- Devel - libvirt List Archives

Re: [libvirt] [RFC] block I/O throttling: how to enable in libvirt
by Zhi Yong Wu 04 Sep '11

04 Sep '11

On Fri, Sep 02, 2011 at 08:34:19AM -0500, Adam Litke wrote: >From: Adam Litke <agl(a)us.ibm.com> >To: Zhi Yong Wu <wuzhy(a)linux.vnet.ibm.com> >Cc: libvir-list(a)redhat.com >Subject: Re: [libvirt] [RFC] block I/O throttling: how to enable in libvirt >Message-ID: <20110902133419.GW15683(a)aglitke.rchland.ibm.com> >References: <20110901050531.GB17963(a)f15.cn.ibm.com> > <20110902021447.GD19143(a)f15.cn.ibm.com> >MIME-Version: 1.0 >Content-Type: text/plain; charset=us-ascii >Content-Disposition: inline >In-Reply-To: <20110902021447.GD19143(a)f15.cn.ibm.com> >User-Agent: Mutt/1.5.21 (2010-09-15) >X-Brightmail-Tracker: AAAAAA== >X-Xagent-From: agl(a)us.ibm.com >X-Xagent-To: wuzhy(a)linux.vnet.ibm.com >X-Xagent-Gateway: vmsdvma.vnet.ibm.com (XAGENTU8 at VMSDVMA) > >On Fri, Sep 02, 2011 at 10:14:48AM +0800, Zhi Yong Wu wrote: >> On Thu, Sep 01, 2011 at 01:05:31PM +0800, Zhi Yong Wu wrote: >> HI, Adam, >> Now stefan, Daniel, and Gui all suggest extending blkiotune to keep libivrt unified interface. What do you think of it? > >It seems like it would be nice to extend the blkiotune API for this use case, >but there is one problem. The blkiotune interface operates at the VM level and >the QEMU throttling is at the disk level. So if you want per-disk throttling >granularity the blkiotune API may not be suitable. I see 3 possible paths >forward: > >1) Accept the blkiotune limitation of a global setting and add a throttle option >that assigns the same throttle value to all disks. > > <blkiotune> > <shares>250</shares> > <throttle>1024</throttle> > </blkiotune> > > + Nice for simplicity and ease of implementation > - May be too simplistic (can't set different throttling for different disks) > > >2) Extend the blkiotune API to allow per-disk tunings: > > <blkiotune> > <shares>250</shares> > <throttle device='virtio-disk0'>1024</throttle> > </blkiotune> > > + Use an existing API > - Unfortunately it doesn't look like this can be done cleanly > > >3) Use a new API virDomainSetBlkDevParameters() to set device specific >tuning parameters: > ><blkiotune> > <shares>250</shares> ></blkiotune> >... ><disk type='file' device='disk'> > <driver name='qemu' type='raw'/> > <source file='...'/> > <target dev='vda' bus='virtio'/> > <blkdevtune> > <throttle>1024</throttle> Several I/O throttling tags should be supported, such as thrbps, thrbpsrd, thrbpswr, thriops, etc. > </blkdevtune> ></disk> > > + Throttle can be specified with the disk it is effecting > - Proliferation of tuning APIs > > >Obviously #1 is the best if you can accept the limitations of the API. #2 would #1 is not accepted by us. If it is adopted, the advantage of QEMU I/O throttling is not used fully. >be nice if we could figure out how to extend virTypedParameter to represent a I guess that #2 is less extendable than #3. >(disk, param) pair. I see #3 as a last resort. Did I miss any other solutions. To be honest, i prefer to #3. At first, i plan to check how blkiotune works currently. >Did I just manage to completely over-complicate this? :) NO, they are nice. thanks. Regards, Zhiyong Wu > >-- >Adam Litke <agl(a)us.ibm.com> >IBM Linux Technology Center

