public inbox for linux-xfs@vger.kernel.org
 help / color / mirror / Atom feed
* xfsrestore: WARNING: corrupt extent header
@ 2012-02-13  4:46 Tommy Wu
  2012-02-13 19:14 ` Bill Kendall
  2012-02-13 23:10 ` [PATCH] xfsdump: handle files with no extent headers Bill Kendall
  0 siblings, 2 replies; 4+ messages in thread
From: Tommy Wu @ 2012-02-13  4:46 UTC (permalink / raw)
  To: xfs

Hi!

   OS: Debian squeeze amd64
   xfsdump 3.0.6
   xfsprogs 3.1.7

   It happen when the xfsdump got message like:
/sbin/xfsdump: WARNING: could not open regular file ino 33905028 mode
0x00008180: Stale NFS file handle: not dumped
/sbin/xfsdump: WARNING: could not get list of non-root attributes for
nondir ino 33905028: Stale NFS file handle (116)
/sbin/xfsdump: WARNING: could not get list of root attributes for
nondir ino 33905028: Stale NFS file handle (116)
/sbin/xfsdump: WARNING: could not get list of secure attributes for
nondir ino 33905028: Stale NFS file handle (116)

   when I restore the dump file, it will get this, and it's not
restore all files in the dump file:
xfsrestore: WARNING: corrupt extent header
xfsrestore: WARNING: unable to resync media file: some portion of dump
will NOT be restored


   for example:
cmd: /sbin/xfsdump -l 0 -o -p 300 -J -F -M weekly_backup -L var -
/dev/sda8 | /bin/gzip -qv | /usr/bin/split -b 1024m --verbose -
/vol/backup/fw1/dump.20120211/var.xfsdump.gz.
/sbin/xfsdump: using file dump (drive_simple) strategy
/sbin/xfsdump: version 3.0.6 (dump format 3.0) - Running single-threaded
/sbin/xfsdump: level 0 dump of fw1.teatime.com.tw:/var
/sbin/xfsdump: dump date: Sat Feb 11 00:17:41 2012
/sbin/xfsdump: session id: 54572680-8778-49e0-b6bd-635747c47de3
/sbin/xfsdump: session label: "var"
/sbin/xfsdump: ino map phase 1: constructing initial dump list
/sbin/xfsdump: ino map phase 2: skipping (no pruning necessary)
/sbin/xfsdump: ino map phase 3: skipping (only one dump stream)
/sbin/xfsdump: ino map construction complete
/sbin/xfsdump: estimated dump size: 598793856 bytes
/sbin/xfsdump: creating dump session media file 0 (media 0, file 0)
/sbin/xfsdump: dumping ino map
/sbin/xfsdump: dumping directories
/sbin/xfsdump: dumping non-directory files
/sbin/xfsdump: WARNING: could not open regular file ino 33905028 mode
0x00008180: Stale NFS file handle: not dumped
/sbin/xfsdump: WARNING: could not get list of non-root attributes for
nondir ino 33905028: Stale NFS file handle (116)
/sbin/xfsdump: WARNING: could not get list of root attributes for
nondir ino 33905028: Stale NFS file handle (116)
/sbin/xfsdump: WARNING: could not get list of secure attributes for
nondir ino 33905028: Stale NFS file handle (116)
/sbin/xfsdump: ending media file
/sbin/xfsdump: media file size 549085392 bytes
/sbin/xfsdump: dump size (non-dir files) : 543886560 bytes
/sbin/xfsdump: dump complete: 46 seconds elapsed
/sbin/xfsdump: Dump Status: SUCCESS
 60.6%
creating file `/vol/backup/fw1/dump.20120211/var.xfsdump.gz.aa'

then test or restore from this dump file:
mail:/vol/backup/fw1/dump.20120211# cat var.xfsdump.gz.aa | gzip -dqv
| xfsrestore -v silent -p 300 -J -t - | grep ^xfsrestore:
xfsrestore: WARNING: corrupt extent header
xfsrestore: WARNING: unable to resync media file: some portion of dump
will NOT be restored


   It seems only happen in /var partition, and it's not always happen.

