LVM2/test t-pvcreate-operation-md.sh

LVM2/test t-pvcreate-operation-md.sh

[mornfall]

CVSROOT:	/cvs/lvm2
Module name:	LVM2
Changes by:	mornfall at sourceware.org	2011-01-05 01:04:47

Modified files:
	test           : t-pvcreate-operation-md.sh 

Log message:
	Fix another bug in t-pvcreate-operation-md (should pass now).

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/LVM2/test/t-pvcreate-operation-md.sh.diff?cvsroot=lvm2&r1=1.12&r2=1.13

--- LVM2/test/t-pvcreate-operation-md.sh	2011/01/05 00:38:41	1.12
+++ LVM2/test/t-pvcreate-operation-md.sh	2011/01/05 01:04:47	1.13
@@ -45,13 +45,13 @@
 
 cleanup_md_and_teardown() {
     cleanup_md
-    teardown
+    aux teardown
 }
 
 # create 2 disk MD raid0 array (stripe_width=128K)
 test -b "$mddev" && exit 200
 mdadm --create --metadata=1.0 $mddev --auto=md --level 0 --raid-devices=2 --chunk 64 $dev1 $dev2
-trap 'aux cleanup_md_and_teardown' EXIT # cleanup this MD device at the end of the test
+trap 'cleanup_md_and_teardown' EXIT # cleanup this MD device at the end of the test
 test -b "$mddev" || exit 200
 
 # Test alignment of PV on MD without any MD-aware or topology-aware detection

LVM2/test t-pvcreate-operation-md.sh

[zkabelac]

CVSROOT:	/cvs/lvm2
Module name:	LVM2
Changes by:	zkabelac at sourceware.org	2011-01-28 16:12:46

Modified files:
	test           : t-pvcreate-operation-md.sh 

Log message:
	Replacei sleep wait with udev settle
	
	Use new udev_wait command instead of unpredictable sleep waiting.
	As with more devices in the system, udev processing is slower.

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/LVM2/test/t-pvcreate-operation-md.sh.diff?cvsroot=lvm2&r1=1.13&r2=1.14

--- LVM2/test/t-pvcreate-operation-md.sh	2011/01/05 01:04:47	1.13
+++ LVM2/test/t-pvcreate-operation-md.sh	2011/01/28 16:12:45	1.14
@@ -34,8 +34,9 @@
 cleanup_md() {
     # sleeps offer hack to defeat: 'md: md127 still in use'
     # see: https://bugzilla.redhat.com/show_bug.cgi?id=509908#c25
-    sleep 2
+    aux udev_wait
     mdadm --stop $mddev || true
+    aux udev_wait
     if [ -b "$mddev" ]; then
         # mdadm doesn't always cleanup the device node
 	sleep 2
@@ -102,6 +103,11 @@
     base_mddev_p=`basename $mddev_p_sysfs_name`
     mddev_p=/dev/${base_mddev_p}
 
+    # in case the system is running without devtmpfs /dev
+    # wait here for created device node on tmpfs
+    aux udev_wait $mddev_p
+    test -b $mddev_p || exit 200
+
     # Checking for 'alignment_offset' in sysfs implies Linux >= 2.6.31
     # but reliable alignment_offset support requires kernel.org Linux >= 2.6.33
     sysfs_alignment_offset=/sys/dev/block/${mddev_maj_min}/${base_mddev_p}/alignment_offset

LVM2/test t-pvcreate-operation-md.sh

[mornfall]

CVSROOT:	/cvs/lvm2
Module name:	LVM2
Changes by:	mornfall at sourceware.org	2011-01-05 00:38:41

Modified files:
	test           : t-pvcreate-operation-md.sh 

Log message:
	Fix typo in t-pvcreate-operation-md.sh.