1 0

[libvirt] [test-API][PATCH] Add 2 methods to generate cdrom & floppy XML for test
by Nan Zhang 04 Sep '11

04 Sep '11

add the following 2 functions to utils/Python/xmlbuilder.py * build_cdrom() * build_floppy() --- utils/Python/xmlbuilder.py | 47 +++++++++++++++++++++++++++++++++++++++++- utils/Python/xmlgenerator.py | 15 +++++++------ 2 files changed, 54 insertions(+), 8 deletions(-) diff --git a/utils/Python/xmlbuilder.py b/utils/Python/xmlbuilder.py index 78230da..5a0f8c8 100644 --- a/utils/Python/xmlbuilder.py +++ b/utils/Python/xmlbuilder.py @@ -112,6 +112,33 @@ class XmlBuilder: self.write_toxml(disk) return disk.toxml() + def build_cdrom(self, params): + if params.get('hdmodel', None) == None: + params['hdmodel'] = 'ide' + + if params['hdmodel'] == 'ide': + target_dev = 'hdc' + elif params['hdmodel'] == 'scsi': + target_dev = 'sdc' + else: + print 'Wrong cdrom model.' + + cdrom = xmlgenerator.disk_xml(params, True) + if params['guesttype'] == 'xenpv': + cdrom.getElementsByTagName("target")[0].setAttribute("dev", "xvdc") + else: + cdrom.getElementsByTagName("target")[0].setAttribute("dev", + target_dev) + if __DEBUG__: + self.write_toxml(cdrom) + return cdrom.toxml() + + def build_floppy(self, params): + floppy = xmlgenerator.floppy_xml(params) + if __DEBUG__: + self.write_toxml(floppy) + return floppy.toxml() + def build_interface(self, params): interface = xmlgenerator.interface_xml(params) if __DEBUG__: @@ -177,11 +204,29 @@ if __name__ == "__main__": print '=' * 30, 'disk xml', '=' * 30 params['guesttype'] = 'kvm' params['guestname'] = 'foo' - params['imagepath'] = '/images' params['hdmodel'] = 'virtio' diskxml = xmlobj.build_disk(params) + #--------------------- + # get cdrom xml string + #--------------------- + print '=' * 30, 'cdrom xml', '=' * 30 + params['guesttype'] = 'kvm' + params['guestname'] = 'foo' + params['hdmodel'] = 'ide' + params['bootcd'] = '/tmp/cdrom.img' + + cdromxml = xmlobj.build_cdrom(params) + + #--------------------- + # get floppy xml string + #--------------------- + print '=' * 30, 'floppy xml', '=' * 30 + params['floppysource'] = '/tmp/floppy.img' + + floppyxml = xmlobj.build_floppy(params) + #-------------------------- # get interface xml string #-------------------------- diff --git a/utils/Python/xmlgenerator.py b/utils/Python/xmlgenerator.py index c59ca9e..d57dd33 100644 --- a/utils/Python/xmlgenerator.py +++ b/utils/Python/xmlgenerator.py @@ -308,7 +308,7 @@ def disk_xml(params, cdrom = False): params['guestname'] elif hypertype == 'xen': params['imagepath'] = '/var/lib/xen/images' - params['fullimagepath'] = '/var/lib/xen/images' + '/' + \ + params['fullimagepath'] = params['imagepath'] + '/' + \ params['guestname'] else: print 'DO NOT supported hypervisor.' @@ -367,14 +367,15 @@ def disk_xml(params, cdrom = False): def floppy_xml(params): # Disk element floppy = xml.dom.minidom.Document() - disk_element = floppy.createElement('disk') - disk_element.setAttribute('device', 'floppy') - floppy.appendChild(disk_element) + floppy_element = floppy.createElement('disk') + floppy_element.setAttribute('type', 'file') + floppy_element.setAttribute('device', 'floppy') + floppy.appendChild(floppy_element) # Source element source_element = floppy.createElement('source') source_element.setAttribute('file', params['floppysource']) - disk_element.appendChild(source_element) + floppy_element.appendChild(source_element) # Target element target_element = floppy.createElement('target') @@ -384,11 +385,11 @@ def floppy_xml(params): target_element.setAttribute('dev', 'fda') target_element.setAttribute('bus', 'fdc') - disk_element.appendChild(target_element) + floppy_element.appendChild(target_element) # Readonly readonly = floppy.createElement('readonly') - disk_element.appendChild(readonly) + floppy_element.appendChild(readonly) return floppy -- 1.7.4.4