-- 

Tommy Wu

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

^ permalink raw reply	[flat|nested] 4+ messages in thread
* Re: xfsrestore: WARNING: corrupt extent header
  2012-02-13  4:46 xfsrestore: WARNING: corrupt extent header Tommy Wu
@ 2012-02-13 19:14 ` Bill Kendall
  2012-02-13 20:26   ` Bill Kendall
  2012-02-13 23:10 ` [PATCH] xfsdump: handle files with no extent headers Bill Kendall
  1 sibling, 1 reply; 4+ messages in thread
From: Bill Kendall @ 2012-02-13 19:14 UTC (permalink / raw)
  To: Tommy Wu; +Cc: xfs

Could you please send me the output of xfsrestore using -v 4?
It'll probably be a lot of output, so compressing and attaching
it would be best.

Thanks,
Bill

On 02/12/2012 10:46 PM, Tommy Wu wrote:
> Hi!
>
>     OS: Debian squeeze amd64
>     xfsdump 3.0.6
>     xfsprogs 3.1.7
>
>     It happen when the xfsdump got message like:
> /sbin/xfsdump: WARNING: could not open regular file ino 33905028 mode
> 0x00008180: Stale NFS file handle: not dumped
> /sbin/xfsdump: WARNING: could not get list of non-root attributes for
> nondir ino 33905028: Stale NFS file handle (116)
> /sbin/xfsdump: WARNING: could not get list of root attributes for
> nondir ino 33905028: Stale NFS file handle (116)
> /sbin/xfsdump: WARNING: could not get list of secure attributes for
> nondir ino 33905028: Stale NFS file handle (116)
>
>     when I restore the dump file, it will get this, and it's not
> restore all files in the dump file:
> xfsrestore: WARNING: corrupt extent header
> xfsrestore: WARNING: unable to resync media file: some portion of dump
> will NOT be restored
>
>
>     for example:
> cmd: /sbin/xfsdump -l 0 -o -p 300 -J -F -M weekly_backup -L var -
> /dev/sda8 | /bin/gzip -qv | /usr/bin/split -b 1024m --verbose -
> /vol/backup/fw1/dump.20120211/var.xfsdump.gz.
> /sbin/xfsdump: using file dump (drive_simple) strategy
> /sbin/xfsdump: version 3.0.6 (dump format 3.0) - Running single-threaded
> /sbin/xfsdump: level 0 dump of fw1.teatime.com.tw:/var
> /sbin/xfsdump: dump date: Sat Feb 11 00:17:41 2012
> /sbin/xfsdump: session id: 54572680-8778-49e0-b6bd-635747c47de3
> /sbin/xfsdump: session label: "var"
> /sbin/xfsdump: ino map phase 1: constructing initial dump list
> /sbin/xfsdump: ino map phase 2: skipping (no pruning necessary)
> /sbin/xfsdump: ino map phase 3: skipping (only one dump stream)
> /sbin/xfsdump: ino map construction complete
> /sbin/xfsdump: estimated dump size: 598793856 bytes
> /sbin/xfsdump: creating dump session media file 0 (media 0, file 0)
> /sbin/xfsdump: dumping ino map
> /sbin/xfsdump: dumping directories
> /sbin/xfsdump: dumping non-directory files
> /sbin/xfsdump: WARNING: could not open regular file ino 33905028 mode
> 0x00008180: Stale NFS file handle: not dumped
> /sbin/xfsdump: WARNING: could not get list of non-root attributes for
> nondir ino 33905028: Stale NFS file handle (116)
> /sbin/xfsdump: WARNING: could not get list of root attributes for
> nondir ino 33905028: Stale NFS file handle (116)
> /sbin/xfsdump: WARNING: could not get list of secure attributes for
> nondir ino 33905028: Stale NFS file handle (116)
> /sbin/xfsdump: ending media file
> /sbin/xfsdump: media file size 549085392 bytes
> /sbin/xfsdump: dump size (non-dir files) : 543886560 bytes
> /sbin/xfsdump: dump complete: 46 seconds elapsed
> /sbin/xfsdump: Dump Status: SUCCESS
>   60.6%
> creating file `/vol/backup/fw1/dump.20120211/var.xfsdump.gz.aa'
>
> then test or restore from this dump file:
> mail:/vol/backup/fw1/dump.20120211# cat var.xfsdump.gz.aa | gzip -dqv
> | xfsrestore -v silent -p 300 -J -t - | grep ^xfsrestore:
> xfsrestore: WARNING: corrupt extent header
> xfsrestore: WARNING: unable to resync media file: some portion of dump
> will NOT be restored
>
>
>     It seems only happen in /var partition, and it's not always happen.
>

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

^ permalink raw reply	[flat|nested] 4+ messages in thread
* Re: xfsrestore: WARNING: corrupt extent header
  2012-02-13 19:14 ` Bill Kendall
@ 2012-02-13 20:26   ` Bill Kendall
  0 siblings, 0 replies; 4+ messages in thread
From: Bill Kendall @ 2012-02-13 20:26 UTC (permalink / raw)
  To: Tommy Wu; +Cc: xfs

No need to send the debug output, I was able to reproduce
the problem.

