On 5/21/10 6:50 AM, "Stefan Berger" wrote:
> This patch may get 802.1Qbh devices working. I am adding some code to
> poll for the status of an 802.1Qbh device and loop for a while until the
> status indicates success. This part for sure needs more work and
> testing...
I think we can drop this p
* Stefan Berger (stef...@linux.vnet.ibm.com) wrote:
> This is so far a more or less experimental patch that adds code for
> communicating with the external lldpad daemon using the recently added
> netlink messages. It's communicating using multicast netlink messages in
> case of 802.1Qbg and unicas
* Stefan Berger (stef...@linux.vnet.ibm.com) wrote:
> This patch introduces a dependency on libnl, which subsequent patches
> will then use.
>
> Changes from V1 to V2:
> - added diffstats
> - following changes in tree
>
> Signed-off-by: Stefan Berger
This looks good to me
Acked-by: Chris Wrigh
* Stefan Berger (stef...@linux.vnet.ibm.com) wrote:
> Below is David Alan's original patch with lots of changes.
>
> In particular, it now parses the following two XML descriptions, one
> for 802.1Qbg and 802.1Qbh and stored the data internally. The actual
> triggering of the switch setup protoco
* src/util/bitmap.h (includes): Placate cppi.
---
Pushing as obvious, to keep 'make syntax-check' happy.
src/util/bitmap.h |6 +++---
1 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/src/util/bitmap.h b/src/util/bitmap.h
index be897ff..08515d1 100644
--- a/src/util/bitmap.h
+++
Below is David Alan's original patch with lots of changes.
In particular, it now parses the following two XML descriptions, one
for 802.1Qbg and 802.1Qbh and stored the data internally. The actual
triggering of the switch setup protocol has not been implemented
here but the relevant code to do th
2010/5/21 Eric Blake :
> Otherwise, './configure --prefix=/foo && make &&
> ./configure --prefix=/bar && make' leaves the wrong files
> in libvirtd.init (/foo instead of /bar).
>
> * daemon/Makefile.am (libvirtd.init): Add dependency on
> config.status. Reported by Cole Robinson.
> ---
>
> Thanks
Otherwise, './configure --prefix=/foo && make &&
./configure --prefix=/bar && make' leaves the wrong files
in libvirtd.init (/foo instead of /bar).
* daemon/Makefile.am (libvirtd.init): Add dependency on
config.status. Reported by Cole Robinson.
---
Thanks for the simple recipe for testing; and
On 05/21/2010 11:05 AM, Cole Robinson wrote:
> Volume detection in the scsi backend was duplicating code already
> present in storage_backend.c. Let's drop the duplicate code.
>
> Also, change the shared function name to be less generic, and remove
> some error squashing in the other call site.
A
On 05/21/2010 11:05 AM, Cole Robinson wrote:
> We were squashing error messages in a few cases. Recode to follow common
> ret = -1 convention.
>
> v2: Handle more error squashing issues further up in MakeNewVol and
> CreateVols. Use ret = -1 convention in MakeVols.
ACK; looks better this time
On 05/21/2010 11:05 AM, Cole Robinson wrote:
> Spurious / in a pool target path makes life difficult for apps using the
> GetVolByPath, and doing other path based comparisons with pools. This
> has caused a few issues for virt-manager users:
>
> https://bugzilla.redhat.com/show_bug.cgi?id=494005
>
On 05/21/2010 02:17 PM, Eric Blake wrote:
> On 05/21/2010 11:03 AM, Cole Robinson wrote:
>> If a directory pool contains pipes or sockets, a pool start can fail or hang:
>>
>> https://bugzilla.redhat.com/show_bug.cgi?id=589577
>>
>> We already try to avoid these special files, but only attempt afte
On 05/21/2010 11:03 AM, Cole Robinson wrote:
> If a directory pool contains pipes or sockets, a pool start can fail or hang:
>
> https://bugzilla.redhat.com/show_bug.cgi?id=589577
>
> We already try to avoid these special files, but only attempt after
> opening the path, which is where the proble
2010/5/20 Eric Blake :
> On 05/19/2010 04:42 PM, Matthias Bolte wrote:
>>> @@ -46,7 +46,8 @@ qemuhelpdata = \
>>> qemu-0.10.5 \
>>> qemu-0.12.1 \
>>> qemu-kvm-0.10.5 \
>>> - qemu-kvm-0.11.0-rc2
>>> + qemu-kvm-0.11.0-rc2 \
>>> + qemu-kvm-0.12.3
>>>
>>> EXTRA_D
Eric Blake wrote:
> On 05/21/2010 09:29 AM, Jim Fehlig wrote:
>
>> This and your other comments are addressed in V3.
