src/util/virstoragefile.c | 2 +- tests/virstoragetest.c | 4 ++++ 2 files changed, 5 insertions(+), 1 deletion(-)
When parsing legacy NBD backing file strings such as
'nbd:unix:/tmp/sock:exportname=/' we'd fail to set the transport to
VIR_STORAGE_NET_HOST_TRANS_UNIX. This started to be a problem once we
actually started to generate config of the backing store on the command
line with -blockdev as the JSON code would try to format it as TCP and
fail with:
internal error: argument key 'host' must not have null value
Set the type properly and add a test.
This bug was found by the libguestfs test suite in:
https://bugzilla.redhat.com/show_bug.cgi?id=1791614
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
---
src/util/virstoragefile.c | 2 +-
tests/virstoragetest.c | 4 ++++
2 files changed, 5 insertions(+), 1 deletion(-)
diff --git a/src/util/virstoragefile.c b/src/util/virstoragefile.c
index 1397f532fd..7a2af0ad94 100644
--- a/src/util/virstoragefile.c
+++ b/src/util/virstoragefile.c
@@ -2964,7 +2964,7 @@ virStorageSourceParseNBDColonString(const char *nbdstr,
}
src->hosts->socket = g_strdup(backing[2]);
-
+ src->hosts->transport = VIR_STORAGE_NET_HOST_TRANS_UNIX;
} else {
src->hosts->name = g_strdup(backing[1]);
diff --git a/tests/virstoragetest.c b/tests/virstoragetest.c
index 2862758752..370e19252b 100644
--- a/tests/virstoragetest.c
+++ b/tests/virstoragetest.c
@@ -1258,6 +1258,10 @@ mymain(void)
"<source protocol='nbd' name=':test'>\n"
" <host name='example.org' port='6000'/>\n"
"</source>\n");
+ TEST_BACKING_PARSE("nbd:unix:/tmp/sock:exportname=/",
+ "<source protocol='nbd' name='/'>\n"
+ " <host transport='unix' socket='/tmp/sock'/>\n"
+ "</source>\n");
TEST_BACKING_PARSE("nbd://example.org:1234",
"<source protocol='nbd'>\n"
" <host name='example.org' port='1234'/>\n"
--
2.24.1
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
On Thu, Jan 16, 2020 at 12:41:37PM +0100, Peter Krempa wrote:
> When parsing legacy NBD backing file strings such as
> 'nbd:unix:/tmp/sock:exportname=/' we'd fail to set the transport to
> VIR_STORAGE_NET_HOST_TRANS_UNIX. This started to be a problem once we
> actually started to generate config of the backing store on the command
> line with -blockdev as the JSON code would try to format it as TCP and
> fail with:
>
> internal error: argument key 'host' must not have null value
>
> Set the type properly and add a test.
>
> This bug was found by the libguestfs test suite in:
>
> https://bugzilla.redhat.com/show_bug.cgi?id=1791614
The bug was found by Ming Xie of the virt-v2v QE team.
The patch itself looks fine as far as I can tell. But I'm having real
problems actually testing it. Can you suggest any way to test libvirt
parsing these URIs?
My current method (which doesn't work for reasons that I don't
understand) is:
(1) Compile libvirt from source.
(2) Run ./run src/libvirtd &
(3) Create an NBD server + overlay file:
rm /tmp/sock
nbdkit -U /tmp/sock memory 1G
qemu-img create -f qcow2 overlay.qcow2 -b nbd:unix:/tmp/sock:exportname=/ -F raw
(4) Try to boot a libvirt guest using the overlay:
virt-install --import --name test --disk path=overlay.qcow2,format=raw --memory 1024
But for some reason libvirt just ignores the overlay:
2020-01-16 13:00:44.980+0000: 2378711: warning : virStorageBackendVolOpen:1527 : ignoring missing file 'nbd:unix:/tmp/sock'
I verified that the overlay works in programs like qemu-img so it
doesn't seem to be a problem with the overlay itself.
Rich.