Bill

On 02/13/2012 01:14 PM, Bill Kendall wrote:
> Could you please send me the output of xfsrestore using -v 4?
> It'll probably be a lot of output, so compressing and attaching
> it would be best.
>
> Thanks,
> Bill
>
> On 02/12/2012 10:46 PM, Tommy Wu wrote:
>> Hi!
>>
>> OS: Debian squeeze amd64
>> xfsdump 3.0.6
>> xfsprogs 3.1.7
>>
>> It happen when the xfsdump got message like:
>> /sbin/xfsdump: WARNING: could not open regular file ino 33905028 mode
>> 0x00008180: Stale NFS file handle: not dumped
>> /sbin/xfsdump: WARNING: could not get list of non-root attributes for
>> nondir ino 33905028: Stale NFS file handle (116)
>> /sbin/xfsdump: WARNING: could not get list of root attributes for
>> nondir ino 33905028: Stale NFS file handle (116)
>> /sbin/xfsdump: WARNING: could not get list of secure attributes for
>> nondir ino 33905028: Stale NFS file handle (116)
>>
>> when I restore the dump file, it will get this, and it's not
>> restore all files in the dump file:
>> xfsrestore: WARNING: corrupt extent header
>> xfsrestore: WARNING: unable to resync media file: some portion of dump
>> will NOT be restored
>>
>>
>> for example:
>> cmd: /sbin/xfsdump -l 0 -o -p 300 -J -F -M weekly_backup -L var -
>> /dev/sda8 | /bin/gzip -qv | /usr/bin/split -b 1024m --verbose -
>> /vol/backup/fw1/dump.20120211/var.xfsdump.gz.
>> /sbin/xfsdump: using file dump (drive_simple) strategy
>> /sbin/xfsdump: version 3.0.6 (dump format 3.0) - Running single-threaded
>> /sbin/xfsdump: level 0 dump of fw1.teatime.com.tw:/var
>> /sbin/xfsdump: dump date: Sat Feb 11 00:17:41 2012
>> /sbin/xfsdump: session id: 54572680-8778-49e0-b6bd-635747c47de3
>> /sbin/xfsdump: session label: "var"
>> /sbin/xfsdump: ino map phase 1: constructing initial dump list
>> /sbin/xfsdump: ino map phase 2: skipping (no pruning necessary)
>> /sbin/xfsdump: ino map phase 3: skipping (only one dump stream)
>> /sbin/xfsdump: ino map construction complete
>> /sbin/xfsdump: estimated dump size: 598793856 bytes
>> /sbin/xfsdump: creating dump session media file 0 (media 0, file 0)
>> /sbin/xfsdump: dumping ino map
>> /sbin/xfsdump: dumping directories
>> /sbin/xfsdump: dumping non-directory files
>> /sbin/xfsdump: WARNING: could not open regular file ino 33905028 mode
>> 0x00008180: Stale NFS file handle: not dumped
>> /sbin/xfsdump: WARNING: could not get list of non-root attributes for
>> nondir ino 33905028: Stale NFS file handle (116)
>> /sbin/xfsdump: WARNING: could not get list of root attributes for
>> nondir ino 33905028: Stale NFS file handle (116)
>> /sbin/xfsdump: WARNING: could not get list of secure attributes for
>> nondir ino 33905028: Stale NFS file handle (116)
>> /sbin/xfsdump: ending media file
>> /sbin/xfsdump: media file size 549085392 bytes
>> /sbin/xfsdump: dump size (non-dir files) : 543886560 bytes
>> /sbin/xfsdump: dump complete: 46 seconds elapsed
>> /sbin/xfsdump: Dump Status: SUCCESS
>> 60.6%
>> creating file `/vol/backup/fw1/dump.20120211/var.xfsdump.gz.aa'
>>
>> then test or restore from this dump file:
>> mail:/vol/backup/fw1/dump.20120211# cat var.xfsdump.gz.aa | gzip -dqv
>> | xfsrestore -v silent -p 300 -J -t - | grep ^xfsrestore:
>> xfsrestore: WARNING: corrupt extent header
>> xfsrestore: WARNING: unable to resync media file: some portion of dump
>> will NOT be restored
>>
>>
>> It seems only happen in /var partition, and it's not always happen.
>>
>
> _______________________________________________
> xfs mailing list
> xfs@oss.sgi.com
> http://oss.sgi.com/mailman/listinfo/xfs