>>
>>
>
> Oops, just realized we missed one thing - if you haven't pushed yet,
> then please squash in a change to use ATTRIBUTE_RETURN_CHECK in bitmap.h
> for the three func
On 05/21/2010 01:21 PM, Eric Blake wrote:
On 05/21/2010 09:28 AM, Cole Robinson wrote:
Any interest in doing this with netlink instead? (I've got this "thing"
against parsing text files to get information if it can be retrieved via
a nice clean API). If so, I think I can whip up the equivale
I'm stepping back, for now, on my suggestion of using libnl to do this.
The proposed method works, and solves a serious problem. Using libnl
would also work, but would create a new dependency (which can be a bit
hairy with libnl - eg RHEL5 has libnl-1.0 which is API-incompatible with
libnl-1.1,
On 05/21/2010 07:16 AM, Paolo Smiraglia wrote:
> Hi guys, I'm working on a pre migration checks implementation.
>
> The patch files attached provide a framework to execute
> pre-migration-checks before the domain migration.
First, thanks for the efforts (sometimes, as open source developers, we
g
On 05/21/2010 09:28 AM, Cole Robinson wrote:
>> Any interest in doing this with netlink instead? (I've got this "thing"
>> against parsing text files to get information if it can be retrieved via
>> a nice clean API). If so, I think I can whip up the equivalent code with
>> libnl calls, but prob
On 05/20/2010 03:17 PM, Eric Blake wrote:
> On 05/20/2010 10:04 AM, Cole Robinson wrote:
>> /etc/sysconfig/libvirtd has a few environment variables for configuring
>> libvirt SDL audio. The libvirtd process doesn't see these, however, because
>> they are never exported. Let's export the variables a
On 05/20/2010 06:25 PM, Eric Blake wrote:
> On 05/20/2010 01:23 PM, Cole Robinson wrote:
>> If a directory pool contains pipes or sockets, a pool start can fail or hang:
>>
>> https://bugzilla.redhat.com/show_bug.cgi?id=589577
>>
>> We already try to avoid these special files, but only attempt afte
On 05/20/2010 03:28 PM, Eric Blake wrote:
> On 05/20/2010 10:04 AM, Cole Robinson wrote:
>> Spurious / in a pool target path makes life difficult for apps using the
>> GetVolByPath, and doing other path based comparisons with pools. This
>> has caused a few issues for virt-manager users:
>>
>> http
Spurious / in a pool target path makes life difficult for apps using the
GetVolByPath, and doing other path based comparisons with pools. This
has caused a few issues for virt-manager users:
https://bugzilla.redhat.com/show_bug.cgi?id=494005
https://bugzilla.redhat.com/show_bug.cgi?id=593565
Add
We were squashing error messages in a few cases. Recode to follow common
ret = -1 convention.
v2: Handle more error squashing issues further up in MakeNewVol and
CreateVols. Use ret = -1 convention in MakeVols.
Signed-off-by: Cole Robinson
---
src/storage/storage_backend_mpath.c | 40
Volume detection in the scsi backend was duplicating code already
present in storage_backend.c. Let's drop the duplicate code.