2 1

[libvirt] [PATCH] libvirtd: create run dir when running at non-root user
by xuhj＠linux.vnet.ibm.com 04 Sep '11

04 Sep '11

From: Xu He Jie <xuhj(a)linux.vnet.ibm.com> Signed-off-by: Xu He Jie <xuhj(a)linux.vnet.ibm.com> When libvirtd is running at non-root user, it won't create ${HOME}/.libvirt. It will show error message: 17:44:16.838: 7035: error : virPidFileAcquirePath:322 : Failed to open pid file --- daemon/libvirtd.c | 46 ++++++++++++++++++++++++++++++++-------------- 1 files changed, 32 insertions(+), 14 deletions(-) diff --git a/daemon/libvirtd.c b/daemon/libvirtd.c index 423c3d7..e0004c7 100644 --- a/daemon/libvirtd.c +++ b/daemon/libvirtd.c @@ -1249,6 +1249,7 @@ int main(int argc, char **argv) { bool privileged = geteuid() == 0 ? true : false; bool implicit_conf = false; bool use_polkit_dbus; + char *run_dir = NULL; struct option opts[] = { { "verbose", no_argument, &verbose, 1}, @@ -1403,21 +1404,35 @@ int main(int argc, char **argv) { /* Ensure the rundir exists (on tmpfs on some systems) */ if (privileged) { - const char *rundir = LOCALSTATEDIR "/run/libvirt"; - mode_t old_umask; - - old_umask = umask(022); - if (mkdir (rundir, 0755)) { - if (errno != EEXIST) { - char ebuf[1024]; - VIR_ERROR(_("unable to create rundir %s: %s"), rundir, - virStrerror(errno, ebuf, sizeof(ebuf))); - ret = VIR_DAEMON_ERR_RUNDIR; - umask(old_umask); - goto cleanup; - } + run_dir = strdup(LOCALSTATEDIR "/run/libvirt"); + if (!run_dir) { + VIR_ERROR(_("Can't allocate memory")); + goto cleanup; + } + } + else { + char *user_dir = NULL; + + if (!(user_dir = virGetUserDirectory(geteuid()))) { + VIR_ERROR(_("Can't determine user directory")); + goto cleanup; + } + + if (virAsprintf(&run_dir, "%s/.libvirt/", user_dir) < 0) { + VIR_ERROR(_("Can't allocate memory")); + VIR_FREE(user_dir); + goto cleanup; } - umask(old_umask); + + VIR_FREE(user_dir); + } + + if (virFileMakePath(run_dir) < 0) { + char ebuf[1024]; + VIR_ERROR(_("unable to create rundir %s: %s"), run_dir, + virStrerror(errno, ebuf, sizeof(ebuf))); + ret = VIR_DAEMON_ERR_RUNDIR; + goto cleanup; } /* Try to claim the pidfile, exiting if we can't */ @@ -1571,6 +1586,9 @@ cleanup: VIR_FREE(sock_file_ro); VIR_FREE(pid_file); VIR_FREE(remote_config_file); + if (run_dir) + VIR_FREE(run_dir); + daemonConfigFree(config); virLogShutdown(); -- 1.7.4.1

3 3

[libvirt] [PATCH 1/2] Fix error detection in device change
by Philipp Hahn 02 Sep '11

