similar to: Re: [PATCH v3 0/6] launch: direct: Disable qemu locking when opening drives readonly.

On Wed, Sep 13, 2017 at 11:25:32AM +0300, Roman Kagan wrote: > On Tue, Sep 12, 2017 at 06:04:18PM +0100, Richard W.M. Jones wrote: > > v2 -> v3: > > > > - I addressed everything that Pino mentioned last time. > > > > - It's tricky to get a stable run when multiple copies of qemu are > > involved, because the same cache files get overwritten by

On Thu, Sep 14, 2017 at 04:15:10PM +0300, Roman Kagan wrote: > What ABI is broken by this? It's used all over, major tools like virt-tail simply assume that it works this way. I'm not disputing that locking in QEMU is valuable, and we can certainly add a flag to obey locking when readonly=true. > In fact it was the amount of reports of weird misbehavior of all sorts > that made

--- lib/launch-direct.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/lib/launch-direct.c b/lib/launch-direct.c index 9f85013f9..00cb25077 100644 --- a/lib/launch-direct.c +++ b/lib/launch-direct.c @@ -66,6 +66,7 @@ struct backend_direct_data { pid_t recoverypid; /* Recovery process PID. */ struct version qemu_version; /* qemu version (0 if

On Tuesday, 12 September 2017 14:29:16 CEST Richard W.M. Jones wrote: > @@ -255,11 +256,13 @@ add_drive_standard_params (guestfs_h *g, struct backend_direct_data *data, > } > else { > /* Writable qcow2 overlay on top of read-only drive. */ > - append_list_format ("file=%s", drv->overlay); > + append_list_format ("file.file.filename=%s",

On Wed, Sep 06, 2017 at 10:41:24AM +0300, Roman Kagan wrote: > On Tue, Sep 05, 2017 at 03:55:08PM +0100, Richard W.M. Jones wrote: > > Since qemu 2.10, qemu locks all drives even when they are opened > > readonly. This disables this locking in this safe case. > > > > Incomplete fix for > > https://bugzilla.redhat.com/show_bug.cgi?id=1417306 > > --- >

On Tuesday, 5 September 2017 16:55:08 CEST Richard W.M. Jones wrote: > Since qemu 2.10, qemu locks all drives even when they are opened > readonly. This disables this locking in this safe case. > > Incomplete fix for > https://bugzilla.redhat.com/show_bug.cgi?id=1417306 > --- I guess that, according to the other message in this thread, the commit message will be amended to use

Since qemu 2.10, qemu locks all drives even when they are opened readonly. This disables this locking in this safe case. Incomplete fix for https://bugzilla.redhat.com/show_bug.cgi?id=1417306 --- lib/launch-direct.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/lib/launch-direct.c b/lib/launch-direct.c index 3b848165c..e5465539d 100644 --- a/lib/launch-direct.c +++

Incomplete fix for https://bugzilla.redhat.com/show_bug.cgi?id=1417306 The full fix is waiting for a libvirt change, but this can still go upstream. Rich.

v2 -> v3: - I addressed everything that Pino mentioned last time. - It's tricky to get a stable run when multiple copies of qemu are involved, because the same cache files get overwritten by parallel libguestfs. So I changed the names of the cache files to include the qemu binary key (size, mtime), which removes this conflict. This is in new patch 4/6. Rich.

Patches 1-4 are almost the same as they are when previously posted here: https://www.redhat.com/archives/libguestfs/2017-September/msg00039.html Patch 5 actually uses the mandatory locking test to turn off locking in the narrow case where a drive is opened readonly, and then only for the drive being inspected. Passes ordinary tests (‘check-direct’ and ‘check-valgrind-direct’). Rich.