Also, change the shared function name to be less generic, and remove
some error squashing in the other call site.
v2: Rebased around other storage changes
Signed-off-by:
If a directory pool contains pipes or sockets, a pool start can fail or hang:
https://bugzilla.redhat.com/show_bug.cgi?id=589577
We already try to avoid these special files, but only attempt after
opening the path, which is where the problems lie. Unify volume opening
into a single function which
On 05/21/2010 07:22 AM, Chris Lalancette wrote:
>>> -virCheckFlags(VIR_MIGRATE_NON_SHARED_DISK | VIR_MIGRATE_NON_SHARED_INC,
>>> - -1);
>>> -
>>
>> Why are you removing the valid flag check altogether? Is it a matter of
>> adding more valid flags, or is this just a forwarding
On 05/21/2010 05:47 AM, Jiri Denemark wrote:
> The event is already generated and sent by qemudDomainObjStart, no need
> to do anything about here.
> ---
> src/qemu/qemu_driver.c |3 ---
> 1 files changed, 0 insertions(+), 3 deletions(-)
>
> diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu
On 05/20/2010 03:57 PM, Chris Lalancette wrote:
> We've been running into a lot of situations where
> virGetHostname() is returning "localhost", where a plain
> gethostname() would have returned the correct thing. This
> is because virGetHostname() is *always* trying to canonicalize
> the name ret
On 05/21/2010 10:00 AM, Jim Fehlig wrote:
> +if (virBitmapGetBit(driver->reservedVNCPorts,
> +i - QEMU_VNC_PORT_MIN, &used) < 0)
> +VIR_DEBUG("virBitmapGetBit failed on bit %d", i -
> QEMU_VNC_PORT_MIN);
> +
> +if (used)
> +contin
On 05/21/2010 09:29 AM, Jim Fehlig wrote:
> This and your other comments are addressed in V3.
>
Oops, just realized we missed one thing - if you haven't pushed yet,
then please squash in a change to use ATTRIBUTE_RETURN_CHECK in bitmap.h
for the three functions that return an int.
--
Eric Blake
On 05/21/2010 09:29 AM, Jim Fehlig wrote:
>
> This and your other comments are addressed in V3.
>
> V3:
> - Check for overflow in virBitmapAlloc
> - Fix copy and paste bug in virBitmapAlloc
> - Use size_t in prototypes
> - Add ATTRIBUTE_NONNULL in prototypes where appropriate
> and re
On 05/20/2010 06:10 PM, Eric Blake wrote:
> On 05/20/2010 01:23 PM, Cole Robinson wrote:
>> We were squashing error messages in a few cases. Recode to follow common
>> ret = -1 convention.
>
> s/squashing/duplicating/, but I did check that the only caller,
> virStorageBackendMpathNewVol does print
Eric Blake wrote:
> On 05/20/2010 02:45 PM, Jim Fehlig wrote:
>
>> for (i = QEMU_VNC_PORT_MIN; i < QEMU_VNC_PORT_MAX; i++) {
>> int fd;
>> int reuse = 1;
>> struct sockaddr_in addr;
>> +bool used = false;
>> +
>> +virBitmapGetBit(driver->reservedVN
Eric Blake wrote:
> On 05/20/2010 02:45 PM, Jim Fehlig wrote:
>
>> +virBitmapPtr virBitmapAlloc(size_t size)
>> +{
>> +virBitmapPtr bitmap;
>> +size_t sz = (size + VIR_BITMAP_BITS_PER_UNIT - 1) /
>> +VIR_BITMAP_BITS_PER_UNIT;
>>
>
> Check for overflow:
> if (SIZE_MAX - V
On 05/21/2010 01:23 AM, Laine Stump wrote:
> On 05/20/2010 07:41 PM, Cole Robinson wrote:
>> Fedora bug https://bugzilla.redhat.com/show_bug.cgi?id=235961
>>
>> If using the default virtual network, an easy way to lose guest network
>> connectivity is to install libvirt inside the VM. The autostart
On Fri, May 21, 2010 at 01:23:32AM -0400, Laine Stump wrote:
> On 05/20/2010 07:41 PM, Cole Robinson wrote:
> >Fedora bug https://bugzilla.redhat.com/show_bug.cgi?id=235961
> >
> >If using the default virtual network, an easy way to lose guest network
> >connectivity is to install libvirt inside th
This patch may get 802.1Qbh devices working. I am adding some code to
poll for the status of an 802.1Qbh device and loop for a while until the
status indicates success. This part for sure needs more work and
testing...