02 Sep '11

According to qemu-kvm/qerror.c all messages start with a capital "Device ", but the current code only scans for the lower case "device ". This results in "virDomainUpdateDeviceFlags()" to not detect locked CD-ROMs and reporting success even in the case of a failure: # virsh qemu-monitor-command "$VM" change\ drive-ide0-0-0\ \"/var/lib/libvirt/images/ucs_2.4-0-sec4-20110714145916-dvd-amd64.iso\" Device 'drive-ide0-0-0' is locked # virsh update-device "$VM" /dev/stdin <<<"<disk type='file' device='cdrom'><driver name='qemu' type='raw'/><source file='/var/lib/libvirt/images/ucs_2.4-0-sec4-20110714145916-dvd-amd64.iso'/><target dev='hda' bus='ide'/><readonly/><alias name='ide0-0-0'/><address type='drive' controller='0' bus='0' unit='0'/></disk>" Device updated successfully Signed-off-by: Philipp Hahn <hahn(a)univention.de> --- src/qemu/qemu_monitor_text.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/qemu/qemu_monitor_text.c b/src/qemu/qemu_monitor_text.c index 52d924a..98d9169 100644 --- a/src/qemu/qemu_monitor_text.c +++ b/src/qemu/qemu_monitor_text.c @@ -1064,7 +1064,7 @@ int qemuMonitorTextChangeMedia(qemuMonitorPtr mon, /* If the command failed qemu prints: * device not found, device is locked ... * No message is printed on success it seems */ - if (strstr(reply, "device ")) { + if (c_strcasestr(reply, "device ")) { qemuReportError(VIR_ERR_OPERATION_FAILED, _("could not change media on %s: %s"), devname, reply); goto cleanup; -- 1.7.1

2 4

[libvirt] [v3 00/14] USB improvements and new redirdev
by Marc-André Lureau 02 Sep '11

02 Sep '11

Hei, Here we go with the third version of this patch series that leads to support for various USB controllers, USB hubs, and USB redirection devices. Major change since v2 is the creation of a new element 'redirdev' instead of reusing the existing 'hostdev'. Marc-André Lureau (14): Add various USB devices QEMU_CAPS Split virDomainControllerModel to virDomainControllerModelSCSI Add USB controller models Add a new controller type 'usb' with optionnal 'model' USB controller can have a PCI address child element USB devices gain a new USB address child element Add USB companion controllers support Add USB hub device Modify USB port to be defined as a port path qemu: don't reserve slot 1 if a PIIX3 USB controller is defined there qemu: Don't append 0 at usb id, so that it is compatible with legacy -usb Add a usb1 & usb2 qemuxml2argv test Add "redirdev" redirection device Learn to use spicevmc as a redirection type for usb-redir docs/formatdomain.html.in | 105 +++++- docs/schemas/domain.rng | 100 ++++- src/conf/domain_audit.c | 65 +++ src/conf/domain_audit.h | 5 + src/conf/domain_conf.c | 443 +++++++++++++++++++- src/conf/domain_conf.h | 104 +++++- src/esx/esx_driver.c | 8 +- src/libvirt_private.syms | 11 +- src/qemu/qemu_capabilities.c | 28 ++ src/qemu/qemu_capabilities.h | 9 + src/qemu/qemu_command.c | 342 ++++++++++++++- src/qemu/qemu_command.h | 14 +- src/qemu/qemu_driver.c | 7 + src/qemu/qemu_hotplug.c | 55 +++- src/qemu/qemu_hotplug.h | 3 + src/vmx/vmx.c | 32 +- tests/qemuhelptest.c | 17 +- .../qemuxml2argv-input-usbmouse-addr.args | 1 + .../qemuxml2argv-input-usbmouse-addr.xml | 27 ++ .../qemuxml2argv-usb-controller.args | 1 + .../qemuxml2argv-usb-controller.xml | 16 + tests/qemuxml2argvdata/qemuxml2argv-usb-hub.args | 1 + tests/qemuxml2argvdata/qemuxml2argv-usb-hub.xml | 19 + .../qemuxml2argv-usb-ich9-companion.args | 6 + .../qemuxml2argv-usb-ich9-companion.xml | 30 ++ .../qemuxml2argv-usb-ich9-ehci-addr.args | 1 + .../qemuxml2argv-usb-ich9-ehci-addr.xml | 18 + .../qemuxml2argv-usb-piix3-controller.args | 1 + .../qemuxml2argv-usb-piix3-controller.xml | 16 + tests/qemuxml2argvdata/qemuxml2argv-usb-ports.args | 1 + tests/qemuxml2argvdata/qemuxml2argv-usb-ports.xml | 31 ++ tests/qemuxml2argvdata/qemuxml2argv-usb-redir.args | 10 + tests/qemuxml2argvdata/qemuxml2argv-usb-redir.xml | 40 ++ tests/qemuxml2argvdata/qemuxml2argv-usb1-usb2.args | 15 + tests/qemuxml2argvdata/qemuxml2argv-usb1-usb2.xml | 74 ++++ tests/qemuxml2argvtest.c | 30 ++ tests/qemuxml2xmltest.c | 2 + tests/xml2vmxtest.c | 2 +- 38 files changed, 1598 insertions(+), 92 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-input-usbmouse-addr.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-input-usbmouse-addr.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-controller.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-controller.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-hub.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-hub.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-ich9-companion.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-ich9-companion.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-ich9-ehci-addr.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-ich9-ehci-addr.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-piix3-controller.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-piix3-controller.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-ports.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-ports.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-redir.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb-redir.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb1-usb2.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-usb1-usb2.xml -- 1.7.6

