linux-raid.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH 0/2] Tests for add/remove of internal bitmaps
@ 2012-05-29 15:09 Jes.Sorensen
  2012-05-29 15:09 ` [PATCH 1/2] New tests testing the addition of bitmaps Jes.Sorensen
                   ` (2 more replies)
  0 siblings, 3 replies; 4+ messages in thread
From: Jes.Sorensen @ 2012-05-29 15:09 UTC (permalink / raw)
  To: neilb; +Cc: dledford, linux-raid

From: Jes Sorensen <Jes.Sorensen@redhat.com>

Hi,

Sometime ago Neil fixed a bug where the kernel would OOPS when adding
a bitmap to an existing array and using it immediately. We hit this
more recently in RHEL so I figured I better write a proper testcase
for it.

The second patch is checking that removing bitmaps also works, though
I didn't have a specific case triggered by that. I just added them for
completeness.

I added the patches to the 05 prefix as that is where the other bitmap
related test cases were. Feel free to reassign them if you like.

Cheers,
Jes
 

Jes Sorensen (2):
  New tests testing the addition of bitmaps
  New tests testing removal of internal bitmaps

 tests/05r1-add-internalbitmap        |   20 ++++++++++++++++++++
 tests/05r1-add-internalbitmap-v1a    |   20 ++++++++++++++++++++
 tests/05r1-add-internalbitmap-v1b    |   20 ++++++++++++++++++++
 tests/05r1-add-internalbitmap-v1c    |   20 ++++++++++++++++++++
 tests/05r1-remove-internalbitmap     |   18 ++++++++++++++++++
 tests/05r1-remove-internalbitmap-v1a |   18 ++++++++++++++++++
 tests/05r1-remove-internalbitmap-v1b |   18 ++++++++++++++++++
 tests/05r1-remove-internalbitmap-v1c |   18 ++++++++++++++++++
 8 files changed, 152 insertions(+)
 create mode 100644 tests/05r1-add-internalbitmap
 create mode 100644 tests/05r1-add-internalbitmap-v1a
 create mode 100644 tests/05r1-add-internalbitmap-v1b
 create mode 100644 tests/05r1-add-internalbitmap-v1c
 create mode 100644 tests/05r1-remove-internalbitmap
 create mode 100644 tests/05r1-remove-internalbitmap-v1a
 create mode 100644 tests/05r1-remove-internalbitmap-v1b
 create mode 100644 tests/05r1-remove-internalbitmap-v1c

-- 
1.7.10.2


^ permalink raw reply	[flat|nested] 4+ messages in thread

* [PATCH 1/2] New tests testing the addition of bitmaps
  2012-05-29 15:09 [PATCH 0/2] Tests for add/remove of internal bitmaps Jes.Sorensen
@ 2012-05-29 15:09 ` Jes.Sorensen
  2012-05-29 15:09 ` [PATCH 2/2] New tests testing removal of internal bitmaps Jes.Sorensen
  2012-05-30  0:47 ` [PATCH 0/2] Tests for add/remove " NeilBrown
  2 siblings, 0 replies; 4+ messages in thread
From: Jes.Sorensen @ 2012-05-29 15:09 UTC (permalink / raw)
  To: neilb; +Cc: dledford, linux-raid

From: Jes Sorensen <Jes.Sorensen@redhat.com>

This set of tests verifies that it is possible to add an internal
bitmap to an existing array, and that the device can be written to
after the bitmap is added. This should catch cases such as the one
fixed by 4474ca42e2577563a919fd3ed782e2ec55bf11a2

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
---
 tests/05r1-add-internalbitmap     |   20 ++++++++++++++++++++
 tests/05r1-add-internalbitmap-v1a |   20 ++++++++++++++++++++
 tests/05r1-add-internalbitmap-v1b |   20 ++++++++++++++++++++
 tests/05r1-add-internalbitmap-v1c |   20 ++++++++++++++++++++
 4 files changed, 80 insertions(+)
 create mode 100644 tests/05r1-add-internalbitmap
 create mode 100644 tests/05r1-add-internalbitmap-v1a
 create mode 100644 tests/05r1-add-internalbitmap-v1b
 create mode 100644 tests/05r1-add-internalbitmap-v1c

diff --git a/tests/05r1-add-internalbitmap b/tests/05r1-add-internalbitmap
new file mode 100644
index 0000000..80ce9f5
--- /dev/null
+++ b/tests/05r1-add-internalbitmap
@@ -0,0 +1,20 @@
+# 
+# create a raid1 without any bitmap, add the bitmap and then write to
+# the device. This should catch the case where the bitmap is created
+# but not reloaded correctly, such as the case fixed by
+# 4474ca42e2577563a919fd3ed782e2ec55bf11a2
+#
+mdadm --create --run $md0 --metadata=0.9 --level=1 -n2 --delay=1 $dev1 $dev2
+check wait
+check nobitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -Gb internal --bitmap-chunk=4 $md0
+check bitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -S $md0
+
+# Re-assemble the array and verify the bitmap is still present
+mdadm --assemble $md0 $dev1 $dev2
+check bitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -S $md0
diff --git a/tests/05r1-add-internalbitmap-v1a b/tests/05r1-add-internalbitmap-v1a
new file mode 100644
index 0000000..b7c80cd
--- /dev/null
+++ b/tests/05r1-add-internalbitmap-v1a
@@ -0,0 +1,20 @@
+# 
+# create a raid1 without any bitmap, add the bitmap and then write to
+# the device. This should catch the case where the bitmap is created
+# but not reloaded correctly, such as the case fixed by
+# 4474ca42e2577563a919fd3ed782e2ec55bf11a2
+#
+mdadm --create --run $md0 --metadata=1.0 --level=1 -n2 --delay=1 $dev1 $dev2
+check wait
+check nobitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -Gb internal --bitmap-chunk=4 $md0
+check bitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -S $md0
+
+# Re-assemble the array and verify the bitmap is still present
+mdadm --assemble $md0 $dev1 $dev2
+check bitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -S $md0
diff --git a/tests/05r1-add-internalbitmap-v1b b/tests/05r1-add-internalbitmap-v1b
new file mode 100644
index 0000000..b077fa7
--- /dev/null
+++ b/tests/05r1-add-internalbitmap-v1b
@@ -0,0 +1,20 @@
+# 
+# create a raid1 without any bitmap, add the bitmap and then write to
+# the device. This should catch the case where the bitmap is created
+# but not reloaded correctly, such as the case fixed by
+# 4474ca42e2577563a919fd3ed782e2ec55bf11a2
+#
+mdadm --create --run $md0 --metadata=1.1 --level=1 -n2 --delay=1 $dev1 $dev2
+check wait
+check nobitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -Gb internal --bitmap-chunk=4 $md0
+check bitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -S $md0
+
+# Re-assemble the array and verify the bitmap is still present
+mdadm --assemble $md0 $dev1 $dev2
+check bitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -S $md0
diff --git a/tests/05r1-add-internalbitmap-v1c b/tests/05r1-add-internalbitmap-v1c
new file mode 100644
index 0000000..7b39784
--- /dev/null
+++ b/tests/05r1-add-internalbitmap-v1c
@@ -0,0 +1,20 @@
+# 
+# create a raid1 without any bitmap, add the bitmap and then write to
+# the device. This should catch the case where the bitmap is created
+# but not reloaded correctly, such as the case fixed by
+# 4474ca42e2577563a919fd3ed782e2ec55bf11a2
+#
+mdadm --create --run $md0 --metadata=1.2 --level=1 -n2 --delay=1 $dev1 $dev2
+check wait
+check nobitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -Gb internal --bitmap-chunk=4 $md0
+check bitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -S $md0
+
+# Re-assemble the array and verify the bitmap is still present
+mdadm --assemble $md0 $dev1 $dev2
+check bitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -S $md0
-- 
1.7.10.2


^ permalink raw reply related	[flat|nested] 4+ messages in thread

* [PATCH 2/2] New tests testing removal of internal bitmaps
  2012-05-29 15:09 [PATCH 0/2] Tests for add/remove of internal bitmaps Jes.Sorensen
  2012-05-29 15:09 ` [PATCH 1/2] New tests testing the addition of bitmaps Jes.Sorensen
@ 2012-05-29 15:09 ` Jes.Sorensen
  2012-05-30  0:47 ` [PATCH 0/2] Tests for add/remove " NeilBrown
  2 siblings, 0 replies; 4+ messages in thread
From: Jes.Sorensen @ 2012-05-29 15:09 UTC (permalink / raw)
  To: neilb; +Cc: dledford, linux-raid

From: Jes Sorensen <Jes.Sorensen@redhat.com>

To match the add-bitmap tests, here is a set of tests checking the
removal of bitmaps.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
---
 tests/05r1-remove-internalbitmap     |   18 ++++++++++++++++++
 tests/05r1-remove-internalbitmap-v1a |   18 ++++++++++++++++++
 tests/05r1-remove-internalbitmap-v1b |   18 ++++++++++++++++++
 tests/05r1-remove-internalbitmap-v1c |   18 ++++++++++++++++++
 4 files changed, 72 insertions(+)
 create mode 100644 tests/05r1-remove-internalbitmap
 create mode 100644 tests/05r1-remove-internalbitmap-v1a
 create mode 100644 tests/05r1-remove-internalbitmap-v1b
 create mode 100644 tests/05r1-remove-internalbitmap-v1c

diff --git a/tests/05r1-remove-internalbitmap b/tests/05r1-remove-internalbitmap
new file mode 100644
index 0000000..4e7728c
--- /dev/null
+++ b/tests/05r1-remove-internalbitmap
@@ -0,0 +1,18 @@
+# 
+# create a raid1 with bitmap, remove the bitmap and verify it is still
+# gone when re-assembling the array
+#
+mdadm --create --run $md0 --metadata=0.9 --level=1 -n2 --bitmap internal --bitmap-chunk=4 --delay=1 $dev1 $dev2
+check wait
+check bitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -Gb none $md0
+check nobitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -S $md0
+
+# Re-assemble the array and verify the bitmap is still present
+mdadm --assemble $md0 $dev1 $dev2
+check nobitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -S $md0
diff --git a/tests/05r1-remove-internalbitmap-v1a b/tests/05r1-remove-internalbitmap-v1a
new file mode 100644
index 0000000..93c7730
--- /dev/null
+++ b/tests/05r1-remove-internalbitmap-v1a
@@ -0,0 +1,18 @@
+# 
+# create a raid1 with bitmap, remove the bitmap and verify it is still
+# gone when re-assembling the array
+#
+mdadm --create --run $md0 --metadata=1.0 --level=1 -n2 --bitmap internal --bitmap-chunk=4 --delay=1 $dev1 $dev2
+check wait
+check bitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -Gb none $md0
+check nobitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -S $md0
+
+# Re-assemble the array and verify the bitmap is still present
+mdadm --assemble $md0 $dev1 $dev2
+check nobitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -S $md0
diff --git a/tests/05r1-remove-internalbitmap-v1b b/tests/05r1-remove-internalbitmap-v1b
new file mode 100644
index 0000000..705c8ee
--- /dev/null
+++ b/tests/05r1-remove-internalbitmap-v1b
@@ -0,0 +1,18 @@
+# 
+# create a raid1 with bitmap, remove the bitmap and verify it is still
+# gone when re-assembling the array
+#
+mdadm --create --run $md0 --metadata=1.1 --level=1 -n2 --bitmap internal --bitmap-chunk=4 --delay=1 $dev1 $dev2
+check wait
+check bitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -Gb none $md0
+check nobitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -S $md0
+
+# Re-assemble the array and verify the bitmap is still present
+mdadm --assemble $md0 $dev1 $dev2
+check nobitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -S $md0
diff --git a/tests/05r1-remove-internalbitmap-v1c b/tests/05r1-remove-internalbitmap-v1c
new file mode 100644
index 0000000..6a14d96
--- /dev/null
+++ b/tests/05r1-remove-internalbitmap-v1c
@@ -0,0 +1,18 @@
+# 
+# create a raid1 with bitmap, remove the bitmap and verify it is still
+# gone when re-assembling the array
+#
+mdadm --create --run $md0 --metadata=1.2 --level=1 -n2 --bitmap internal --bitmap-chunk=4 --delay=1 $dev1 $dev2
+check wait
+check bitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -Gb none $md0
+check nobitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -S $md0
+
+# Re-assemble the array and verify the bitmap is still present
+mdadm --assemble $md0 $dev1 $dev2
+check nobitmap
+testdev $md0 1 $mdsize1b 64
+mdadm -S $md0
-- 
1.7.10.2


^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH 0/2] Tests for add/remove of internal bitmaps
  2012-05-29 15:09 [PATCH 0/2] Tests for add/remove of internal bitmaps Jes.Sorensen
  2012-05-29 15:09 ` [PATCH 1/2] New tests testing the addition of bitmaps Jes.Sorensen
  2012-05-29 15:09 ` [PATCH 2/2] New tests testing removal of internal bitmaps Jes.Sorensen
@ 2012-05-30  0:47 ` NeilBrown
  2 siblings, 0 replies; 4+ messages in thread
From: NeilBrown @ 2012-05-30  0:47 UTC (permalink / raw)
  To: Jes.Sorensen; +Cc: dledford, linux-raid

[-- Attachment #1: Type: text/plain, Size: 1996 bytes --]

On Tue, 29 May 2012 17:09:57 +0200 Jes.Sorensen@redhat.com wrote:

> From: Jes Sorensen <Jes.Sorensen@redhat.com>
> 
> Hi,
> 
> Sometime ago Neil fixed a bug where the kernel would OOPS when adding
> a bitmap to an existing array and using it immediately. We hit this
> more recently in RHEL so I figured I better write a proper testcase
> for it.
> 
> The second patch is checking that removing bitmaps also works, though
> I didn't have a specific case triggered by that. I just added them for
> completeness.
> 
> I added the patches to the 05 prefix as that is where the other bitmap
> related test cases were. Feel free to reassign them if you like.
> 
> Cheers,
> Jes

Thanks!  Applied (With a couple of trailing spaces removed :-)

(and the --no-error fix applied too)

NeilBrown

>  
> 
> Jes Sorensen (2):
>   New tests testing the addition of bitmaps
>   New tests testing removal of internal bitmaps
> 
>  tests/05r1-add-internalbitmap        |   20 ++++++++++++++++++++
>  tests/05r1-add-internalbitmap-v1a    |   20 ++++++++++++++++++++
>  tests/05r1-add-internalbitmap-v1b    |   20 ++++++++++++++++++++
>  tests/05r1-add-internalbitmap-v1c    |   20 ++++++++++++++++++++
>  tests/05r1-remove-internalbitmap     |   18 ++++++++++++++++++
>  tests/05r1-remove-internalbitmap-v1a |   18 ++++++++++++++++++
>  tests/05r1-remove-internalbitmap-v1b |   18 ++++++++++++++++++
>  tests/05r1-remove-internalbitmap-v1c |   18 ++++++++++++++++++
>  8 files changed, 152 insertions(+)
>  create mode 100644 tests/05r1-add-internalbitmap
>  create mode 100644 tests/05r1-add-internalbitmap-v1a
>  create mode 100644 tests/05r1-add-internalbitmap-v1b
>  create mode 100644 tests/05r1-add-internalbitmap-v1c
>  create mode 100644 tests/05r1-remove-internalbitmap
>  create mode 100644 tests/05r1-remove-internalbitmap-v1a
>  create mode 100644 tests/05r1-remove-internalbitmap-v1b
>  create mode 100644 tests/05r1-remove-internalbitmap-v1c
> 


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 828 bytes --]

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2012-05-30  0:47 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-05-29 15:09 [PATCH 0/2] Tests for add/remove of internal bitmaps Jes.Sorensen
2012-05-29 15:09 ` [PATCH 1/2] New tests testing the addition of bitmaps Jes.Sorensen
2012-05-29 15:09 ` [PATCH 2/2] New tests testing removal of internal bitmaps Jes.Sorensen
2012-05-30  0:47 ` [PATCH 0/2] Tests for add/remove " NeilBrown

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).