I am recycling link_dump from a previous patch.
Changes from V1 to V2:
- Follo
This is so far a more or less experimental patch that adds code for
communicating with the external lldpad daemon using the recently added
netlink messages. It's communicating using multicast netlink messages in
case of 802.1Qbg and unicast using 802.1Qbh. In the multicast case I am
currently expec
This patch introduces a dependency on libnl, which subsequent patches
will then use.
Changes from V1 to V2:
- added diffstats
- following changes in tree
Signed-off-by: Stefan Berger
---
configure.ac| 24
libvirt.spec.in | 14 ++
src/Makefile.am |
Below is David Alan's original patch with lots of changes.
In particular, it now parses the following two XML descriptions, one
for 802.1Qbg and 802.1Qbh and stored the data internally. The actual
triggering of the switch setup protocol has not been implemented
here but the relevant code to do th
On 05/20/2010 05:30 PM, Eric Blake wrote:
> On 05/20/2010 01:57 PM, Chris Lalancette wrote:
>> We shouldn't be checking validity in domain_conf, since
>> it can be used by multiple different hosts and hypervisors.
>> Remove the check completely.
>>
>>
>> -static bool
>> -isValidIfname(const char
On 05/20/2010 05:35 PM, Eric Blake wrote:
> On 05/20/2010 01:57 PM, Chris Lalancette wrote:
>> Basic live migration was broken by the commit that added
>> non-shared block support. It also added a couple of
>> built-in memory leaks. This patch fixes these problems
>> and allows normal live migrat
Hi guys, I'm working on a pre migration checks implementation.
The patch files attached provide a framework to execute
pre-migration-checks before the domain migration.
1 - FRAMEWORK OVERVIEW
==
On Thu, 2010-05-20 at 16:55 -0400, Stefan Berger wrote:
> On Thu, 2010-05-20 at 13:32 -0700, Scott Feldman wrote:
> > On 5/20/10 1:10 PM, "Stefan Berger" wrote:
> >
> > >
> >
> > Can we use something more generic than since VSI is a .bg term?
> > Something like virtual_port?
>
> Fine wi
2010/5/21 Jiri Denemark :
>> >>> The problem desribed here:
>> >>> https://bugzilla.redhat.com/show_bug.cgi?id=593266
>> >>>
>> >>> Please update http://www.libvirt.org/formatdomain.html and describe
>> >>> NEW xml tags, like for hard drive and all cpu
>> >>> tags, other if any.
>> >>
>> >> W
The event is already generated and sent by qemudDomainObjStart, no need
to do anything about here.
---
src/qemu/qemu_driver.c |3 ---
1 files changed, 0 insertions(+), 3 deletions(-)
diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c
index 2f644c9..5c8ab38 100644
--- a/src/qemu/qemu
> >>> The problem desribed here:
> >>> https://bugzilla.redhat.com/show_bug.cgi?id=593266
> >>>
> >>> Please update http://www.libvirt.org/formatdomain.html and describe
> >>> NEW xml tags, like for hard drive and all cpu
> >>> tags, other if any.
> >>
> >> What cpu tag are you missing from t
> > Are we sure that init scripts called during '/sbin/shutdown' are
> > normally called with 'stop' rather than 'shutdown' argument?
>
> Hmm, I don't think anything calls init scripts with 'shutdown' argument, but
> it might probably be better to rename as 'guestshutdown' to match
> 'gueststatus
> The fix required some refactoring in various places. More details about why
> and
> what had to be reworked can be found in each of the patches.
>
> Refactor qemudDomainRestore
> Factor out def assignment to existing domain from virDomainAssignDef
> Refactor qemudDomainStart
> Autostart
50 matches
Mail list logo