5 26

[libvirt] [PATCH 0/8] Report disk latency info
by Osier Yang 02 Sep '11

02 Sep '11

This patch series introduces a new API to report the disk latency related information, which is supported by upstream QEMU just a few days ago (commit c488c7f649, Thu Aug 25). Per previous dicussion on the ABI compatiblity, and API design principle, the new API is defined with style: typedef struct _virDomainBlockStatsFlags virDomainBlockStatsFlagsStruct; typedef virDomainBlockStatsFlagsStruct *virDomainBlockStatsFlagsPtr; struct _virDomainBlockStatsFlags { char field[VIR_DOMAIN_BLOCK_STATS_FIELD_LENGTH]; long long value; }; int virDomainBlockStatsFlags (virDomainPtr dom, const char *path, virDomainBlockStatsFlagsPtr params, int *nparams, unsigned int flags) Other points: 1) With the new API, output of virsh command "domblkstat" is different, this might affect the existed scripts. But considering we are even introducing new feilds, so this can be fine? 2) qemuMonitorJSONGetBlockStatsInfo set "*errs = 0" before, it will cause "domblkstat" always print things like "vda errs 0". However, QEMU doesn't support this field. Fixed it in these patches (*errs = -1). And new API qemuDomainBlockStatsFlags won't even set a field for "errs". 3) Is it deserved to update gendispatch.pl to generate remote codes for functions take argument like "virNodeCPUStatsPtr params", "virNodeMemoryStatsPtr params"? All of these argument points to structs has same structure. Perhaps we can define an alias for these structs and generate the remote codes just like for "virTypedParameterPtr params". [PATCH 1/8] latency: Define new public API and structure [PATCH 2/8] latency: Define the internal driver callback [PATCH 3/8] latency: Implemente the public API [PATCH 4/8] latency: Wire up the remote protocol [PATCH 5/8] latency: Update monitor functions for new latency fields [PATCH 6/8] latency: Implemente internal API for qemu driver [PATCH 7/8] latency: Expose the new API for Python binding [PATCH 8/8] latency: Update cmdBlkStats to use new API Regards, Osier

