Libguestfs - Aug 2014 - [PATCH 0/3] libguestfs: improve OpenSUSE support

Hi,

this series for libguestfs improves the support for OpenSUSE.

There are couple of changes (in testing-only stuff) needed to help
running the tests properly, and an update of the packagelist.


Pino Toscano (3):
  builder: do not use xz --block-size for the test images
  tests: make the scratch disk used for scratch much larger
  appliance: initial packagelist for OpenSUSE

 appliance/packagelist.in | 21 +++++++++++++++++++++
 builder/Makefile.am      |  4 ++--
 generator/actions.ml     |  2 +-
 tests/c-api/tests-main.c |  2 +-
 4 files changed, 25 insertions(+), 4 deletions(-)

-- 
1.9.3

Do not use the --block-size parameter of xz when compressing the test
images, as that option does not exist in xz < 5.1.
---
 builder/Makefile.am | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/builder/Makefile.am b/builder/Makefile.am
index 6a5cb01..726ca60 100644
--- a/builder/Makefile.am
+++ b/builder/Makefile.am
@@ -224,12 +224,12 @@ fedora.qcow2: ../tests/guests/fedora.img
 
 fedora.qcow2.xz: fedora.qcow2
 	rm -f $@ $@-t
-	xz --best --block-size=16777216 -c $< > $@-t
+	xz --best -c $< > $@-t
 	mv $@-t $@
 
 %.xz: ../tests/guests/%.img
 	rm -f $@ $@-t
-	xz --best --block-size=16777216 -c $< > $@-t
+	xz --best -c $< > $@-t
 	mv $@-t $@
 
 TESTS = \
-- 
1.9.3

50M can be not enough when uploading various copies of unstripped
static binaries from the system, which can fill up such a small disk.
---
 generator/actions.ml     | 2 +-
 tests/c-api/tests-main.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/generator/actions.ml b/generator/actions.ml
index 9570d9b..245da6f 100644
--- a/generator/actions.ml
+++ b/generator/actions.ml
@@ -9607,7 +9607,7 @@ device is stopped, but it is not destroyed or
zeroed." };
           "check_hash (ret, \"PART_ENTRY_NUMBER\",
\"1\") == 0 && "^
           "check_hash (ret, \"PART_ENTRY_TYPE\",
\"0x83\") == 0 && "^
           "check_hash (ret, \"PART_ENTRY_OFFSET\",
\"128\") == 0 && "^
-          "check_hash (ret, \"PART_ENTRY_SIZE\",
\"102145\") == 0"), [];
+          "check_hash (ret, \"PART_ENTRY_SIZE\",
\"1023745\") == 0"), [];
     ];
     shortdesc = "print block device attributes";
     longdesc = "\
diff --git a/tests/c-api/tests-main.c b/tests/c-api/tests-main.c
index e2e86f5..e81e15e 100644
--- a/tests/c-api/tests-main.c
+++ b/tests/c-api/tests-main.c
@@ -436,7 +436,7 @@ create_handle (void)
     exit (EXIT_FAILURE);
   }
 
-  if (guestfs_add_drive_scratch (g, 52428800, -1) == -1) {
+  if (guestfs_add_drive_scratch (g, 524288000, -1) == -1) {
     printf ("FAIL: guestfs_add_drive_scratch\n");
     exit (EXIT_FAILURE);
   }
-- 
1.9.3

Mostly complete and making most of the features work.

The notable lack is ntfsprogs and ntfs-3g, which due to the way
mount.ntfs is handled (i.e. using alternatives) do not work properly.
---
 appliance/packagelist.in | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/appliance/packagelist.in b/appliance/packagelist.in
index 3aeafc6..b0a713a 100644
--- a/appliance/packagelist.in
+++ b/appliance/packagelist.in
@@ -125,6 +125,27 @@ ifelse(ARCHLINUX,1,
   zfs-fuse
 )
 
+ifelse(SUSE,1,
+  augeas
+  dnl It seems no other augeas package depends on it.
+  augeas-lenses
+  btrfsprogs
+  cryptsetup
+  e2fsprogs
+  genisoimage
+  gptfdisk
+  hivex
+  iproute2
+  iputils
+  libcap2
+  libselinux1
+  reiserfs
+  syslinux
+  systemd
+  vim
+  xz
+)
+
 ifelse(FRUGALWARE,1,
   augeas
   btrfs-progs
-- 
1.9.3

On Tue, Aug 19, 2014 at 04:01:04PM +0200, Pino Toscano
wrote:
> Do not use the --block-size parameter of xz when compressing the test
> images, as that option does not exist in xz < 5.1.
> ---
>  builder/Makefile.am | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/builder/Makefile.am b/builder/Makefile.am
> index 6a5cb01..726ca60 100644
> --- a/builder/Makefile.am
> +++ b/builder/Makefile.am
> @@ -224,12 +224,12 @@ fedora.qcow2: ../tests/guests/fedora.img
>  
>  fedora.qcow2.xz: fedora.qcow2
>  	rm -f $@ $@-t
> -	xz --best --block-size=16777216 -c $< > $@-t
> +	xz --best -c $< > $@-t
>  	mv $@-t $@
>  
>  %.xz: ../tests/guests/%.img
>  	rm -f $@ $@-t
> -	xz --best --block-size=16777216 -c $< > $@-t
> +	xz --best -c $< > $@-t
>  	mv $@-t $@
>  
>  TESTS = \
Hmmm, but OK.

I don't understand the reason why these distros ship ancient
versions of xz TBH ...

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-builder quickly builds VMs from scratch
http://libguestfs.org/virt-builder.1.html

On Tue, Aug 19, 2014 at 04:01:05PM +0200, Pino Toscano
wrote:
> 50M can be not enough when uploading various copies of unstripped
> static binaries from the system, which can fill up such a small disk.
> ---
>  generator/actions.ml     | 2 +-
>  tests/c-api/tests-main.c | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/generator/actions.ml b/generator/actions.ml
> index 9570d9b..245da6f 100644
> --- a/generator/actions.ml
> +++ b/generator/actions.ml
> @@ -9607,7 +9607,7 @@ device is stopped, but it is not destroyed or
zeroed." };
>            "check_hash (ret, \"PART_ENTRY_NUMBER\",
\"1\") == 0 && "^
>            "check_hash (ret, \"PART_ENTRY_TYPE\",
\"0x83\") == 0 && "^
>            "check_hash (ret, \"PART_ENTRY_OFFSET\",
\"128\") == 0 && "^
> -          "check_hash (ret, \"PART_ENTRY_SIZE\",
\"102145\") == 0"), [];
> +          "check_hash (ret, \"PART_ENTRY_SIZE\",
\"1023745\") == 0"), [];
>      ];
>      shortdesc = "print block device attributes";
>      longdesc = "\
> diff --git a/tests/c-api/tests-main.c b/tests/c-api/tests-main.c
> index e2e86f5..e81e15e 100644
> --- a/tests/c-api/tests-main.c
> +++ b/tests/c-api/tests-main.c
> @@ -436,7 +436,7 @@ create_handle (void)
>      exit (EXIT_FAILURE);
>    }
>  
> -  if (guestfs_add_drive_scratch (g, 52428800, -1) == -1) {
> +  if (guestfs_add_drive_scratch (g, 524288000, -1) == -1) {
>      printf ("FAIL: guestfs_add_drive_scratch\n");
>      exit (EXIT_FAILURE);
>    }
> -- 
> 1.9.3
ACK, but please also change the documentation: See src/guestfs.pod
under the section "ADDING TESTS FOR AN API ACTION".

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-top is 'top' for virtual machines.  Tiny program with many
powerful monitoring features, net stats, disk stats, logging, etc.
http://people.redhat.com/~rjones/virt-top

On Tue, Aug 19, 2014 at 04:01:06PM +0200, Pino Toscano
wrote:
> Mostly complete and making most of the features work.
> 
> The notable lack is ntfsprogs and ntfs-3g, which due to the way
> mount.ntfs is handled (i.e. using alternatives) do not work properly.
> ---
>  appliance/packagelist.in | 21 +++++++++++++++++++++
>  1 file changed, 21 insertions(+)
> 
> diff --git a/appliance/packagelist.in b/appliance/packagelist.in
> index 3aeafc6..b0a713a 100644
> --- a/appliance/packagelist.in
> +++ b/appliance/packagelist.in
> @@ -125,6 +125,27 @@ ifelse(ARCHLINUX,1,
>    zfs-fuse
>  )
>  
> +ifelse(SUSE,1,
> +  augeas
> +  dnl It seems no other augeas package depends on it.
> +  augeas-lenses
> +  btrfsprogs
> +  cryptsetup
> +  e2fsprogs
> +  genisoimage
> +  gptfdisk
> +  hivex
> +  iproute2
> +  iputils
> +  libcap2
> +  libselinux1
> +  reiserfs
> +  syslinux
> +  systemd
> +  vim
> +  xz
> +)
> +
>  ifelse(FRUGALWARE,1,
>    augeas
>    btrfs-progs
> -- 
> 1.9.3
ACK.

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-top is 'top' for virtual machines.  Tiny program with many
powerful monitoring features, net stats, disk stats, logging, etc.
http://people.redhat.com/~rjones/virt-top