> Signed-off-by: Peter Krempa <pkrempa@redhat.com>
> ---
> src/util/virstoragefile.c | 2 +-
> tests/virstoragetest.c | 4 ++++
> 2 files changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/src/util/virstoragefile.c b/src/util/virstoragefile.c
> index 1397f532fd..7a2af0ad94 100644
> --- a/src/util/virstoragefile.c
> +++ b/src/util/virstoragefile.c
> @@ -2964,7 +2964,7 @@ virStorageSourceParseNBDColonString(const char *nbdstr,
> }
>
> src->hosts->socket = g_strdup(backing[2]);
> -
> + src->hosts->transport = VIR_STORAGE_NET_HOST_TRANS_UNIX;
> } else {
> src->hosts->name = g_strdup(backing[1]);
>
> diff --git a/tests/virstoragetest.c b/tests/virstoragetest.c
> index 2862758752..370e19252b 100644
> --- a/tests/virstoragetest.c
> +++ b/tests/virstoragetest.c
> @@ -1258,6 +1258,10 @@ mymain(void)
> "<source protocol='nbd' name=':test'>\n"
> " <host name='example.org' port='6000'/>\n"
> "</source>\n");
> + TEST_BACKING_PARSE("nbd:unix:/tmp/sock:exportname=/",
> + "<source protocol='nbd' name='/'>\n"
> + " <host transport='unix' socket='/tmp/sock'/>\n"
> + "</source>\n");
> TEST_BACKING_PARSE("nbd://example.org:1234",
> "<source protocol='nbd'>\n"
> " <host name='example.org' port='1234'/>\n"
> --
> 2.24.1
>
> --
> libvir-list mailing list
> libvir-list@redhat.com
> https://www.redhat.com/mailman/listinfo/libvir-list
On Thu, Jan 16, 2020 at 01:13:38PM +0000, Richard W.M. Jones wrote: > On Thu, Jan 16, 2020 at 12:41:37PM +0100, Peter Krempa wrote: > > When parsing legacy NBD backing file strings such as > > 'nbd:unix:/tmp/sock:exportname=/' we'd fail to set the transport to > > VIR_STORAGE_NET_HOST_TRANS_UNIX. This started to be a problem once we > > actually started to generate config of the backing store on the command > > line with -blockdev as the JSON code would try to format it as TCP and > > fail with: > > > > internal error: argument key 'host' must not have null value > > > > Set the type properly and add a test. > > > > This bug was found by the libguestfs test suite in: > > > > https://bugzilla.redhat.com/show_bug.cgi?id=1791614 > > The bug was found by Ming Xie of the virt-v2v QE team. > > The patch itself looks fine as far as I can tell. But I'm having real > problems actually testing it. Can you suggest any way to test libvirt > parsing these URIs? > > My current method (which doesn't work for reasons that I don't > understand) is: > > (1) Compile libvirt from source. > > (2) Run ./run src/libvirtd & > > (3) Create an NBD server + overlay file: > > rm /tmp/sock > nbdkit -U /tmp/sock memory 1G > qemu-img create -f qcow2 overlay.qcow2 -b nbd:unix:/tmp/sock:exportname=/ -F raw > > (4) Try to boot a libvirt guest using the overlay: > > virt-install --import --name test --disk path=overlay.qcow2,format=raw --memory 1024 > > But for some reason libvirt just ignores the overlay: > > 2020-01-16 13:00:44.980+0000: 2378711: warning : virStorageBackendVolOpen:1527 : ignoring missing file 'nbd:unix:/tmp/sock' > > I verified that the overlay works in programs like qemu-img so it > doesn't seem to be a problem with the overlay itself. The problem was too old qemu. With qemu 4.2 I can verify that the patch is working, so: Tested-by: Richard W.M. Jones <rjones@redhat.com> 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 Thu, Jan 16, 2020 at 13:13:38 +0000, Richard W.M. Jones wrote:
> On Thu, Jan 16, 2020 at 12:41:37PM +0100, Peter Krempa wrote:
> > When parsing legacy NBD backing file strings such as
> > 'nbd:unix:/tmp/sock:exportname=/' we'd fail to set the transport to
> > VIR_STORAGE_NET_HOST_TRANS_UNIX. This started to be a problem once we
> > actually started to generate config of the backing store on the command
> > line with -blockdev as the JSON code would try to format it as TCP and
> > fail with:
> >
> > internal error: argument key 'host' must not have null value
> >
> > Set the type properly and add a test.
> >
> > This bug was found by the libguestfs test suite in:
> >
> > https://bugzilla.redhat.com/show_bug.cgi?id=1791614
>
> The bug was found by Ming Xie of the virt-v2v QE team.
I'll add a
Reported-by: Ming Xie <mxie@redhat.com>
>
> The patch itself looks fine as far as I can tell. But I'm having real
> problems actually testing it. Can you suggest any way to test libvirt
> parsing these URIs?
>
> My current method (which doesn't work for reasons that I don't
> understand) is:
>
> (1) Compile libvirt from source.
>
> (2) Run ./run src/libvirtd &
>
> (3) Create an NBD server + overlay file:
>
> rm /tmp/sock
> nbdkit -U /tmp/sock memory 1G
> qemu-img create -f qcow2 overlay.qcow2 -b nbd:unix:/tmp/sock:exportname=/ -F raw
>
> (4) Try to boot a libvirt guest using the overlay:
>
> virt-install --import --name test --disk path=overlay.qcow2,format=raw --memory 1024
>
> But for some reason libvirt just ignores the overlay:
Do you also use qemu-4.2? The bug manifests itself only when -blockdev
is used. That's why it was dormant at least since 2014. I did not bother
looking further into the history.
At any rate, if it's inconvenient to install qemu-4.2 you can
force-enable blockdev for testing using:
<domain type='kvm' xmlns:qemu='http://libvirt.org/schemas/domain/qemu/1.0'>
...
<qemu:capabilities>
<qemu:add capability='blockdev'/>
</qemu:capabilities>
</domain>
>
> 2020-01-16 13:00:44.980+0000: 2378711: warning : virStorageBackendVolOpen:1527 : ignoring missing file 'nbd:unix:/tmp/sock'
>
> I verified that the overlay works in programs like qemu-img so it
> doesn't seem to be a problem with the overlay itself.
>
> Rich.
>
> > Signed-off-by: Peter Krempa <pkrempa@redhat.com>
> > ---
> > src/util/virstoragefile.c | 2 +-
> > tests/virstoragetest.c | 4 ++++
> > 2 files changed, 5 insertions(+), 1 deletion(-)
> >
> > diff --git a/src/util/virstoragefile.c b/src/util/virstoragefile.c
> > index 1397f532fd..7a2af0ad94 100644
> > --- a/src/util/virstoragefile.c
> > +++ b/src/util/virstoragefile.c
> > @@ -2964,7 +2964,7 @@ virStorageSourceParseNBDColonString(const char *nbdstr,
> > }
> >
> > src->hosts->socket = g_strdup(backing[2]);
> > -
> > + src->hosts->transport = VIR_STORAGE_NET_HOST_TRANS_UNIX;
> > } else {
> > src->hosts->name = g_strdup(backing[1]);
> >
> > diff --git a/tests/virstoragetest.c b/tests/virstoragetest.c
> > index 2862758752..370e19252b 100644
> > --- a/tests/virstoragetest.c
> > +++ b/tests/virstoragetest.c
> > @@ -1258,6 +1258,10 @@ mymain(void)
> > "<source protocol='nbd' name=':test'>\n"
> > " <host name='example.org' port='6000'/>\n"
> > "</source>\n");
> > + TEST_BACKING_PARSE("nbd:unix:/tmp/sock:exportname=/",
> > + "<source protocol='nbd' name='/'>\n"
> > + " <host transport='unix' socket='/tmp/sock'/>\n"
> > + "</source>\n");
> > TEST_BACKING_PARSE("nbd://example.org:1234",
> > "<source protocol='nbd'>\n"
> > " <host name='example.org' port='1234'/>\n"
> > --
> > 2.24.1
> >
> > --
> > libvir-list mailing list
> > libvir-list@redhat.com
> > https://www.redhat.com/mailman/listinfo/libvir-list
On 1/16/20 5:41 AM, Peter Krempa wrote: > When parsing legacy NBD backing file strings such as > 'nbd:unix:/tmp/sock:exportname=/' we'd fail to set the transport to > VIR_STORAGE_NET_HOST_TRANS_UNIX. This started to be a problem once we > actually started to generate config of the backing store on the command > line with -blockdev as the JSON code would try to format it as TCP and > fail with: > > internal error: argument key 'host' must not have null value > > Set the type properly and add a test. > > This bug was found by the libguestfs test suite in: > > https://bugzilla.redhat.com/show_bug.cgi?id=1791614 > > Signed-off-by: Peter Krempa <pkrempa@redhat.com> > --- > src/util/virstoragefile.c | 2 +- > tests/virstoragetest.c | 4 ++++ > 2 files changed, 5 insertions(+), 1 deletion(-) > Reviewed-by: Eric Blake <eblake@redhat.com> -- Eric Blake, Principal Software Engineer Red Hat, Inc. +1-919-301-3226 Virtualization: qemu.org | libvirt.org
© 2016 - 2026 Red Hat, Inc.