3 24

[libvirt] [PATCH 0/3 v4] Add filesystem pool formatting
by Osier Yang 02 Sep '11

02 Sep '11

The following patches add the ability to format filesystem pools when the appropriate flags are passed to pool build. This patch set introduces two new flags: VIR_STORAGE_POOL_BUILD_NO_OVERWRITE causes the build to probe for an existing pool of the requested type. The build operation formats the filesystem if it does not find an existing filesystem of that type. VIR_STORAGE_POOL_BUILD_OVERWRITE causes the build to format unconditionally. This patch set is mainly based on v3 by Dave Allan. http://www.redhat.com/archives/libvir-list/2010-June/msg00042.html [PATCH 1/3] storage: Add mkfs and libblkid to build system [PATCH 2/3] storage: Add fs pool formatting [PATCH 3/3] storage: Add virsh support for fs pool formating Regards, Osier

2 9

[libvirt] Preparing for next release 0.9.5
by Daniel Veillard 02 Sep '11

02 Sep '11

To try to keep good habits and stay predictable I would suggest to enter development freeze this week-end to plan an upstream release on Friday 2nd Sep or early the following week. I may also switch the libvirt.org server to a new box at that time, but it's not really urgent, th new box would be less powerful, but with overall better network reliability. Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

1 1

[libvirt] [RFC] block I/O throttling: how to enable in libvirt
by Zhi Yong Wu 02 Sep '11

