/faulty journal in mdadm. Since
> the array is readonly with journal missing, it is necessary to
> restart the array after recreating journal.
>
> Song Liu (3):
>   add sysfs_array_state to struct mdinfo
>   [mdadm] recreate journal in mdadm
>   make .spec file work

I didn't get that third patch???

The patches generally look good, but it bothers me that you have to stop
and restart the array for the added journal to work.  Can we just teach
the kernel to accept a new journal on an active array, and then use that
functionality in mdadm?
That is certainly what I want in the longer term and I'm not sure it
makes sense to have a half-way "--add-journal" that doesn't do the whole
job.

Thanks,
NeilBrown


>
>  Assemble.c                  |  8 +++++++-
>  Manage.c                    | 32 +++++++++++++++++++++++++++++++-
>  ReadMe.c                    |  3 ++-
>  mdadm.c                     |  8 ++++++++
>  mdadm.h                     |  5 +++++
>  mdadm.spec                  |  8 ++++----
>  super1.c                    |  3 ++-
>  sysfs.c                     |  7 +++++++
>  udev-md-raid-assembly.rules |  2 +-
>  9 files changed, 67 insertions(+), 9 deletions(-)
>
> --
> 2.4.6

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 818 bytes --]

RE: [PATCH 0/3] mdadm: re-add journal to array with bad journal

[Song Liu]

> -----Original Message-----
> From: NeilBrown [mailto:neilb@suse.com]
> Sent: Thursday, December 10, 2015 2:06 PM
> To: Song Liu <songliubraving@fb.com>; linux-raid@vger.kernel.org
> Cc: dan.j.williams@intel.com; Shaohua Li <shli@fb.com>; hch@infradead.org;
> Kernel Team <Kernel-team@fb.com>; Song Liu <songliubraving@fb.com>
> Subject: Re: [PATCH 0/3] mdadm: re-add journal to array with bad journal
> 
> On Wed, Dec 02 2015, Song Liu <songliubraving@fb.com> wrote:
> 
> > These patches try to recreate missing/faulty journal in mdadm. Since
> > the array is readonly with journal missing, it is necessary to restart
> > the array after recreating journal.
> >
> > Song Liu (3):
> >   add sysfs_array_state to struct mdinfo
> >   [mdadm] recreate journal in mdadm
> >   make .spec file work
> 
> I didn't get that third patch???

That's my local hack to .spec file. Didn't mean to send it. 

> 
> The patches generally look good, but it bothers me that you have to stop and
> restart the array for the added journal to work.  Can we just teach the kernel to
> accept a new journal on an active array, and then use that functionality in
> mdadm?
> That is certainly what I want in the longer term and I'm not sure it makes sense
> to have a half-way "--add-journal" that doesn't do the whole job.

We will look into hot_add option.  

Thanks,
Song