Ancient qemu 1.5 (in RHEL 7) does not understand the file.file.filename= and file.driver= parameters. Go back to using the old-style file= and format= parameters when we're not trying to set the file.backing.file.locking=off parameter. Fixes commit 9fe592808ccfd9ed184b88ca9c6cad2e1798dee3. Thanks: Yongkui Guo, Václav Kadlčík. --- lib/launch-direct.c | 31 +++++++++++++++++++++---------- 1

On Mon, Apr 18, 2016 at 02:49:07PM +0200, Pino Toscano wrote: > The drive used for the appliance is a raw (sparse) disk: specify that > explicitly in its -drive qemu command line options, so qemu can skip the > autodetection of its format and save a tiny bit of time. > --- > src/launch-direct.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git

Signed-off-by: Roman Kagan <rkagan@virtuozzo.com> --- v2v/v2v.ml | 32 ++++++++++++++++++++------------ 1 file changed, 20 insertions(+), 12 deletions(-) diff --git a/v2v/v2v.ml b/v2v/v2v.ml index c28905d..23bd708 100644 --- a/v2v/v2v.ml +++ b/v2v/v2v.ml @@ -60,19 +60,9 @@ let rec main () = let overlays = create_overlays source.s_disks in let targets = init_targets overlays source

On Mon, Oct 02, 2017 at 11:50:42AM +0300, Maxim Kozover wrote: > Hi Richard! > Could you tell me, please, if there is a way to relaunch libguestfs with > the same overlay from the last run? The documentation tells the overlay is > removed when the handle to libguestfs is closed. Right, the overlay is created by libguestfs itself and thrown away when the handle is closed:

From: "Richard W.M. Jones" <rjones at redhat.com> You can now add remote NBD drives using: ><fs> add-drive "" format:raw protocol:nbd server:localhost (Note that you also need to add port:NNNN if the server is running on a non-standard port). The corresponding qemu-nbd service can be started by doing: qemu-nbd disk.img -t This commit also adds a test. ---

I'm using qemu-nbd to setup the device with an offset into the image. Using libguestfs version 1.21.22 Code for adding the drive: char *server[] = { "localhost:3000", NULL }; if (guestfs_add_drive_opts (g, "" /* export name - see below */, GUESTFS_ADD_DRIVE_OPTS_FORMAT, "raw", GUESTFS_ADD_DRIVE_OPTS_PROTOCOL, "nbd",

I notice that the qemu driver doesn't support snapshot drives (-drive file=foo,snapshot=on). This is important for libguestfs. Currently libguestfs hacks this using <qemu:arg>. That works fine for static disks in the libvirt XML, but lack of direct support in libvirt is a blocker for adding hotplugging to libguestfs. In qemu, the snapshot=on feature does several things: (a) It

Dr. David Gilbert pointed out to me that the first line of the qemu -help output includes the qemu version, so we don't need to run qemu -version at all. This saves about 0.04s on the launch time when using the direct backend. --- src/launch-direct.c | 54 +++++++++++++++++++---------------------------------- 1 file changed, 19 insertions(+), 35 deletions(-) diff --git a/src/launch-direct.c

Besides, it doesn't need guestfs handle open so move this step earlier in the process. Signed-off-by: Roman Kagan <rkagan@virtuozzo.com> --- v2v/v2v.ml | 77 ++++++++++++++++++++++++++++++++------------------------------ 1 file changed, 40 insertions(+), 37 deletions(-) diff --git a/v2v/v2v.ml b/v2v/v2v.ml index 155eb83..4257b8d 100644 --- a/v2v/v2v.ml +++ b/v2v/v2v.ml @@ -58,6 +58,7

Opening the source VM and amending the properties in its internal representation in accordance with command-line options fit nicely into two isolated functions. Signed-off-by: Roman Kagan <rkagan@virtuozzo.com> --- v2v/v2v.ml | 124 ++++++++++++++++++++++++++++++++----------------------------- 1 file changed, 65 insertions(+), 59 deletions(-) diff --git a/v2v/v2v.ml b/v2v/v2v.ml index