02 Sep '11

----- Forwarded message from Zhi Yong Wu <wuzhy(a)linux.vnet.ibm.com> ----- Date: Thu, 1 Sep 2011 11:55:17 +0800 From: Zhi Yong Wu <wuzhy(a)linux.vnet.ibm.com> To: Stefan Hajnoczi <stefanha(a)gmail.com> Cc: "Daniel P. Berrange" <berrange(a)redhat.com>, Stefan Hajnoczi <stefanha(a)gmail.com>, Adam Litke <agl(a)us.ibm.com>, Zhi Yong Wu <wuzhy(a)linux.vnet.ibm.com>, QEMU Developers <qemu-devel(a)nongnu.org>, guijianfeng(a)cn.fujitsu.com, hutao(a)cn.fujitsu.com Subject: [RFC] block I/O throttling: how to enable in libvirt Message-ID: <20110901035517.GD16985(a)f15.cn.ibm.com> References: <CAEH94Li_C=BOe2gV8NyM48njYWMBAo9MTGc1eUOh-Y=cODs6VA(a)mail.gmail.com> <CAJSP0QW1CPCokX=F5z7y==vn1S4wH0VtOaQ7oj4kC7f7uQM4MQ(a)mail.gmail.com> <20110830134636.GB29130(a)aglitke.rchland.ibm.com> <CAJSP0QUHm=y8XJC_KXRg7ufFZt3K_XDDfQb--sxjC+c0GjO8qg(a)mail.gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <CAJSP0QUHm=y8XJC_KXRg7ufFZt3K_XDDfQb--sxjC+c0GjO8qg(a)mail.gmail.com> User-Agent: Mutt/1.5.21 (2010-09-15) X-Xagent-From: wuzhy(a)linux.vnet.ibm.com X-Xagent-To: wuzhy(a)linux.vnet.ibm.com X-Xagent-Gateway: vmsdvma.vnet.ibm.com (XAGENTU7 at VMSDVMA) On Wed, Aug 31, 2011 at 08:18:19AM +0100, Stefan Hajnoczi wrote: >Subject: Re: The design choice for how to enable block I/O throttling > function in libvirt >From: Stefan Hajnoczi <stefanha(a)gmail.com> >To: Adam Litke <agl(a)us.ibm.com> >Cc: libvir-list(a)redhat.com, "Daniel P. Berrange" <berrange(a)redhat.com>, Zhi > Yong Wu <wuzhy(a)linux.vnet.ibm.com>, Zhi Yong Wu <zwu.kernel(a)gmail.com> >Content-Type: text/plain; charset=ISO-8859-1 >Content-Transfer-Encoding: quoted-printable >X-Brightmail-Tracker: AAAAAA== >X-Xagent-From: stefanha(a)gmail.com >X-Xagent-To: wuzhy(a)linux.vnet.ibm.com >X-Xagent-Gateway: bldgate.vnet.ibm.com (XAGENTU7 at BLDGATE) > >On Tue, Aug 30, 2011 at 2:46 PM, Adam Litke <agl(a)us.ibm.com> wrote: >> On Tue, Aug 30, 2011 at 09:53:33AM +0100, Stefan Hajnoczi wrote: >>> On Tue, Aug 30, 2011 at 3:55 AM, Zhi Yong Wu <zwu.kernel(a)gmail.com> wrote: >>> > I am trying to enable block I/O throttling function in libvirt. But >>> > currently i met some design questions, and don't make sure if we >>> > should extend blkiotune to support block I/O throttling or introduce >>> > one new libvirt command "blkiothrottle" to cover it or not. If you >>> > have some better idea, pls don't hesitate to drop your comments. >>> >>> A little bit of context: this discussion is about adding libvirt >>> support for QEMU disk I/O throttling. >> >> Thanks for the additional context Stefan. >> >>> Today libvirt supports the cgroups blkio-controller, which handles >>> proportional shares and throughput/iops limits on host block devices. >>> blkio-controller does not support network file systems (NFS) or other >>> QEMU remote block drivers (curl, Ceph/rbd, sheepdog) since they are >>> not host block devices. QEMU I/O throttling works with all types of >>> -drive and therefore complements blkio-controller. >> >> The first question that pops into my mind is: Should a user need to understand >> when to use the cgroups blkio-controller vs. the QEMU I/O throttling method? In >> my opinion, it would be nice if libvirt had a single interface for block I/O >> throttling and libvirt would decide which mechanism to use based on the type of >> device and the specific limits that need to be set. > >Yes, I agree it would be simplest to pick the right mechanism, >depending on the type of throttling the user wants. More below. > >>> I/O throttling can be applied independently to each -drive attached to >>> a guest and supports throughput/iops limits. For more information on >>> this QEMU feature and a comparison with blkio-controller, see Ryan >>> Harper's KVM Forum 2011 presentation: >> >>> http://www.linux-kvm.org/wiki/images/7/72/2011-forum-keep-a-limit-on-it-io-… >> >> From the presentation, it seems that both the cgroups method the the qemu method >> offer comparable control (assuming a block device) so it might possible to apply >> either method from the same API in a transparent manner. Am I correct or are we >> suggesting that the Qemu throttling approach should always be used for Qemu >> domains? > >QEMU I/O throttling does not provide a proportional share mechanism. >So you cannot assign weights to VMs and let them receive a fraction of >the available disk time. That is only supported by cgroups >blkio-controller because it requires a global view which QEMU does not >have. > >So I think the two are complementary: > >If proportional share should be used on a host block device, use >cgroups blkio-controller. >Otherwise use QEMU I/O throttling. Stefan, Do you agree with introducing one new libvirt command blkiothrottle now? If so, i will work on the code draft to make it work. Daniel and other maintainers, If you are available, can you make some comments for us?:) Regards, Zhi Yong Wu > >Stefan ----- End forwarded message -----

4 10

[libvirt] [test-API][PATCH] Add libvirtd restart test case
by Wayne Sun 02 Sep '11