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

^ permalink raw reply	[flat|nested] 4+ messages in thread
* [PATCH] xfsdump: handle files with no extent headers
  2012-02-13  4:46 xfsrestore: WARNING: corrupt extent header Tommy Wu
  2012-02-13 19:14 ` Bill Kendall
@ 2012-02-13 23:10 ` Bill Kendall
  1 sibling, 0 replies; 4+ messages in thread
From: Bill Kendall @ 2012-02-13 23:10 UTC (permalink / raw)
  To: xfs

Normally the first file header for a given file in xfsdump is followed
by one or more extent headers along with the file data. If there is a
problem opening the file, the file header will not be dumped and the
file will not be created during a restore. This is working as designed.

However if the inode indicates the file has extended attributes, a file
header will be dumped followed by an extended attribute header.  Since
this is the first file header xfsrestore sees for the file, it expects
to see an extent header and bails out complaining about an extent header
checksum error. This patch changes xfsrestore to look for the extended
attribute flag on the file header even if it's the first file header
seen for the file. The result is a zero-length file will be restored
along with the extended attributes, if they were successfully backed up.
More importantly, xfsrestore will continue on to restore the rest of the
backup.

This patch also changes xfsdump so that if it fails to open a file, it
does not try to dump the extended attributes. This prevents xfsrestore
from creating zero-length files in the situation described above.

Signed-off-by: Bill Kendall <wkendall@sgi.com>
---
 dump/content.c    |   13 ++++++++++---
 restore/content.c |   29 ++++++++++++++++++++++-------
 2 files changed, 32 insertions(+), 10 deletions(-)

diff --git a/dump/content.c b/dump/content.c
index 3a7f508..f915a8b 100644
--- a/dump/content.c
+++ b/dump/content.c
@@ -270,7 +270,8 @@ static rv_t dump_file_reg( drive_t *drivep,
 			   context_t *contextp,
 			   content_inode_hdr_t *scwhdrp,
 			   jdm_fshandle_t *,
-			   xfs_bstat_t * );
+			   xfs_bstat_t *,
+			   bool_t *);
 static rv_t dump_file_spec( drive_t *drivep,
 			    context_t *contextp,
 			    jdm_fshandle_t *,
@@ -3690,6 +3691,7 @@ dump_file( void *arg1,
 				       cwhdrp->ch_specific;
 	startpt_t *startptp = &scwhdrp->cih_startpt;
 	startpt_t *endptp = &scwhdrp->cih_endpt;
+	bool_t file_skipped = BOOL_FALSE;
 	intgen_t state;
 	rv_t rv;
 
@@ -3835,7 +3837,8 @@ dump_file( void *arg1,
 				    contextp,
 				    scwhdrp,
 				    fshandlep,
-				    statp );
+				    statp,
+				    &file_skipped );
 		if ( statp->bs_ino > contextp->cc_stat_lastino ) {
 			lock( );
 			sc_stat_nondirdone++;
@@ -3883,6 +3886,8 @@ dump_file( void *arg1,
 
 	if ( rv == RV_OK
 	     &&
+	     file_skipped == BOOL_FALSE
+	     &&
 	     sc_dumpextattrpr
 	     &&
 	     ( statp->bs_xflags & XFS_XFLAG_HASATTR )) {
@@ -3903,7 +3908,8 @@ dump_file_reg( drive_t *drivep,
 	       context_t *contextp,
 	       content_inode_hdr_t *scwhdrp,
 	       jdm_fshandle_t *fshandlep,
-	       xfs_bstat_t *statp )
+	       xfs_bstat_t *statp,
+	       bool_t *file_skippedp )
 {
 	startpt_t *startptp = &scwhdrp->cih_startpt;
 	startpt_t *endptp = &scwhdrp->cih_endpt;
@@ -3996,6 +4002,7 @@ dump_file_reg( drive_t *drivep,
 		      statp->bs_ino,
 		      statp->bs_mode,
 		      strerror( errno ));
+		*file_skippedp = BOOL_TRUE;
 		return RV_OK;
 	}
 