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/LVM2/test/t-pvcreate-operation-md.sh.diff?cvsroot=lvm2&r1=1.11&r2=1.12

--- LVM2/test/t-pvcreate-operation-md.sh	2011/01/05 00:16:20	1.11
+++ LVM2/test/t-pvcreate-operation-md.sh	2011/01/05 00:38:41	1.12
@@ -20,7 +20,7 @@
 
 . lib/test
 
-prepare_lvmconf 'devices/filter = [ "a|/dev/md.*|", "a/dev\/mapper\/.*$/", "r/.*/" ]'
+aux lvmconf 'devices/filter = [ "a|/dev/md.*|", "a/dev\/mapper\/.*$/", "r/.*/" ]'
 aux prepare_devs 2
 
 # Have MD use a non-standard name to avoid colliding with an existing MD device

LVM2/test t-pvcreate-operation-md.sh

[snitzer]

CVSROOT:	/cvs/lvm2
Module name:	LVM2
Changes by:	snitzer at sourceware.org	2010-08-21 15:43:45

Modified files:
	test           : t-pvcreate-operation-md.sh 

Log message:
	Verify that pvcreate --dataalignment really does override the topology
	detected alignment.
	
	NOTE: lvm2 doesn't detect MD 1.2 metadata (now the default on RHEL6) so
	for now I'm forcing 1.0 metadata.  This was needed to be able to reuse
	the existing loop devices but recreate the md device with different
	raid0 striping.

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/LVM2/test/t-pvcreate-operation-md.sh.diff?cvsroot=lvm2&r1=1.9&r2=1.10

--- LVM2/test/t-pvcreate-operation-md.sh	2010/08/12 04:56:05	1.9
+++ LVM2/test/t-pvcreate-operation-md.sh	2010/08/21 15:43:45	1.10
@@ -41,13 +41,17 @@
 	sleep 2
 	rm -f $mddev
     fi
+}
+
+cleanup_md_and_teardown() {
+    cleanup_md
     teardown
 }
 
 # create 2 disk MD raid0 array (stripe_width=128K)
 test -b "$mddev" && exit 200
-mdadm --create $mddev --auto=md --level 0 --raid-devices=2 --chunk 64 $dev1 $dev2
-trap 'aux cleanup_md' EXIT # cleanup this MD device at the end of the test
+mdadm --create --metadata=1.0 $mddev --auto=md --level 0 --raid-devices=2 --chunk 64 $dev1 $dev2
+trap 'aux cleanup_md_and_teardown' EXIT # cleanup this MD device at the end of the test
 test -b "$mddev" || exit 200
 
 # Test alignment of PV on MD without any MD-aware or topology-aware detection
@@ -113,3 +117,27 @@
 	pvremove $mddev_p
     fi
 fi
+
+# Test newer topology-aware alignment detection w/ --dataalignment override
+if [ $linux_minor -ge 33 ]; then
+    cleanup_md
+    pvcreate -f $dev1
+    pvcreate -f $dev2
+
+    # create 2 disk MD raid0 array (stripe_width=2M)
+    test -b "$mddev" && exit 200
+    mdadm --create --metadata=1.0 $mddev --auto=md --level 0 --raid-devices=2 --chunk 1024 $dev1 $dev2
+    test -b "$mddev" || exit 200
+
+    # optimal_io_size=2097152, minimum_io_size=1048576
+    pv_align="2.00m"
+    pvcreate --metadatasize 128k \
+	--config 'devices { md_chunk_alignment=0 }' $mddev
+    check_pv_field_ $mddev pe_start $pv_align
+
+    # now verify pe_start alignment override using --dataalignment
+    pv_align="192.00k"
+    pvcreate --dataalignment 64k --metadatasize 128k \
+	--config 'devices { md_chunk_alignment=0 }' $mddev
+    check_pv_field_ $mddev pe_start $pv_align
+fi

LVM2/test t-pvcreate-operation-md.sh

[Mike Snitzer]

On Sat, Aug 21 2010 at 11:43am -0400,
snitzer at sourceware.org <snitzer@sourceware.org> wrote:

> CVSROOT:	/cvs/lvm2
> Module name:	LVM2
> Changes by:	snitzer at sourceware.org	2010-08-21 15:43:45
> 
> Modified files:
> 	test           : t-pvcreate-operation-md.sh 
> 
> Log message:
> 	Verify that pvcreate --dataalignment really does override the topology
> 	detected alignment.
> 	
> 	NOTE: lvm2 doesn't detect MD 1.2 metadata (now the default on RHEL6) so
> 	for now I'm forcing 1.0 metadata.  This was needed to be able to reuse
> 	the existing loop devices but recreate the md device with different
> 	raid0 striping.

Turns out there is nothing wrong with LVM2's MD 1.x superblock
detection (1.0, 1.1, and 1.2 are detected perfectly).

I figured out why using MD v1.0 metadata worked for this most recently
added test though:

MD's v1.0 superblock is placed at the end of the disk.  v1.1 and v1.2
are placed at the beginning of the disk.

If v1.1 or v1.2 superblock are used then the partition table of the
previous MD v1.2 device is preserved.

But when an MD v1.0 superblock is used for this new test it allows
pvcreate of the new MD v1.0 device to overwrite the partition table of
the previous v1.2 MD device.

Otherwise, if/when v1.1 or v1.2 superblock are used, for the new MD
device, you get (the failure that caused me to force the use of v1.0):

#device/dev-cache.c:329         /dev/md/md_lvm_test0: Added to device cache
#device/dev-io.c:487         Opened /dev/md/md_lvm_test0 RO
#device/dev-io.c:260       /dev/md/md_lvm_test0: size is 131072 sectors
#device/dev-io.c:134         /dev/md/md_lvm_test0: block size is 4096 bytes
#filters/filter.c:167         /dev/md/md_lvm_test0: Skipping: Partition table signature found
#device/dev-io.c:533         Closed /dev/md/md_lvm_test0


> --- LVM2/test/t-pvcreate-operation-md.sh	2010/08/12 04:56:05	1.9
> +++ LVM2/test/t-pvcreate-operation-md.sh	2010/08/21 15:43:45	1.10
> @@ -41,13 +41,17 @@
>  	sleep 2
>  	rm -f $mddev
>      fi
> +}
> +
> +cleanup_md_and_teardown() {
> +    cleanup_md
>      teardown
>  }
>  
>  # create 2 disk MD raid0 array (stripe_width=128K)
>  test -b "$mddev" && exit 200
> -mdadm --create $mddev --auto=md --level 0 --raid-devices=2 --chunk 64 $dev1 $dev2
> -trap 'aux cleanup_md' EXIT # cleanup this MD device at the end of the test
> +mdadm --create --metadata=1.0 $mddev --auto=md --level 0 --raid-devices=2 --chunk 64 $dev1 $dev2
> +trap 'aux cleanup_md_and_teardown' EXIT # cleanup this MD device at the end of the test
>  test -b "$mddev" || exit 200
>  
>  # Test alignment of PV on MD without any MD-aware or topology-aware detection
> @@ -113,3 +117,27 @@
>  	pvremove $mddev_p
>      fi
>  fi
> +
> +# Test newer topology-aware alignment detection w/ --dataalignment override
> +if [ $linux_minor -ge 33 ]; then
> +    cleanup_md
> +    pvcreate -f $dev1
> +    pvcreate -f $dev2
> +
> +    # create 2 disk MD raid0 array (stripe_width=2M)
> +    test -b "$mddev" && exit 200
> +    mdadm --create --metadata=1.0 $mddev --auto=md --level 0 --raid-devices=2 --chunk 1024 $dev1 $dev2
> +    test -b "$mddev" || exit 200
> +
> +    # optimal_io_size=2097152, minimum_io_size=1048576
> +    pv_align="2.00m"
> +    pvcreate --metadatasize 128k \
> +	--config 'devices { md_chunk_alignment=0 }' $mddev
> +    check_pv_field_ $mddev pe_start $pv_align
> +
> +    # now verify pe_start alignment override using --dataalignment
> +    pv_align="192.00k"
> +    pvcreate --dataalignment 64k --metadatasize 128k \
> +	--config 'devices { md_chunk_alignment=0 }' $mddev
> +    check_pv_field_ $mddev pe_start $pv_align
> +fi

LVM2/test t-pvcreate-operation-md.sh

[snitzer]

CVSROOT:	/cvs/lvm2
Module name:	LVM2
Changes by:	snitzer at sourceware.org	2010-08-12 04:56:06

Modified files:
	test           : t-pvcreate-operation-md.sh 

Log message:
	fix t-pvcreate-operation-md.sh to require kernel.org Linux >= 2.6.33 for
	the final alignment_offset check.  In the future, might look to check
	for the RHEL6 kernel too.

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/LVM2/test/t-pvcreate-operation-md.sh.diff?cvsroot=lvm2&r1=1.8&r2=1.9

--- LVM2/test/t-pvcreate-operation-md.sh	2010/08/12 04:11:49	1.8
+++ LVM2/test/t-pvcreate-operation-md.sh	2010/08/12 04:56:05	1.9
@@ -99,14 +99,17 @@
     mddev_p=/dev/${base_mddev_p}
 
     # Checking for 'alignment_offset' in sysfs implies Linux >= 2.6.31
+    # but reliable alignment_offset support requires kernel.org Linux >= 2.6.33
     sysfs_alignment_offset=/sys/dev/block/${mddev_maj_min}/${base_mddev_p}/alignment_offset
     [ -f $sysfs_alignment_offset -a $linux_minor -ge 33 ] && \
 	alignment_offset=`cat $sysfs_alignment_offset` || \
 	alignment_offset=0
 
-    # default alignment is 1M, add alignment_offset
-    pv_align=$((1048576+$alignment_offset))B
-    pvcreate --metadatasize 128k $mddev_p
-    check_pv_field_ $mddev_p pe_start $pv_align "--units b"
-    pvremove $mddev_p
+    if [ $alignment_offset -gt 0 ]; then    
+        # default alignment is 1M, add alignment_offset
+	pv_align=$((1048576+$alignment_offset))B
+	pvcreate --metadatasize 128k $mddev_p
+	check_pv_field_ $mddev_p pe_start $pv_align "--units b"
+	pvremove $mddev_p
+    fi
 fi

LVM2/test t-pvcreate-operation-md.sh

[mornfall]

CVSROOT:	/cvs/lvm2
Module name:	LVM2
Changes by:	mornfall at sourceware.org	2010-07-21 14:12:48

Modified files:
	test           : t-pvcreate-operation-md.sh 

Log message:
	Don't fail t-pvcreate-operation-md if mdadm is broken.

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/LVM2/test/t-pvcreate-operation-md.sh.diff?cvsroot=lvm2&r1=1.6&r2=1.7

--- LVM2/test/t-pvcreate-operation-md.sh	2010/05/24 15:28:45	1.6
+++ LVM2/test/t-pvcreate-operation-md.sh	2010/07/21 14:12:47	1.7
@@ -35,7 +35,7 @@
     # sleeps offer hack to defeat: 'md: md127 still in use'
     # see: https://bugzilla.redhat.com/show_bug.cgi?id=509908#c25
     sleep 2
-    mdadm --stop $mddev
+    mdadm --stop $mddev || true
     if [ -b "$mddev" ]; then
         # mdadm doesn't always cleanup the device node
 	sleep 2

LVM2/test t-pvcreate-operation-md.sh

[mornfall]

CVSROOT:	/cvs/lvm2
Module name:	LVM2
Changes by:	mornfall at sourceware.org	2010-05-24 15:28:45

Modified files:
	test           : t-pvcreate-operation-md.sh 

Log message:
	Skip t-pvcreate-operation-md if mdadm fails to create the required device.

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/LVM2/test/t-pvcreate-operation-md.sh.diff?cvsroot=lvm2&r1=1.5&r2=1.6

--- LVM2/test/t-pvcreate-operation-md.sh	2010/04/13 21:42:44	1.5
+++ LVM2/test/t-pvcreate-operation-md.sh	2010/05/24 15:28:45	1.6
@@ -45,9 +45,10 @@
 }
 
 # create 2 disk MD raid0 array (stripe_width=128K)
-[ -b "$mddev" ] && exit 200
+test -b "$mddev" && exit 200
 mdadm --create $mddev --auto=md --level 0 --raid-devices=2 --chunk 64 $dev1 $dev2
 trap 'aux cleanup_md' EXIT # cleanup this MD device at the end of the test
+test -b "$mddev" || exit 200
 
 # Test alignment of PV on MD without any MD-aware or topology-aware detection
 # - should treat $mddev just like any other block device

LVM2/test t-pvcreate-operation-md.sh

[wysochanski]

CVSROOT:	/cvs/lvm2
Module name:	LVM2
Changes by:	wysochanski at sourceware.org	2010-04-13 21:42:45

Modified files:
	test           : t-pvcreate-operation-md.sh 

Log message:
	Fix teardown in t-pvcreate-operation-md.sh nightly test.

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/LVM2/test/t-pvcreate-operation-md.sh.diff?cvsroot=lvm2&r1=1.4&r2=1.5

--- LVM2/test/t-pvcreate-operation-md.sh	2010/01/19 15:59:34	1.4
+++ LVM2/test/t-pvcreate-operation-md.sh	2010/04/13 21:42:44	1.5
@@ -41,7 +41,7 @@
 	sleep 2
 	rm -f $mddev
     fi
-    teardown_
+    teardown
 }
 
 # create 2 disk MD raid0 array (stripe_width=128K)

LVM2/test t-pvcreate-operation-md.sh

[snitzer]

CVSROOT:	/cvs/lvm2
Module name:	LVM2
Changes by:	snitzer at sourceware.org	2010-01-19 15:59:34

Modified files:
	test           : t-pvcreate-operation-md.sh 

Log message:
	update test/t-pvcreate-operation-md.sh attempt loading raid0.ko if raid0
	isn't already available (in /proc/mdstat).
	
	switch to requiring 2.6.33 for the alignment_offset tests; 2.6.{31,32}
	alignment_offset values aren't reliable.  2.6.33 _should_ have mkp's
	alignment_offset fixes but so far it doesn't (as of 2.6.33-rc4).

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/LVM2/test/t-pvcreate-operation-md.sh.diff?cvsroot=lvm2&r1=1.3&r2=1.4

--- LVM2/test/t-pvcreate-operation-md.sh	2009/09/30 16:13:53	1.3
+++ LVM2/test/t-pvcreate-operation-md.sh	2010/01/19 15:59:34	1.4
@@ -14,7 +14,9 @@
 which perl || exit 200
 which awk || exit 200
 which cut || exit 200
-test -f /proc/mdstat || exit 200
+
+test -f /proc/mdstat && grep -q raid0 /proc/mdstat || \
+modprobe raid0 || exit 200
 
 . ./test-utils.sh
 
@@ -66,7 +68,8 @@
 linux_minor=$(echo `uname -r` | cut -d'.' -f3 | cut -d'-' -f1)
 
 # Test newer topology-aware alignment detection
-if [ $linux_minor -gt 31 ]; then
+# - first added to 2.6.31 but not "reliable" until 2.6.33
+if [ $linux_minor -ge 33 ]; then
     pv_align="256.00k"
     pvcreate --metadatasize 128k \
 	--config 'devices { md_chunk_alignment=0 }' $mddev
@@ -74,7 +77,7 @@
 fi
 
 # partition MD array directly, depends on blkext in Linux >= 2.6.28
-if [ $linux_minor -gt 27 ]; then
+if [ $linux_minor -ge 28 ]; then
     # create one partition
     sfdisk $mddev <<EOF
 ,,83
@@ -95,7 +98,7 @@
 
     # Checking for 'alignment_offset' in sysfs implies Linux >= 2.6.31
     sysfs_alignment_offset=/sys/dev/block/${mddev_maj_min}/${base_mddev_p}/alignment_offset
-    [ -f $sysfs_alignment_offset ] && \
+    [ -f $sysfs_alignment_offset -a $linux_minor -ge 33 ] && \
 	alignment_offset=`cat $sysfs_alignment_offset` || \
 	alignment_offset=0
 

LVM2/test t-pvcreate-operation-md.sh

[mbroz]

CVSROOT:	/cvs/lvm2
Module name:	LVM2
Changes by:	mbroz at sourceware.org	2009-09-13 19:28:00

Modified files:
	test           : t-pvcreate-operation-md.sh 

Log message:
	Do not run mdadm tests if MD module is not loaded.

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/LVM2/test/t-pvcreate-operation-md.sh.diff?cvsroot=lvm2&r1=1.1&r2=1.2

--- LVM2/test/t-pvcreate-operation-md.sh	2009/08/04 16:02:39	1.1
+++ LVM2/test/t-pvcreate-operation-md.sh	2009/09/13 19:28:00	1.2
@@ -14,6 +14,7 @@
 which perl || exit 200
 which awk || exit 200
 which cut || exit 200
+test -f /proc/mdstat || exit 200
 
 . ./test-utils.sh
 

Re: LVM2/test t-pvcreate-operation-md.sh

[Mike Snitzer]

On Sun, Sep 13 2009 at  3:28pm -0400,
mbroz at sourceware.org <mbroz@sourceware.org> wrote:

> CVSROOT:	/cvs/lvm2
> Module name:	LVM2
> Changes by:	mbroz at sourceware.org	2009-09-13 19:28:00
> 
> Modified files:
> 	test           : t-pvcreate-operation-md.sh 
> 
> Log message:
> 	Do not run mdadm tests if MD module is not loaded.
> 
> Patches:
> http://sourceware.org/cgi-bin/cvsweb.cgi/LVM2/test/t-pvcreate-operation-md.sh.diff?cvsroot=lvm2&r1=1.1&r2=1.2
> 
> --- LVM2/test/t-pvcreate-operation-md.sh	2009/08/04 16:02:39	1.1
> +++ LVM2/test/t-pvcreate-operation-md.sh	2009/09/13 19:28:00	1.2
> @@ -14,6 +14,7 @@
>  which perl || exit 200
>  which awk || exit 200
>  which cut || exit 200
> +test -f /proc/mdstat || exit 200
>  
>  . ./test-utils.sh
>  
> 

We might try loading the raid0 module if it isn't loaded (grep for raid0
in /proc/mdstat)?  And if that fails skip this test?

Re: LVM2/test t-pvcreate-operation-md.sh

[Milan Broz]

Mike Snitzer wrote:
> We might try loading the raid0 module if it isn't loaded (grep for raid0
> in /proc/mdstat)?  And if that fails skip this test?

I have intentionally all modules unloaded (and do not want sometimes run md tests),
raid0 is loaded automatically, if md_mod is loaded (then /proc/mdstat exists).

If md_mod is not there, test fails. For most systems it is loaded in initrd anyway.

That was just quick hack, I think there is problem in mdadm, because it created
some device node even if md module is not present - but I have no time to debug it now.

Milan

LVM2/test t-pvcreate-operation-md.sh

[Mike Snitzer]

On Mon, Sep 14 2009 at  1:11pm -0400,
Milan Broz <mbroz@redhat.com> wrote:

> Mike Snitzer wrote:
> > We might try loading the raid0 module if it isn't loaded (grep for raid0
> > in /proc/mdstat)?  And if that fails skip this test?
> 
> I have intentionally all modules unloaded (and do not want sometimes run md tests),
> raid0 is loaded automatically, if md_mod is loaded (then /proc/mdstat exists).

md-mod doesn't provide the raid0 personality; raid0 does.  I've changed
the test to attempt loading raid0 if it isn't loaded.
 
> If md_mod is not there, test fails. For most systems it is loaded in initrd anyway.
> 
> That was just quick hack, I think there is problem in mdadm, because it created
> some device node even if md module is not present - but I have no time to debug it now.

I'd be interested to know if you still have problems with the test.  If
so we should get to the bottom of it (maybe require a minimum mdadm
version?).

t-pvcreate-operation-md.sh does provide needed testing of LVM2's MD
support so it shouldn't be skipped.

Mike