02 Sep '11

* libvirtd restart should not affect the running domains. This test check the libvirtd status before and after libvirtd restart, and also by checking the domain pid to confirm the domain is not affected. --- repos/libvirtd/restart.py | 143 +++++++++++++++++++++++++++++++++++++++++++++ 1 files changed, 143 insertions(+), 0 deletions(-) create mode 100644 repos/libvirtd/restart.py diff --git a/repos/libvirtd/restart.py b/repos/libvirtd/restart.py new file mode 100644 index 0000000..15dd43c --- /dev/null +++ b/repos/libvirtd/restart.py @@ -0,0 +1,143 @@ +#!/usr/bin/evn python +""" Restart libvirtd testing. A running guest is required in this test. + During libvirtd restart, the guest remains running and not affected + by libvirtd restart. + libvirtd:restart + guestname + #GUESTNAME# +""" + +__author__ = 'Wayne Sun: gsun(a)redhat.com' +__date__ = 'Thu Aug 4, 2011' +__version__ = '0.1.0' +__credits__ = 'Copyright (C) 2011 Red Hat, Inc.' +__all__ = ['restart'] + +import os +import re +import sys +import time + +def append_path(path): + """Append root path of package""" + if path not in sys.path: + sys.path.append(path) + +pwd = os.getcwd() +result = re.search('(.*)libvirt-test-API', pwd) +append_path(result.group(0)) + +from lib import connectAPI +from lib import domainAPI +from utils.Python import utils + +VIRSH_LIST = "virsh list --all" +RESTART_CMD = "service libvirtd restart" + +def check_params(params): + """Verify inputing parameter dictionary""" + logger = params['logger'] + keys = ['guestname'] + for key in keys: + if key not in params: + logger.error("%s is required" %key) + return 1 + return 0 + +def libvirtd_check(util, logger): + """check libvirtd status + """ + cmd = "service libvirtd status" + ret, out = util.exec_cmd(cmd, shell=True) + if ret != 0: + logger.error("failed to get libvirtd status") + return 1 + else: + logger.info(out[0]) + + logger.info(VIRSH_LIST) + ret, out = util.exec_cmd(VIRSH_LIST, shell=True) + if ret != 0: + logger.error("failed to get virsh list result") + return 1 + else: + for i in range(len(out)): + logger.info(out[i]) + + return 0 + +def get_domain_pid(util, logger, guestname): + """get the pid of running domain""" + logger.info("get the pid of running domain %s" % guestname) + get_pid_cmd = "cat /var/run/libvirt/qemu/%s.pid" % guestname + ret, pid = util.exec_cmd(get_pid_cmd, shell=True) + if ret: + logger.error("fail to get the pid of runnings domain %s" % \ + guestname) + return 1, "" + else: + logger.info("the pid of domain %s is %s" % \ + (guestname, pid[0])) + return 0, pid[0] + +def restart(params): + """restart libvirtd test""" + # Initiate and check parameters + params_check_result = check_params(params) + if params_check_result: + return 1 + + logger = params['logger'] + guestname = params['guestname'] + util = utils.Utils() + uri = util.get_uri('127.0.0.1') + + conn = connectAPI.ConnectAPI() + virconn = conn.open(uri) + domobj = domainAPI.DomainAPI(virconn) + state = domobj.get_state(guestname) + conn.close() + + if(state == "shutoff"): + logger.info("guest is shutoff, if u want to run this case, \ + guest must be running") + return 1 + + logger.info("check the libvirtd status:") + result = libvirtd_check(util, logger) + if result: + return 1 + + ret, pid_before = get_domain_pid(util, logger, guestname) + if ret: + return 1 + + logger.info("restart libvirtd service:") + ret, out = util.exec_cmd(RESTART_CMD, shell=True) + if ret != 0: + logger.error("failed to restart libvirtd") + for i in range(len(out)): + logger.error(out[i]) + return 1 + else: + for i in range(len(out)): + logger.info(out[i]) + + logger.info("recheck libvirtd status:") + result = libvirtd_check(util, logger) + if result: + return 1 + + ret, pid_after = get_domain_pid(util, logger, guestname) + if ret: + return 1 + + if pid_before != pid_after: + logger.error("%s pid changed during libvirtd restart" % \ + guestname) + return 1 + else: + logger.info("domain pid not change, %s keeps running during \ + libvirtd restart" % guestname) + + return 0 -- 1.7.1

3 2