diff --git a/restore/content.c b/restore/content.c
index a9e0b20..a773552 100644
--- a/restore/content.c
+++ b/restore/content.c
@@ -763,6 +763,7 @@ static rv_t read_filehdr( drive_t *drivep, filehdr_t *fhdrp, bool_t fhcs );
 static rv_t restore_file( drive_t *drivep,
 			  filehdr_t *fhdrp,
 			  bool_t ehcs,
+			  bool_t ahcs,
 			  char *path1,
 			  char *path2 );
 static bool_t restore_reg( drive_t *drivep,
@@ -3425,7 +3426,7 @@ applynondirdump( drive_t *drivep,
 			strctxp->sc_path[0] = '\0';
 			strctxp->sc_fd = -1;
 
-			rv = restore_file( drivep, fhdrp, ehcs, path1, path2 );
+			rv = restore_file( drivep, fhdrp, ehcs, ahcs, path1, path2 );
 
 		} else if ( fhdrp->fh_flags & FILEHDR_FLAGS_EXTATTR ) {
 			rv = restore_extattr( drivep,
@@ -7153,6 +7154,7 @@ struct cb_context {
 	filehdr_t *cb_fhdrp;
 	rv_t cb_rv;
 	bool_t cb_ehcs;
+	bool_t cb_ahcs;
 	char *cb_path1;
 	char *cb_path2;
 };
@@ -7165,6 +7167,7 @@ static rv_t
 restore_file( drive_t *drivep,
 	      filehdr_t *fhdrp,
 	      bool_t ehcs,
+	      bool_t ahcs,
 	      char *path1,
 	      char *path2 )
 {
@@ -7180,6 +7183,7 @@ restore_file( drive_t *drivep,
 	context.cb_fhdrp = fhdrp;
 	context.cb_rv = RV_OK;
 	context.cb_ehcs = ehcs;
+	context.cb_ahcs = ahcs;
 	context.cb_path1 = path1;
 	context.cb_path2 = path2;
 	rv = tree_cb_links( bstatp->bs_ino,
@@ -7212,6 +7216,7 @@ restore_file_cb( void *cp, bool_t linkpr, char *path1, char *path2 )
 	bstat_t *bstatp = &fhdrp->fh_stat;
 	rv_t *rvp = &contextp->cb_rv;
 	bool_t ehcs = contextp->cb_ehcs;
+	bool_t ahcs = contextp->cb_ahcs;
 	stream_context_t *strctxp = (stream_context_t *)drivep->d_strmcontextp;
 
 	int rval;
@@ -7237,12 +7242,22 @@ restore_file_cb( void *cp, bool_t linkpr, char *path1, char *path2 )
 			ok = restore_reg( drivep, fhdrp, rvp, path1 );
 			if (!ok)
 				return ok;
-			ok = restore_extent_group( drivep,
-						   fhdrp,
-						   path1,
-						   strctxp->sc_fd,
-						   ehcs,
-						   rvp );
+			if ( fhdrp->fh_flags & FILEHDR_FLAGS_EXTATTR ) {
+				*rvp = restore_extattr( drivep,
+							fhdrp,
+							path1,
+							ahcs,
+							BOOL_FALSE, /* isdirpr */
+							BOOL_FALSE, /* onlydoreadpr */
+							DAH_NULL );
+			} else {
+				ok = restore_extent_group( drivep,
+							   fhdrp,
+							   path1,
+							   strctxp->sc_fd,
+							   ehcs,
+							   rvp );
+			}
 			return ok;
 		case S_IFBLK:
 		case S_IFCHR:
-- 
1.7.0.4

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

^ permalink raw reply related	[flat|nested] 4+ messages in thread
end of thread, other threads:[~2012-02-13 23:10 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-02-13  4:46 xfsrestore: WARNING: corrupt extent header Tommy Wu
2012-02-13 19:14 ` Bill Kendall
2012-02-13 20:26   ` Bill Kendall
2012-02-13 23:10 ` [PATCH] xfsdump: handle files with no extent headers Bill Kendall

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox