- Devel - libvirt List Archives

[libvirt] [RFC PATCH] set tap mtu
by Chris Wright 18 Nov '08

18 Nov '08

Below is a simple PoC for setting a large MTU size on a tap device. With this we are able to improve net i/o throughput substantially (~40% improvement on TX and ~130% improvement on RX). This is just RFC because it's hardcoded to an MTU of 9000 for any tap device. Thoughts on the best way to add this kind of support? diff -up libvirt-0.4.6/src/bridge.c~tap-mtu libvirt-0.4.6/src/bridge.c --- libvirt-0.4.6/src/bridge.c~tap-mtu 2008-08-29 05:19:52.000000000 -0700 +++ libvirt-0.4.6/src/bridge.c 2008-11-17 18:37:56.000000000 -0800 @@ -276,6 +276,38 @@ brDeleteInterface(brControl *ctl ATTRIBU #endif /** + * ifSetMtu: + * @ctl: bridge control pointer + * @ifname: interface name to set MTU for + * @mtu: MTU value + * + * This function sets the @mtu for a given interface @ifname. Typically + * used on a tap device to set up for Jumbo Frames. + * + * Returns 0 in case of success or an errno code in case of failure. + */ +static int +ifSetMtu(brControl *ctl, const char *ifname, int mtu) +{ + struct ifreq ifr; + int len; + + if (!ctl || !ifname) + return EINVAL; + + if ((len = strlen(ifname)) >= BR_IFNAME_MAXLEN) + return EINVAL; + + memset(&ifr, 0, sizeof(struct ifreq)); + + strncpy(ifr.ifr_name, ifname, len); + ifr.ifr_name[len] = '\0'; + ifr.ifr_mtu = mtu; + + return ioctl(ctl->fd, SIOCSIFMTU, &ifr) == 0 ? 0 : errno; +} + +/** * brAddTap: * @ctl: bridge control pointer * @bridge: the bridge name @@ -334,6 +366,8 @@ brAddTap(brControl *ctl, } if (ioctl(fd, TUNSETIFF, &try) == 0) { + if ((errno = ifSetMtu(ctl, try.ifr_name, 9000))) + goto error; if ((errno = brAddInterface(ctl, bridge, try.ifr_name))) goto error; if ((errno = brSetInterfaceUp(ctl, try.ifr_name, 1)))

4 3

[libvirt] [PATCH 2/2] Java bindings for domain events
by David Lively 18 Nov '08

18 Nov '08

This patch allows the remote driver to work with an asynchronous EventImpl (it's the only one using an externally-supplied one), assuming libvirt is compiled with pthread support. (Without pthreads, this code is harmless in a single-threaded environment.) Basically it uses a mutex to protect reads from the RPC socket in such a way that message reads (in their entirety) are done atomically (otherwise the remoteDomainEventFired() can race the call() code that reads replies & events). In addition, I update the EventImpl handle to prevent remoteDomainEventFired() from being called everytime a reply is sent. (This helps us dispatch events in a timely manner, though it's not strictly necessary. Without it, any events coming in during a call() won't be dispatched until the call drops the socket lock (because remoteDomainEventFired() will be stuck awaiting the lock). Dave

2 2

[libvirt] ANNOUNCE: Ruby bindings for libvirt, version 0.1.0
by David Lutterkort 18 Nov '08

18 Nov '08

I am pleased to announce the release of version 0.1.0 of the ruby-libvirt bindings. NEWS: * Add binding for virConnectFindStoragePoolSources (clalance) * Fix dom_migrate (clalance) * Add the MIGRATE_LIVE (enum virDomainMigrateFlags) flag * Slight improvements of the unit tests The main site for the bindings is http://libvirt.org/ruby API docs can be found at http://libvirt.org/ruby/api/index.html Tarballs/src RPM's are at http://libvirt.org/ruby/download Packages for Fedora and EPEL5 are in the works and should show up on a friendly mirror near you soon. David

1 0

[libvirt] PATCH: Support events for define/undefine in QEMU
by Daniel P. Berrange 17 Nov '08

17 Nov '08

This patch adds support for the DEFINED and UNDEFINED events in the QEMU drive. This was more involved than I expected, because when the daemon gets SIGHUP it re-loads config files and we need to broadcast events for each new config file it finds. So I had to add a callback to the internal virDomainLoadAllConfigs method. The LoadConfig method also had a bogus line of code resetting the domain state to SHUTOFF, which made all your active VMs suddenlly appear inactive, after the daemon got SIGHUP ! Daniel diff --git a/src/domain_conf.c b/src/domain_conf.c --- a/src/domain_conf.c +++ b/src/domain_conf.c @@ -3242,18 +3242,20 @@ virDomainObjPtr virDomainLoadConfig(virC virDomainObjListPtr doms, const char *configDir, const char *autostartDir, - const char *name) + const char *name, + virDomainLoadConfigNotify notify, + void *opaque) { char *configFile = NULL, *autostartLink = NULL; virDomainDefPtr def = NULL; virDomainObjPtr dom; int autostart; + int newVM = 1; if ((configFile = virDomainConfigFile(conn, configDir, name)) == NULL) goto error; if ((autostartLink = virDomainConfigFile(conn, autostartDir, name)) == NULL) goto error; - if ((autostart = virFileLinkPointsTo(autostartLink, configFile)) < 0) goto error; @@ -3261,11 +3263,16 @@ virDomainObjPtr virDomainLoadConfig(virC if (!(def = virDomainDefParseFile(conn, caps, configFile))) goto error; + if (virDomainFindByName(doms, def->name)) + newVM = 0; + if (!(dom = virDomainAssignDef(conn, doms, def))) goto error; - dom->state = VIR_DOMAIN_SHUTOFF; dom->autostart = autostart; + + if (notify) + (*notify)(dom, newVM, opaque); return dom; @@ -3280,7 +3287,9 @@ int virDomainLoadAllConfigs(virConnectPt virCapsPtr caps, virDomainObjListPtr doms, const char *configDir, - const char *autostartDir) + const char *autostartDir, + virDomainLoadConfigNotify notify, + void *opaque) { DIR *dir; struct dirent *entry; @@ -3310,7 +3319,9 @@ int virDomainLoadAllConfigs(virConnectPt doms, configDir, autostartDir, - entry->d_name); + entry->d_name, + notify, + opaque); if (dom) dom->persistent = 1; } diff --git a/src/domain_conf.h b/src/domain_conf.h --- a/src/domain_conf.h +++ b/src/domain_conf.h @@ -549,18 +549,26 @@ int virDomainSaveConfig(virConnectPtr co const char *configDir, virDomainDefPtr def); +typedef void (*virDomainLoadConfigNotify)(virDomainObjPtr dom, + int newDomain, + void *opaque); + virDomainObjPtr virDomainLoadConfig(virConnectPtr conn, virCapsPtr caps, virDomainObjListPtr doms, const char *configDir, const char *autostartDir, - const char *name); + const char *name, + virDomainLoadConfigNotify notify, + void *opaque); int virDomainLoadAllConfigs(virConnectPtr conn, virCapsPtr caps, virDomainObjListPtr doms, const char *configDir, - const char *autostartDir); + const char *autostartDir, + virDomainLoadConfigNotify notify, + void *opaque); int virDomainDeleteConfig(virConnectPtr conn, const char *configDir, diff --git a/src/lxc_driver.c b/src/lxc_driver.c --- a/src/lxc_driver.c +++ b/src/lxc_driver.c @@ -1017,7 +1017,8 @@ static int lxcStartup(void) lxc_driver->caps, &lxc_driver->domains, lxc_driver->configDir, - lxc_driver->autostartDir) < 0) { + lxc_driver->autostartDir, + NULL, NULL) < 0) { lxcShutdown(); return -1; } diff --git a/src/qemu_driver.c b/src/qemu_driver.c --- a/src/qemu_driver.c +++ b/src/qemu_driver.c @@ -226,7 +226,8 @@ qemudStartup(void) { qemu_driver->caps, &qemu_driver->domains, qemu_driver->configDir, - qemu_driver->autostartDir) < 0) { + qemu_driver->autostartDir, + NULL, NULL) < 0) { qemudShutdown(); return -1; } @@ -241,6 +242,16 @@ qemudStartup(void) { VIR_FREE(base); VIR_FREE(qemu_driver); return -1; +} + +static void qemudNotifyLoadDomain(virDomainObjPtr vm, int newVM, void *opaque) +{ + struct qemud_driver *driver = opaque; + + if (newVM) + qemudDomainEventDispatch(driver, vm, + VIR_DOMAIN_EVENT_DEFINED, + VIR_DOMAIN_EVENT_DEFINED_ADDED); } /** @@ -258,7 +269,8 @@ qemudReload(void) { qemu_driver->caps, &qemu_driver->domains, qemu_driver->configDir, - qemu_driver->autostartDir); + qemu_driver->autostartDir, + qemudNotifyLoadDomain, qemu_driver); qemudAutostartConfigs(qemu_driver); @@ -2356,9 +2368,14 @@ static virDomainPtr qemudDomainDefine(vi virDomainDefPtr def; virDomainObjPtr vm; virDomainPtr dom; + int newVM = 1; if (!(def = virDomainDefParseString(conn, driver->caps, xml))) return NULL; + + vm = virDomainFindByName(&driver->domains, def->name); + if (vm) + newVM = 0; if (!(vm = virDomainAssignDef(conn, &driver->domains, @@ -2375,6 +2392,12 @@ static virDomainPtr qemudDomainDefine(vi vm); return NULL; } + + qemudDomainEventDispatch(driver, vm, + VIR_DOMAIN_EVENT_DEFINED, + newVM ? + VIR_DOMAIN_EVENT_DEFINED_ADDED : + VIR_DOMAIN_EVENT_DEFINED_UPDATED); dom = virGetDomain(conn, vm->def->name, vm->def->uuid); if (dom) dom->id = vm->def->id; @@ -2405,6 +2428,8 @@ static int qemudDomainUndefine(virDomain if (virDomainDeleteConfig(dom->conn, driver->configDir, driver->autostartDir, vm) < 0) return -1; + + qemudDomainEventDispatch(driver, vm, VIR_DOMAIN_EVENT_UNDEFINED, 0); virDomainRemoveInactive(&driver->domains, vm); -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 2

[libvirt] PATCH: Add domain event "detail" information
by Daniel P. Berrange 17 Nov '08

17 Nov '08

As per our earlier discussion today, this patch expands the callback for domain events so that it also gets a event type specific 'detail' field. This is also kept as an int, and we define enumerations for the possible values associated with each type. If a event type has no detail, 0 is passed. The RESTORED and SAVED event types disappear in this patch and just become another piece of 'detail' to the STOPPED and STARTED events. I have also renamed ADDED & REMOVED to DEFINED and UNDEFINED to match terminology we have elsewhere & because the names were confusing me Easiest to just look at the final set: typedef enum { VIR_DOMAIN_EVENT_DEFINED = 0, VIR_DOMAIN_EVENT_UNDEFINED = 1, VIR_DOMAIN_EVENT_STARTED = 2, VIR_DOMAIN_EVENT_SUSPENDED = 3, VIR_DOMAIN_EVENT_RESUMED = 4, VIR_DOMAIN_EVENT_STOPPED = 5, } virDomainEventType; typedef enum { VIR_DOMAIN_EVENT_STARTED_BOOTED = 0, /* Normal startup from boot */ VIR_DOMAIN_EVENT_STARTED_MIGRATED = 1, /* Incoming migration from another host */ VIR_DOMAIN_EVENT_STARTED_RESTORED = 2, /* Restored from a state file */ } virDomainEventStartedDetailType; typedef enum { VIR_DOMAIN_EVENT_STOPPED_SHUTDOWN = 0, /* Normal shutdown */ VIR_DOMAIN_EVENT_STOPPED_DESTROYED = 1, /* Forced poweroff from host */ VIR_DOMAIN_EVENT_STOPPED_CRASHED = 2, /* Guest crashed */ VIR_DOMAIN_EVENT_STOPPED_MIGRATED = 3, /* Migrated off to another host */ VIR_DOMAIN_EVENT_STOPPED_SAVED = 4, /* Saved to a state file */ VIR_DOMAIN_EVENT_STOPPED_FAILED = 5, /* Host emulator/mgmt failed */ } virDomainEventStoppedDetailType; typedef enum { VIR_DOMAIN_EVENT_DEFINED_ADDED = 0, /* Newly created config file */ VIR_DOMAIN_EVENT_DEFINED_UPDATED = 1, /* Changed config file */ } virDomainEventDefinedDetailType; I don't make use of 'CRASHED' in QEMU driver yet. It might be useful in Xen though - when a PV guest crashes, Xen stops the domain running, but leaves it there in a shutoff state, but marked as crashed. Now using the C event-test program you can see the effects: myDomainEventCallback1 EVENT: Domain F9x86_64(2) Started Booted myDomainEventCallback2 EVENT: Domain F9x86_64(2) Started Booted myDomainEventCallback1 EVENT: Domain F9x86_64(-1) Stopped Destroyed myDomainEventCallback2 EVENT: Domain F9x86_64(-1) Stopped Destroyed myDomainEventCallback1 EVENT: Domain F9x86_64(3) Started Booted myDomainEventCallback2 EVENT: Domain F9x86_64(3) Started Booted myDomainEventCallback1 EVENT: Domain F9x86_64(3) Suspended myDomainEventCallback2 EVENT: Domain F9x86_64(3) Suspended myDomainEventCallback1 EVENT: Domain F9x86_64(3) Resumed myDomainEventCallback2 EVENT: Domain F9x86_64(3) Resumed myDomainEventCallback1 EVENT: Domain F9x86_64(-1) Stopped Shutdown myDomainEventCallback2 EVENT: Domain F9x86_64(-1) Stopped Shutdown Of the following sequence of actions virsh start F9x86_64 virsh destroy F9x86_64 virsh start F9x86_64 virsh suspend F9x86_64 virsh resume F9x86_64 virsh shutdown F9x86_64 For the last 'shutdown' operation, you'll see the same if you just run a graceful shutdown inside the guest itself. NB, I've not tested saved/restored because my install of KVM is not new enough to support that correctly, but I expect it to work without trouble. Likewise for migration. A word about migration... - The destination host first gets a STARTED event, with detail MIGRATED when it starts running - The source host then gets a STOPPED event with detail MIGRATED when it completes - The destination host then gets a RESUMED event, on success, and a STOPPED event with detail FAILED if migration aborts. Daniel examples/domain-events/events-c/event-test.c | 78 +++++++++++++++---- examples/domain-events/events-python/event-test.py | 8 +- include/libvirt/libvirt.h | 29 ++++++- include/libvirt/libvirt.h.in | 29 ++++++- python/libvir.c | 6 + qemud/qemud.h | 1 qemud/remote.c | 22 +++-- qemud/remote_protocol.c | 2 qemud/remote_protocol.h | 1 qemud/remote_protocol.x | 1 src/domain_event.c | 4 - src/domain_event.h | 6 + src/qemu_driver.c | 82 +++++++++++++++------ src/remote_internal.c | 29 ++++--- 18 files changed, 366 insertions(+), 118 deletions(-) diff --git a/examples/domain-events/events-c/event-test.c b/examples/domain-events/events-c/event-test.c --- a/examples/domain-events/events-c/event-test.c +++ b/examples/domain-events/events-c/event-test.c @@ -35,9 +35,9 @@ void *t_opaque = NULL; /* Prototypes */ const char *eventToString(int event); int myDomainEventCallback1 (virConnectPtr conn, virDomainPtr dom, - int event, void *opaque); + int event, int detail, void *opaque); int myDomainEventCallback2 (virConnectPtr conn, virDomainPtr dom, - int event, void *opaque); + int event, int detail, void *opaque); int myEventAddHandleFunc (int fd, int event, virEventHandleCallback cb, void *opaque); void myEventUpdateHandleFunc(int fd, int event); @@ -58,11 +58,11 @@ const char *eventToString(int event) { const char *eventToString(int event) { const char *ret = NULL; switch(event) { - case VIR_DOMAIN_EVENT_ADDED: - ret ="Added"; + case VIR_DOMAIN_EVENT_DEFINED: + ret ="Defined"; break; - case VIR_DOMAIN_EVENT_REMOVED: - ret ="Removed"; + case VIR_DOMAIN_EVENT_UNDEFINED: + ret ="Undefined"; break; case VIR_DOMAIN_EVENT_STARTED: ret ="Started"; @@ -76,14 +76,56 @@ const char *eventToString(int event) { case VIR_DOMAIN_EVENT_STOPPED: ret ="Stopped"; break; - case VIR_DOMAIN_EVENT_SAVED: - ret ="Saved"; - break; - case VIR_DOMAIN_EVENT_RESTORED: - ret ="Restored"; - break; default: ret ="Unknown Event"; + } + return ret; +} + +static const char *eventDetailToString(int event, int detail) { + const char *ret = ""; + switch(event) { + case VIR_DOMAIN_EVENT_DEFINED: + if (detail == VIR_DOMAIN_EVENT_DEFINED_ADDED) + ret = "Added"; + else if (detail == VIR_DOMAIN_EVENT_DEFINED_UPDATED) + ret = "Updated"; + break; + case VIR_DOMAIN_EVENT_STARTED: + switch (detail) { + case VIR_DOMAIN_EVENT_STARTED_BOOTED: + ret = "Booted"; + break; + case VIR_DOMAIN_EVENT_STARTED_MIGRATED: + ret = "Migrated"; + break; + case VIR_DOMAIN_EVENT_STARTED_RESTORED: + ret = "Restored"; + break; + } + break; + case VIR_DOMAIN_EVENT_STOPPED: + switch (detail) { + case VIR_DOMAIN_EVENT_STOPPED_SHUTDOWN: + ret = "Shutdown"; + break; + case VIR_DOMAIN_EVENT_STOPPED_DESTROYED: + ret = "Destroyed"; + break; + case VIR_DOMAIN_EVENT_STOPPED_CRASHED: + ret = "Crashed"; + break; + case VIR_DOMAIN_EVENT_STOPPED_MIGRATED: + ret = "Migrated"; + break; + case VIR_DOMAIN_EVENT_STOPPED_SAVED: + ret = "Failed"; + break; + case VIR_DOMAIN_EVENT_STOPPED_FAILED: + ret = "Failed"; + break; + } + break; } return ret; } @@ -91,20 +133,24 @@ int myDomainEventCallback1 (virConnectPt int myDomainEventCallback1 (virConnectPtr conn ATTRIBUTE_UNUSED, virDomainPtr dom, int event, + int detail, void *opaque ATTRIBUTE_UNUSED) { - printf("%s EVENT: Domain %s(%d) %s\n", __FUNCTION__, virDomainGetName(dom), - virDomainGetID(dom), eventToString(event)); + printf("%s EVENT: Domain %s(%d) %s %s\n", __FUNCTION__, virDomainGetName(dom), + virDomainGetID(dom), eventToString(event), + eventDetailToString(event, detail)); return 0; } int myDomainEventCallback2 (virConnectPtr conn ATTRIBUTE_UNUSED, virDomainPtr dom, int event, + int detail, void *opaque ATTRIBUTE_UNUSED) { - printf("%s EVENT: Domain %s(%d) %s\n", __FUNCTION__, virDomainGetName(dom), - virDomainGetID(dom), eventToString(event)); + printf("%s EVENT: Domain %s(%d) %s %s\n", __FUNCTION__, virDomainGetName(dom), + virDomainGetID(dom), eventToString(event), + eventDetailToString(event, detail)); return 0; } diff --git a/examples/domain-events/events-python/event-test.py b/examples/domain-events/events-python/event-test.py --- a/examples/domain-events/events-python/event-test.py +++ b/examples/domain-events/events-python/event-test.py @@ -32,11 +32,11 @@ def eventToString(event): "Restored" ); return eventStrings[event]; -def myDomainEventCallback1 (conn, dom, event, opaque): - print "myDomainEventCallback1 EVENT: Domain %s(%s) %s" % (dom.name(), dom.ID(), eventToString(event)) +def myDomainEventCallback1 (conn, dom, event, detail, opaque): + print "myDomainEventCallback1 EVENT: Domain %s(%s) %s %d" % (dom.name(), dom.ID(), eventToString(event), detail) -def myDomainEventCallback2 (conn, dom, event, opaque): - print "myDomainEventCallback2 EVENT: Domain %s(%s) %s" % (dom.name(), dom.ID(), eventToString(event)) +def myDomainEventCallback2 (conn, dom, event, detail, opaque): + print "myDomainEventCallback2 EVENT: Domain %s(%s) %s %d" % (dom.name(), dom.ID(), eventToString(event), detail) ##################################################### # EventImpl Functions diff --git a/include/libvirt/libvirt.h b/include/libvirt/libvirt.h --- a/include/libvirt/libvirt.h +++ b/include/libvirt/libvirt.h @@ -1004,21 +1004,41 @@ virDomainPtr virDomainCreateL * a virDomainEventType is emitted during domain lifecycle events */ typedef enum { - VIR_DOMAIN_EVENT_ADDED = 0, - VIR_DOMAIN_EVENT_REMOVED = 1, + VIR_DOMAIN_EVENT_DEFINED = 0, + VIR_DOMAIN_EVENT_UNDEFINED = 1, VIR_DOMAIN_EVENT_STARTED = 2, VIR_DOMAIN_EVENT_SUSPENDED = 3, VIR_DOMAIN_EVENT_RESUMED = 4, VIR_DOMAIN_EVENT_STOPPED = 5, - VIR_DOMAIN_EVENT_SAVED = 6, - VIR_DOMAIN_EVENT_RESTORED = 7, } virDomainEventType; + +typedef enum { + VIR_DOMAIN_EVENT_STARTED_BOOTED = 0, /* Normal startup from boot */ + VIR_DOMAIN_EVENT_STARTED_MIGRATED = 1, /* Incoming migration from another host */ + VIR_DOMAIN_EVENT_STARTED_RESTORED = 2, /* Restored from a state file */ +} virDomainEventStartedDetailType; + +typedef enum { + VIR_DOMAIN_EVENT_STOPPED_SHUTDOWN = 0, /* Normal shutdown */ + VIR_DOMAIN_EVENT_STOPPED_DESTROYED = 1, /* Forced poweroff from host */ + VIR_DOMAIN_EVENT_STOPPED_CRASHED = 2, /* Guest crashed */ + VIR_DOMAIN_EVENT_STOPPED_MIGRATED = 3, /* Migrated off to another host */ + VIR_DOMAIN_EVENT_STOPPED_SAVED = 4, /* Saved to a state file */ + VIR_DOMAIN_EVENT_STOPPED_FAILED = 5, /* Host emulator/mgmt failed */ +} virDomainEventStoppedDetailType; + +typedef enum { + VIR_DOMAIN_EVENT_DEFINED_ADDED = 0, /* Newly created config file */ + VIR_DOMAIN_EVENT_DEFINED_UPDATED = 1, /* Changed config file */ +} virDomainEventDefinedDetailType; + /** * virConnectDomainEventCallback: * @conn: virConnect connection * @dom: The domain on which the event occured * @event: The specfic virDomainEventType which occured + * @detail: event specific detail information * @opaque: opaque user data * * A callback function to be registered, and called when a domain event occurs @@ -1026,6 +1046,7 @@ typedef int (*virConnectDomainEventCallb typedef int (*virConnectDomainEventCallback)(virConnectPtr conn, virDomainPtr dom, int event, + int detail, void *opaque); int virConnectDomainEventRegister(virConnectPtr conn, diff --git a/include/libvirt/libvirt.h.in b/include/libvirt/libvirt.h.in --- a/include/libvirt/libvirt.h.in +++ b/include/libvirt/libvirt.h.in @@ -1004,21 +1004,41 @@ virDomainPtr virDomainCreateL * a virDomainEventType is emitted during domain lifecycle events */ typedef enum { - VIR_DOMAIN_EVENT_ADDED = 0, - VIR_DOMAIN_EVENT_REMOVED = 1, + VIR_DOMAIN_EVENT_DEFINED = 0, + VIR_DOMAIN_EVENT_UNDEFINED = 1, VIR_DOMAIN_EVENT_STARTED = 2, VIR_DOMAIN_EVENT_SUSPENDED = 3, VIR_DOMAIN_EVENT_RESUMED = 4, VIR_DOMAIN_EVENT_STOPPED = 5, - VIR_DOMAIN_EVENT_SAVED = 6, - VIR_DOMAIN_EVENT_RESTORED = 7, } virDomainEventType; + +typedef enum { + VIR_DOMAIN_EVENT_STARTED_BOOTED = 0, /* Normal startup from boot */ + VIR_DOMAIN_EVENT_STARTED_MIGRATED = 1, /* Incoming migration from another host */ + VIR_DOMAIN_EVENT_STARTED_RESTORED = 2, /* Restored from a state file */ +} virDomainEventStartedDetailType; + +typedef enum { + VIR_DOMAIN_EVENT_STOPPED_SHUTDOWN = 0, /* Normal shutdown */ + VIR_DOMAIN_EVENT_STOPPED_DESTROYED = 1, /* Forced poweroff from host */ + VIR_DOMAIN_EVENT_STOPPED_CRASHED = 2, /* Guest crashed */ + VIR_DOMAIN_EVENT_STOPPED_MIGRATED = 3, /* Migrated off to another host */ + VIR_DOMAIN_EVENT_STOPPED_SAVED = 4, /* Saved to a state file */ + VIR_DOMAIN_EVENT_STOPPED_FAILED = 5, /* Host emulator/mgmt failed */ +} virDomainEventStoppedDetailType; + +typedef enum { + VIR_DOMAIN_EVENT_DEFINED_ADDED = 0, /* Newly created config file */ + VIR_DOMAIN_EVENT_DEFINED_UPDATED = 1, /* Changed config file */ +} virDomainEventDefinedDetailType; + /** * virConnectDomainEventCallback: * @conn: virConnect connection * @dom: The domain on which the event occured * @event: The specfic virDomainEventType which occured + * @detail: event specific detail information * @opaque: opaque user data * * A callback function to be registered, and called when a domain event occurs @@ -1026,6 +1046,7 @@ typedef int (*virConnectDomainEventCallb typedef int (*virConnectDomainEventCallback)(virConnectPtr conn, virDomainPtr dom, int event, + int detail, void *opaque); int virConnectDomainEventRegister(virConnectPtr conn, diff --git a/python/libvir.c b/python/libvir.c --- a/python/libvir.c +++ b/python/libvir.c @@ -1538,6 +1538,7 @@ libvirt_virConnectDomainEventCallback(vi libvirt_virConnectDomainEventCallback(virConnectPtr conn ATTRIBUTE_UNUSED, virDomainPtr dom, int event, + int detail, void *opaque) { PyObject *pyobj_ret; @@ -1595,9 +1596,10 @@ libvirt_virConnectDomainEventCallback(vi /* Call the Callback Dispatcher */ pyobj_ret = PyObject_CallMethod(pyobj_conn_inst, (char*)"dispatchDomainEventCallbacks", - (char*)"Oi", + (char*)"Oii", pyobj_dom_inst, - event); + event, + detail); Py_DECREF(pyobj_dom_inst); diff --git a/qemud/qemud.h b/qemud/qemud.h --- a/qemud/qemud.h +++ b/qemud/qemud.h @@ -190,6 +190,7 @@ int remoteRelayDomainEvent (virConnectPt int remoteRelayDomainEvent (virConnectPtr conn ATTRIBUTE_UNUSED, virDomainPtr dom, int event, + int detail, void *opaque); #endif diff --git a/qemud/remote.c b/qemud/remote.c --- a/qemud/remote.c +++ b/qemud/remote.c @@ -80,7 +80,8 @@ static void static void remoteDispatchDomainEventSend (struct qemud_client *client, virDomainPtr dom, - virDomainEventType event); + int event, + int detail); /* This function gets called from qemud when it detects an incoming * remote protocol message. At this point, client->buffer contains @@ -413,15 +414,16 @@ remoteDispatchError (struct qemud_client } int remoteRelayDomainEvent (virConnectPtr conn ATTRIBUTE_UNUSED, - virDomainPtr dom, - int event, - void *opaque) + virDomainPtr dom, + int event, + int detail, + void *opaque) { struct qemud_client *client = opaque; - REMOTE_DEBUG("Relaying domain event %d", event); + REMOTE_DEBUG("Relaying domain event %d %d", event, detail); if(client) { - remoteDispatchDomainEventSend (client, dom, event); + remoteDispatchDomainEventSend (client, dom, event, detail); qemudDispatchClientWrite(client->server,client); } return 0; @@ -3762,8 +3764,9 @@ remoteDispatchDomainEventsDeregister (st static void remoteDispatchDomainEventSend (struct qemud_client *client, - virDomainPtr dom, - virDomainEventType event) + virDomainPtr dom, + int event, + int detail) { remote_message_header rep; XDR xdr; @@ -3799,7 +3802,8 @@ remoteDispatchDomainEventSend (struct qe /* build return data */ make_nonnull_domain (&data.dom, dom); - data.event = (int) event; + data.event = event; + data.detail = detail; if (!xdr_remote_domain_event_ret(&xdr, &data)) { remoteDispatchError (client, NULL, "%s", _("serialise return struct")); diff --git a/qemud/remote_protocol.c b/qemud/remote_protocol.c --- a/qemud/remote_protocol.c +++ b/qemud/remote_protocol.c @@ -2024,6 +2024,8 @@ xdr_remote_domain_event_ret (XDR *xdrs, return FALSE; if (!xdr_int (xdrs, &objp->event)) return FALSE; + if (!xdr_int (xdrs, &objp->detail)) + return FALSE; return TRUE; } diff --git a/qemud/remote_protocol.h b/qemud/remote_protocol.h --- a/qemud/remote_protocol.h +++ b/qemud/remote_protocol.h @@ -1130,6 +1130,7 @@ struct remote_domain_event_ret { struct remote_domain_event_ret { remote_nonnull_domain dom; int event; + int detail; }; typedef struct remote_domain_event_ret remote_domain_event_ret; #define REMOTE_PROGRAM 0x20008086 diff --git a/qemud/remote_protocol.x b/qemud/remote_protocol.x --- a/qemud/remote_protocol.x +++ b/qemud/remote_protocol.x @@ -1007,6 +1007,7 @@ struct remote_domain_event_ret { struct remote_domain_event_ret { remote_nonnull_domain dom; int event; + int detail; }; /*----- Protocol. -----*/ diff --git a/src/domain_event.c b/src/domain_event.c --- a/src/domain_event.c +++ b/src/domain_event.c @@ -198,7 +198,8 @@ int int virDomainEventCallbackQueuePush(virDomainEventQueuePtr evtQueue, virDomainPtr dom, - virDomainEventType event) + int event, + int detail) { virDomainEventPtr domEvent; @@ -214,6 +215,7 @@ virDomainEventCallbackQueuePush(virDomai } domEvent->dom = dom; domEvent->event = event; + domEvent->detail = detail; /* Make space on queue */ if (VIR_REALLOC_N(evtQueue->events, diff --git a/src/domain_event.h b/src/domain_event.h --- a/src/domain_event.h +++ b/src/domain_event.h @@ -58,7 +58,8 @@ int virDomainEventCallbackListRemove(vir */ struct _virDomainEvent { virDomainPtr dom; - virDomainEventType event; + int event; + int detail; }; typedef struct _virDomainEvent virDomainEvent; typedef virDomainEvent *virDomainEventPtr; @@ -72,7 +73,8 @@ typedef virDomainEventQueue *virDomainEv int virDomainEventCallbackQueuePush(virDomainEventQueuePtr evtQueue, virDomainPtr dom, - virDomainEventType event); + int event, + int detail); virDomainEventPtr virDomainEventCallbackQueuePop(virDomainEventQueuePtr evtQueue); diff --git a/src/qemu_driver.c b/src/qemu_driver.c --- a/src/qemu_driver.c +++ b/src/qemu_driver.c @@ -107,7 +107,8 @@ static int qemudSetNonBlock(int fd) { static void qemudDomainEventDispatch (struct qemud_driver *driver, virDomainObjPtr vm, - virDomainEventType evt); + int event, + int detail); static void qemudDispatchVMEvent(int fd, int events, @@ -137,13 +138,19 @@ qemudAutostartConfigs(struct qemud_drive unsigned int i; for (i = 0 ; i < driver->domains.count ; i++) { - if (driver->domains.objs[i]->autostart && - !virDomainIsActive(driver->domains.objs[i]) && - qemudStartVMDaemon(NULL, driver, driver->domains.objs[i], NULL) < 0) { - virErrorPtr err = virGetLastError(); - qemudLog(QEMUD_ERR, _("Failed to autostart VM '%s': %s\n"), - driver->domains.objs[i]->def->name, - err ? err->message : NULL); + virDomainObjPtr vm = driver->domains.objs[i]; + if (vm->autostart && + !virDomainIsActive(vm)) { + int ret = qemudStartVMDaemon(NULL, driver, vm, NULL); + if (ret < 0) { + virErrorPtr err = virGetLastError(); + qemudLog(QEMUD_ERR, _("Failed to autostart VM '%s': %s\n"), + vm->def->name, + err ? err->message : NULL); + } else { + qemudDomainEventDispatch(driver, vm, VIR_DOMAIN_EVENT_STARTED, + VIR_DOMAIN_EVENT_STARTED_BOOTED); + } } } } @@ -945,7 +952,6 @@ static int qemudStartVMDaemon(virConnect qemudShutdownVMDaemon(conn, driver, vm); return -1; } - qemudDomainEventDispatch(driver, vm, VIR_DOMAIN_EVENT_STARTED); } return ret; @@ -1030,6 +1036,9 @@ static int qemudDispatchVMLog(struct qem static int qemudDispatchVMLog(struct qemud_driver *driver, virDomainObjPtr vm, int fd) { if (qemudVMData(driver, vm, fd) < 0) { qemudShutdownVMDaemon(NULL, driver, vm); + qemudDomainEventDispatch(driver, vm, + VIR_DOMAIN_EVENT_STOPPED, + VIR_DOMAIN_EVENT_STOPPED_FAILED); if (!vm->persistent) virDomainRemoveInactive(&driver->domains, vm); @@ -1040,7 +1049,9 @@ static int qemudDispatchVMFailure(struct static int qemudDispatchVMFailure(struct qemud_driver *driver, virDomainObjPtr vm, int fd ATTRIBUTE_UNUSED) { qemudShutdownVMDaemon(NULL, driver, vm); - qemudDomainEventDispatch(driver, vm, VIR_DOMAIN_EVENT_STOPPED); + qemudDomainEventDispatch(driver, vm, + VIR_DOMAIN_EVENT_STOPPED, + VIR_DOMAIN_EVENT_STOPPED_SHUTDOWN); if (!vm->persistent) virDomainRemoveInactive(&driver->domains, vm); @@ -1513,6 +1524,9 @@ static virDomainPtr qemudDomainCreate(vi vm); return NULL; } + qemudDomainEventDispatch(driver, vm, + VIR_DOMAIN_EVENT_STARTED, + VIR_DOMAIN_EVENT_STARTED_BOOTED); dom = virGetDomain(conn, vm->def->name, vm->def->uuid); if (dom) dom->id = vm->def->id; @@ -1543,7 +1557,7 @@ static int qemudDomainSuspend(virDomainP } vm->state = VIR_DOMAIN_PAUSED; qemudDebug("Reply %s", info); - qemudDomainEventDispatch(driver, vm, VIR_DOMAIN_EVENT_SUSPENDED); + qemudDomainEventDispatch(driver, vm, VIR_DOMAIN_EVENT_SUSPENDED, 0); VIR_FREE(info); return 0; } @@ -1572,7 +1586,7 @@ static int qemudDomainResume(virDomainPt } vm->state = VIR_DOMAIN_RUNNING; qemudDebug("Reply %s", info); - qemudDomainEventDispatch(driver, vm, VIR_DOMAIN_EVENT_RESUMED); + qemudDomainEventDispatch(driver, vm, VIR_DOMAIN_EVENT_RESUMED, 0); VIR_FREE(info); return 0; } @@ -1611,7 +1625,9 @@ static int qemudDomainDestroy(virDomainP } qemudShutdownVMDaemon(dom->conn, driver, vm); - qemudDomainEventDispatch(driver, vm, VIR_DOMAIN_EVENT_STOPPED); + qemudDomainEventDispatch(driver, vm, + VIR_DOMAIN_EVENT_STOPPED, + VIR_DOMAIN_EVENT_STOPPED_DESTROYED); if (!vm->persistent) virDomainRemoveInactive(&driver->domains, vm); @@ -1942,10 +1958,12 @@ static int qemudDomainSave(virDomainPtr /* Shut it down */ qemudShutdownVMDaemon(dom->conn, driver, vm); + qemudDomainEventDispatch(driver, vm, + VIR_DOMAIN_EVENT_STOPPED, + VIR_DOMAIN_EVENT_STOPPED_SAVED); if (!vm->persistent) virDomainRemoveInactive(&driver->domains, vm); - qemudDomainEventDispatch(driver, vm, VIR_DOMAIN_EVENT_SAVED); return 0; } @@ -2240,6 +2258,10 @@ static int qemudDomainRestore(virConnect return -1; } + qemudDomainEventDispatch(driver, vm, + VIR_DOMAIN_EVENT_STARTED, + VIR_DOMAIN_EVENT_STARTED_RESTORED); + /* If it was running before, resume it now. */ if (header.was_running) { char *info; @@ -2252,7 +2274,6 @@ static int qemudDomainRestore(virConnect vm->state = VIR_DOMAIN_RUNNING; } - qemudDomainEventDispatch(driver, vm, VIR_DOMAIN_EVENT_RESTORED); return 0; } @@ -2312,6 +2333,7 @@ static int qemudDomainStart(virDomainPtr static int qemudDomainStart(virDomainPtr dom) { struct qemud_driver *driver = (struct qemud_driver *)dom->conn->privateData; virDomainObjPtr vm = virDomainFindByUUID(&driver->domains, dom->uuid); + int ret; if (!vm) { qemudReportError(dom->conn, dom, NULL, VIR_ERR_INVALID_DOMAIN, @@ -2319,7 +2341,13 @@ static int qemudDomainStart(virDomainPtr return -1; } - return qemudStartVMDaemon(dom->conn, driver, vm, NULL); + ret = qemudStartVMDaemon(dom->conn, driver, vm, NULL); + if (ret < 0) + return ret; + qemudDomainEventDispatch(driver, vm, + VIR_DOMAIN_EVENT_STARTED, + VIR_DOMAIN_EVENT_STARTED_BOOTED); + return 0; } @@ -3340,7 +3368,8 @@ qemudDomainEventDeregister (virConnectPt static void qemudDomainEventDispatch (struct qemud_driver *driver, virDomainObjPtr vm, - virDomainEventType evt) + int event, + int detail) { int i; virDomainEventCallbackListPtr cbList; @@ -3354,11 +3383,11 @@ static void qemudDomainEventDispatch (st vm->def->uuid); if (dom) { dom->id = virDomainIsActive(vm) ? vm->def->id : -1; - DEBUG("Dispatching callback %p %p event %d", - cbList->callbacks[i], - cbList->callbacks[i]->cb, evt); + DEBUG("Dispatching callback %p %p event %d detail %d", + cbList->callbacks[i], + cbList->callbacks[i]->cb, event, detail); cbList->callbacks[i]->cb(cbList->callbacks[i]->conn, - dom, evt, + dom, event, detail, cbList->callbacks[i]->opaque); virDomainFree(dom); } @@ -3507,6 +3536,9 @@ qemudDomainMigratePrepare2 (virConnectPt return -1; } + qemudDomainEventDispatch(driver, vm, + VIR_DOMAIN_EVENT_STARTED, + VIR_DOMAIN_EVENT_STARTED_MIGRATED); return 0; } @@ -3578,6 +3610,9 @@ qemudDomainMigratePerform (virDomainPtr /* Clean up the source domain. */ qemudShutdownVMDaemon (dom->conn, driver, vm); + qemudDomainEventDispatch(driver, vm, + VIR_DOMAIN_EVENT_STOPPED, + VIR_DOMAIN_EVENT_STOPPED_MIGRATED); if (!vm->persistent) virDomainRemoveInactive(&driver->domains, vm); @@ -3612,9 +3647,14 @@ qemudDomainMigrateFinish2 (virConnectPtr dom = virGetDomain (dconn, vm->def->name, vm->def->uuid); VIR_FREE(info); vm->state = VIR_DOMAIN_RUNNING; + qemudDomainEventDispatch(driver, vm, + VIR_DOMAIN_EVENT_RESUMED, 0); return dom; } else { qemudShutdownVMDaemon (dconn, driver, vm); + qemudDomainEventDispatch(driver, vm, + VIR_DOMAIN_EVENT_STOPPED, + VIR_DOMAIN_EVENT_STOPPED_FAILED); if (!vm->persistent) virDomainRemoveInactive(&driver->domains, vm); return NULL; diff --git a/src/remote_internal.c b/src/remote_internal.c --- a/src/remote_internal.c +++ b/src/remote_internal.c @@ -5154,7 +5154,7 @@ remoteRegister (void) */ static int remoteDomainReadEvent(virConnectPtr conn, XDR *xdr, - virDomainPtr *dom, int *event) + virDomainPtr *dom, int *event, int *detail) { remote_domain_event_ret ret; memset (&ret, 0, sizeof ret); @@ -5168,6 +5168,7 @@ remoteDomainReadEvent(virConnectPtr conn *dom = get_nonnull_domain(conn,ret.dom); *event = ret.event; + *detail = ret.detail; return 0; } @@ -5176,15 +5177,16 @@ remoteDomainProcessEvent(virConnectPtr c remoteDomainProcessEvent(virConnectPtr conn, XDR *xdr) { virDomainPtr dom; - int event,i; - struct private_data *priv = conn->privateData; - - if(!remoteDomainReadEvent(conn, xdr, &dom, &event)) { + int event, detail, i; + struct private_data *priv = conn->privateData; + + if(!remoteDomainReadEvent(conn, xdr, &dom, &event, &detail)) { DEBUG0("Calling domain event callbacks (no queue)"); for(i=0 ; i < priv->callbackList->count ; i++) { - if( priv->callbackList->callbacks[i] ) - priv->callbackList->callbacks[i]->cb(conn, dom, event, - priv->callbackList->callbacks[i]->opaque); + if (priv->callbackList->callbacks[i] ) + priv->callbackList->callbacks[i]->cb( + conn, dom, event, detail, + priv->callbackList->callbacks[i]->opaque); } } } @@ -5193,13 +5195,13 @@ remoteDomainQueueEvent(virConnectPtr con remoteDomainQueueEvent(virConnectPtr conn, XDR *xdr) { virDomainPtr dom; - int event; - struct private_data *priv = conn->privateData; - - if(!remoteDomainReadEvent(conn, xdr, &dom, &event)) + int event, detail; + struct private_data *priv = conn->privateData; + + if(!remoteDomainReadEvent(conn, xdr, &dom, &event, &detail)) { if( virDomainEventCallbackQueuePush(priv->domainEvents, - dom, event) < 0 ) { + dom, event, detail) < 0 ) { DEBUG("%s", "Error adding event to queue"); } } @@ -5292,6 +5294,7 @@ remoteDomainEventQueueFlush(int timer AT priv->callbackList->callbacks[i]->cb(domEvent->dom->conn, domEvent->dom, domEvent->event, + domEvent->detail, user_data); } } -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

3 6

[libvirt] PATCH: OpenVZ bridge support
by Daniel P. Berrange 17 Nov '08

17 Nov '08

This is an update of the patch http://www.redhat.com/archives/libvir-list/2008-October/msg00326.html To enable bridge support in the OpenVZ driver. As well as the fixes suggested last time, it includes an initial bit of HTML doc for the openvz driver, covering example XML, and the bridge configuration requirements Daniel diff -r 34eb1c9bdd7c docs/drvopenvz.html --- a/docs/drvopenvz.html Mon Nov 10 11:02:51 2008 +0000 +++ b/docs/drvopenvz.html Mon Nov 10 12:08:30 2008 +0000 @@ -122,6 +122,93 @@ </div> <div id="content"> <h1>OpenVZ container driver</h1> + <p> + The OpenVZ driver for libvirt allows use and management of container + based virtualization on a Linux host OS. Prior to using the OpenVZ + driver, the OpenVZ enabled kernel must be installed & booted, and the + OpenVZ userspace tools installed. The libvirt driver has been tested + with OpenVZ 3.0.22, but other 3.0.x versions should also work without + undue trouble. + </p> + <h2>Connections to OpenVZ driver</h2> + <p> + The libvirt OpenVZ driver is a single-instance privileged driver, + with a driver name of 'openvz'. Some example conection URIs for + the libvirt driver are: + </p> + <pre> + openvz:///system (local access) + openvz://example.com/system (remote access, TLS/x509) + openvz+tcp://example.com/system (remote access, SASl/Kerberos) + openvz+ssh://root@example.com/system (remote access, SSH tunnelled) + </pre> + <h2>Notes on bridged networking</h2> + <p> + Bridged networking enables a guest domain (ie container) to have its + network interface connected directly to the host's physical LAN. Before + this can be used there are a couple of configuration pre-requisites for + the host OS. + </p> + <h3>Host network devices</h3> + <p> + One or more of the physical devices must be attached to a bridge. The + process for this varies according to the operating system in use, so + for up to date notes consult the <a href="http://wiki.libvirt.org">Wiki</a> + or your operating system's networking documentation. The basic idea is + that the host OS should end up with a bridge device "br0" containing a + physical device "eth0", or a bonding device "bond0". + </p> + <h3>OpenVZ tools configuration</h3> + <p> + OpenVZ releases later than 3.0.23 ship with a standard network device + setup script that is able to setup bridging, named + <code>/usr/sbin/vznetaddbr</code>. For releases prior to 3.0.23, this + script must be created manually by the host OS adminstrator. The + simplest way is to just download the latest version of this script + from a newer OpenVZ release, or upstream source repository. Then + a generic configuration file <code>/etc/vz/vznetctl.conf</code> + must be created containing + </p> + <pre> +#!/bin/bash +EXTERNAL_SCRIPT="/usr/sbin/vznetaddbr" + </pre> + <p> + The host OS is now ready to allow bridging of guest containers, which + will work whether the container is started with libvirt, or OpenVZ + tools. + </p> + <h2>Example guest domain XML configuration</h2> + <p> + The current libvirt OpenVZ driver has a restriction that the + domain names must match the OpenVZ container VEID, which by + convention start at 100, and are incremented from there. The + choice of OS template to use inside the container is determined + by the <code>filesystem</code> tag, and the template source name + matches the templates known to OpenVZ tools. + </p> + <pre> +<domain type='openvz' id='104'> + <name>104</name> + <uuid>86c12009-e591-a159-6e9f-91d18b85ef78</uuid> + <vcpu>3</vcpu> + <os> + <type>exe</type> + <init>/sbin/init</init> + </os> + <devices> + <filesystem type='template'> + <source name='fedora-9-i386-minimal'/> + <target dir='/'/> + </filesystem> + <interface type='bridge'> + <mac address='00:18:51:5b:ea:bf'/> + <source bridge='br0'/> + <target dev='veth101.0'/> + </interface> + </devices> +</domain> + </pre> </div> </div> <div id="footer"> diff -r 34eb1c9bdd7c docs/drvopenvz.html.in --- a/docs/drvopenvz.html.in Mon Nov 10 11:02:51 2008 +0000 +++ b/docs/drvopenvz.html.in Mon Nov 10 12:08:30 2008 +0000 @@ -1,5 +1,109 @@ -<html> +<html>  <body> <h1>OpenVZ container driver</h1> + + <p> + The OpenVZ driver for libvirt allows use and management of container + based virtualization on a Linux host OS. Prior to using the OpenVZ + driver, the OpenVZ enabled kernel must be installed & booted, and the + OpenVZ userspace tools installed. The libvirt driver has been tested + with OpenVZ 3.0.22, but other 3.0.x versions should also work without + undue trouble. + </p> + + <h2>Connections to OpenVZ driver</h2> + + <p> + The libvirt OpenVZ driver is a single-instance privileged driver, + with a driver name of 'openvz'. Some example conection URIs for + the libvirt driver are: + </p> + + <pre> + openvz:///system (local access) + openvz://example.com/system (remote access, TLS/x509) + openvz+tcp://example.com/system (remote access, SASl/Kerberos) + openvz+ssh://root@example.com/system (remote access, SSH tunnelled) + </pre> + + <h2>Notes on bridged networking</h2> + + <p> + Bridged networking enables a guest domain (ie container) to have its + network interface connected directly to the host's physical LAN. Before + this can be used there are a couple of configuration pre-requisites for + the host OS. + </p> + + <h3>Host network devices</h3> + + <p> + One or more of the physical devices must be attached to a bridge. The + process for this varies according to the operating system in use, so + for up to date notes consult the <a href="http://wiki.libvirt.org">Wiki</a> + or your operating system's networking documentation. The basic idea is + that the host OS should end up with a bridge device "br0" containing a + physical device "eth0", or a bonding device "bond0". + </p> + + <h3>OpenVZ tools configuration</h3> + + <p> + OpenVZ releases later than 3.0.23 ship with a standard network device + setup script that is able to setup bridging, named + <code>/usr/sbin/vznetaddbr</code>. For releases prior to 3.0.23, this + script must be created manually by the host OS adminstrator. The + simplest way is to just download the latest version of this script + from a newer OpenVZ release, or upstream source repository. Then + a generic configuration file <code>/etc/vz/vznetctl.conf</code> + must be created containing + </p> + + <pre> +#!/bin/bash +EXTERNAL_SCRIPT="/usr/sbin/vznetaddbr" + </pre> + + <p> + The host OS is now ready to allow bridging of guest containers, which + will work whether the container is started with libvirt, or OpenVZ + tools. + </p> + + + <h2>Example guest domain XML configuration</h2> + + <p> + The current libvirt OpenVZ driver has a restriction that the + domain names must match the OpenVZ container VEID, which by + convention start at 100, and are incremented from there. The + choice of OS template to use inside the container is determined + by the <code>filesystem</code> tag, and the template source name + matches the templates known to OpenVZ tools. + </p> + + <pre> +<domain type='openvz' id='104'> + <name>104</name> + <uuid>86c12009-e591-a159-6e9f-91d18b85ef78</uuid> + <vcpu>3</vcpu> + <os> + <type>exe</type> + <init>/sbin/init</init> + </os> + <devices> + <filesystem type='template'> + <source name='fedora-9-i386-minimal'/> + <target dir='/'/> + </filesystem> + <interface type='bridge'> + <mac address='00:18:51:5b:ea:bf'/> + <source bridge='br0'/> + <target dev='veth101.0'/> + </interface> + </devices> +</domain> + </pre> + </body> </html> diff -r 34eb1c9bdd7c src/openvz_conf.c --- a/src/openvz_conf.c Mon Nov 10 11:02:51 2008 +0000 +++ b/src/openvz_conf.c Mon Nov 10 12:08:30 2008 +0000 @@ -51,7 +51,7 @@ static char *openvzLocateConfDir(void); static int openvzGetVPSUUID(int vpsid, char *uuidstr); -static int openvzLocateConfFile(int vpsid, char *conffile, int maxlen); +static int openvzLocateConfFile(int vpsid, char *conffile, int maxlen, const char *ext); static int openvzAssignUUIDs(void); int @@ -144,6 +144,8 @@ virCapsPtr openvzCapsInit(void) 0, 0)) == NULL) goto no_memory; + virCapabilitiesSetMacPrefix(caps, (unsigned char[]){ 0x52, 0x54, 0x00 }); + if ((guest = virCapabilitiesAddGuest(caps, "exe", utsname.machine, @@ -168,54 +170,6 @@ no_memory: return NULL; } - -/* function checks MAC address is empty - return 0 - empty - 1 - not -*/ -int openvzCheckEmptyMac(const unsigned char *mac) -{ - int i; - for (i = 0; i < VIR_MAC_BUFLEN; i++) - if (mac[i] != 0x00) - return 1; - - return 0; -} - -/* convert mac address to string - return pointer to string or NULL -*/ -char *openvzMacToString(const unsigned char *mac) -{ - char str[20]; - if (snprintf(str, 18, "%02X:%02X:%02X:%02X:%02X:%02X", - mac[0], mac[1], mac[2], - mac[3], mac[4], mac[5]) >= 18) - return NULL; - - return strdup(str); -} - -/*parse MAC from view: 00:18:51:8F:D9:F3 - return -1 - error - 0 - OK -*/ -static int openvzParseMac(const char *macaddr, unsigned char *mac) -{ - int ret; - ret = sscanf((const char *)macaddr, "%02X:%02X:%02X:%02X:%02X:%02X", - (unsigned int*)&mac[0], - (unsigned int*)&mac[1], - (unsigned int*)&mac[2], - (unsigned int*)&mac[3], - (unsigned int*)&mac[4], - (unsigned int*)&mac[5]) ; - if (ret == 6) - return 0; - - return -1; -} static int openvzReadNetworkConf(virConnectPtr conn, @@ -287,6 +241,9 @@ openvzReadNetworkConf(virConnectPtr conn while (*next != '\0' && *next != ',') next++; if (STRPREFIX(p, "ifname=")) { p += 7; + /* skip in libvirt */ + } else if (STRPREFIX(p, "host_ifname=")) { + p += 12; len = next - p; if (len > 16) { openvzError(conn, VIR_ERR_INTERNAL_ERROR, @@ -294,14 +251,25 @@ openvzReadNetworkConf(virConnectPtr conn goto error; } + if (VIR_ALLOC_N(net->ifname, len+1) < 0) + goto no_memory; + + strncpy(net->ifname, p, len); + net->ifname[len] = '\0'; + } else if (STRPREFIX(p, "bridge=")) { + p += 7; + len = next - p; + if (len > 16) { + openvzError(conn, VIR_ERR_INTERNAL_ERROR, + "%s", _("Too long bridge device name")); + goto error; + } + if (VIR_ALLOC_N(net->data.bridge.brname, len+1) < 0) goto no_memory; strncpy(net->data.bridge.brname, p, len); net->data.bridge.brname[len] = '\0'; - } else if (STRPREFIX(p, "host_ifname=")) { - p += 12; - //skip in libvirt } else if (STRPREFIX(p, "mac=")) { p += 4; len = next - p; @@ -312,14 +280,11 @@ openvzReadNetworkConf(virConnectPtr conn } strncpy(cpy_temp, p, len); cpy_temp[len] = '\0'; - if (openvzParseMac(cpy_temp, net->mac)<0) { + if (virParseMacAddr(cpy_temp, net->mac) < 0) { openvzError(conn, VIR_ERR_INTERNAL_ERROR, "%s", _("Wrong MAC address")); goto error; } - } else if (STRPREFIX(p, "host_mac=")) { - p += 9; - //skip in libvirt } p = ++next; } while (p < token + strlen(token)); @@ -491,6 +456,68 @@ int openvzLoadDomains(struct openvz_driv return -1; } + +int +openvzWriteConfigParam(int vpsid, const char *param, const char *value) +{ + char conf_file[PATH_MAX]; + char temp_file[PATH_MAX]; + char line[PATH_MAX] ; + int fd, temp_fd; + + if (openvzLocateConfFile(vpsid, conf_file, PATH_MAX, "conf")<0) + return -1; + if (openvzLocateConfFile(vpsid, temp_file, PATH_MAX, "tmp")<0) + return -1; + + fd = open(conf_file, O_RDONLY); + if (fd == -1) + return -1; + temp_fd = open(temp_file, O_WRONLY | O_CREAT | O_TRUNC, 0644); + if (temp_fd == -1) { + close(fd); + return -1; + } + + while(1) { + if (openvz_readline(fd, line, sizeof(line)) <= 0) + break; + + if (!STRPREFIX(line, param) && + line[strlen(param)] == '=') { + if (safewrite(temp_fd, line, strlen(line)) != + strlen(line)) + goto error; + } + } + + if (safewrite(temp_fd, param, strlen(param)) < 0 || + safewrite(temp_fd, "=\"", 2) < 0 || + safewrite(temp_fd, value, strlen(value)) < 0 || + safewrite(temp_fd, "\"\n", 2) < 0) + goto error; + + if (close(fd) < 0) + goto error; + fd = -1; + if (close(temp_fd) < 0) + goto error; + temp_fd = -1; + + if (rename(temp_file, conf_file) < 0) + goto error; + + return 0; + +error: + if (fd != -1) + close(fd); + if (temp_fd != -1) + close(temp_fd); + unlink(temp_file); + return -1; +} + /* * Read parameter from container config * sample: 133, "OSTEMPLATE", value, 1024 @@ -508,7 +535,7 @@ openvzReadConfigParam(int vpsid ,const c char * sf, * token; char *saveptr = NULL; - if (openvzLocateConfFile(vpsid, conf_file, PATH_MAX)<0) + if (openvzLocateConfFile(vpsid, conf_file, PATH_MAX, "conf")<0) return -1; value[0] = 0; @@ -548,7 +575,7 @@ openvzReadConfigParam(int vpsid ,const c * 0 - OK */ static int -openvzLocateConfFile(int vpsid, char *conffile, int maxlen) +openvzLocateConfFile(int vpsid, char *conffile, int maxlen, const char *ext) { char * confdir; int ret = 0; @@ -557,7 +584,8 @@ openvzLocateConfFile(int vpsid, char *co if (confdir == NULL) return -1; - if (snprintf(conffile, maxlen, "%s/%d.conf", confdir, vpsid) >= maxlen) + if (snprintf(conffile, maxlen, "%s/%d.%s", + confdir, vpsid, ext ? ext : "conf") >= maxlen) ret = -1; VIR_FREE(confdir); @@ -614,7 +642,7 @@ openvzGetVPSUUID(int vpsid, char *uuidst char iden[1024]; int fd, ret; - if (openvzLocateConfFile(vpsid, conf_file, PATH_MAX)<0) + if (openvzLocateConfFile(vpsid, conf_file, PATH_MAX, "conf")<0) return -1; fd = open(conf_file, O_RDONLY); @@ -654,7 +682,7 @@ openvzSetDefinedUUID(int vpsid, unsigned if (uuid == NULL) return -1; - if (openvzLocateConfFile(vpsid, conf_file, PATH_MAX)<0) + if (openvzLocateConfFile(vpsid, conf_file, PATH_MAX, "conf")<0) return -1; if (openvzGetVPSUUID(vpsid, uuidstr)) @@ -722,4 +750,3 @@ static int openvzAssignUUIDs(void) VIR_FREE(conf_dir); return 0; } - diff -r 34eb1c9bdd7c src/openvz_conf.h --- a/src/openvz_conf.h Mon Nov 10 11:02:51 2008 +0000 +++ b/src/openvz_conf.h Mon Nov 10 12:08:30 2008 +0000 @@ -50,6 +50,8 @@ enum { OPENVZ_WARN, OPENVZ_ERR }; #define VZLIST "/usr/sbin/vzlist" #define VZCTL "/usr/sbin/vzctl" +#define VZCTL_BRIDGE_MIN_VERSION ((3 * 1000 * 1000) + (0 * 1000) + 22 + 1) + struct openvz_driver { virCapsPtr caps; virDomainObjList domains; @@ -60,12 +62,11 @@ int openvzExtractVersion(virConnectPtr c int openvzExtractVersion(virConnectPtr conn, struct openvz_driver *driver); int openvzReadConfigParam(int vpsid ,const char * param, char *value, int maxlen); +int openvzWriteConfigParam(int vpsid, const char *param, const char *value); virCapsPtr openvzCapsInit(void); int openvzLoadDomains(struct openvz_driver *driver); void openvzFreeDriver(struct openvz_driver *driver); int strtoI(const char *str); -int openvzCheckEmptyMac(const unsigned char *mac); -char *openvzMacToString(const unsigned char *mac); int openvzSetDefinedUUID(int vpsid, unsigned char *uuid); #endif /* OPENVZ_CONF_H */ diff -r 34eb1c9bdd7c src/openvz_driver.c --- a/src/openvz_driver.c Mon Nov 10 11:02:51 2008 +0000 +++ b/src/openvz_driver.c Mon Nov 10 12:08:30 2008 +0000 @@ -55,6 +55,7 @@ #include "openvz_conf.h" #include "nodeinfo.h" #include "memory.h" +#include "bridge.h" #define OPENVZ_MAX_ARG 28 #define CMDBUF_LEN 1488 @@ -329,13 +330,56 @@ static int openvzDomainReboot(virDomainP return 0; } +static char * +openvzGenerateVethName(int veid, char *dev_name_ve) +{ + char dev_name[32]; + int ifNo = 0; + + if (sscanf(dev_name_ve, "%*[^0-9]%d", &ifNo) != 1) + return NULL; + if (snprintf(dev_name, sizeof(dev_name), "veth%d.%d", veid, ifNo) < 7) + return NULL; + return strdup(dev_name); +} + +static char * +openvzGenerateContainerVethName(int veid) +{ + int ret; + char temp[1024]; + + /* try to get line "^NETIF=..." from config */ + if ( (ret = openvzReadConfigParam(veid, "NETIF", temp, sizeof(temp))) <= 0) { + snprintf(temp, sizeof(temp), "eth0"); + } else { + char *s; + int max = 0; + + /* get maximum interface number (actually, it is the last one) */ + for (s=strtok(temp, ";"); s; s=strtok(NULL, ";")) { + int x; + + if (sscanf(s, "ifname=eth%d", &x) != 1) return NULL; + if (x > max) max = x; + } + + /* set new name */ + snprintf(temp, sizeof(temp), "eth%d", max+1); + } + return strdup(temp); +} + static int openvzDomainSetNetwork(virConnectPtr conn, const char *vpsid, - virDomainNetDefPtr net) + virDomainNetDefPtr net, + virBufferPtr configBuf) { int rc = 0, narg; const char *prog[OPENVZ_MAX_ARG]; - char *mac = NULL; + char macaddr[VIR_MAC_STRING_BUFLEN]; + struct openvz_driver *driver = (struct openvz_driver *) conn->privateData; + char *opt = NULL; #define ADD_ARG_LIT(thisarg) \ do { \ @@ -367,21 +411,61 @@ openvzDomainSetNetwork(virConnectPtr con ADD_ARG_LIT(vpsid); } - if (openvzCheckEmptyMac(net->mac) > 0) - mac = openvzMacToString(net->mac); - - if (net->type == VIR_DOMAIN_NET_TYPE_BRIDGE && - net->data.bridge.brname != NULL) { - char opt[1024]; + virFormatMacAddr(net->mac, macaddr); + + if (net->type == VIR_DOMAIN_NET_TYPE_BRIDGE) { + virBuffer buf = VIR_BUFFER_INITIALIZER; + char *dev_name_ve; + int veid = strtoI(vpsid); + //--netif_add ifname[,mac,host_ifname,host_mac] ADD_ARG_LIT("--netif_add") ; - strncpy(opt, net->data.bridge.brname, 256); - if (mac != NULL) { - strcat(opt, ","); - strcat(opt, mac); - } + + /* generate interface name in ve and copy it to options */ + dev_name_ve = openvzGenerateContainerVethName(veid); + if (dev_name_ve == NULL) { + openvzError(conn, VIR_ERR_INTERNAL_ERROR, + _("Could not generate eth name for container")); + rc = -1; + goto exit; + } + + /* if user doesn't specified host interface name, + * than we need to generate it */ + if (net->ifname == NULL) { + net->ifname = openvzGenerateVethName(veid, dev_name_ve); + if (net->ifname == NULL) { + openvzError(conn, VIR_ERR_INTERNAL_ERROR, + _("Could not generate veth name")); + rc = -1; + VIR_FREE(dev_name_ve); + goto exit; + } + } + + virBufferAdd(&buf, dev_name_ve, -1); /* Guest dev */ + virBufferVSprintf(&buf, ",%s", macaddr); /* Guest dev mac */ + virBufferVSprintf(&buf, ",%s", net->ifname); /* Host dev */ + virBufferVSprintf(&buf, ",%s", macaddr); /* Host dev mac */ + + if (driver->version >= VZCTL_BRIDGE_MIN_VERSION) { + virBufferVSprintf(&buf, ",%s", net->data.bridge.brname); /* Host bridge */ + } else { + virBufferVSprintf(configBuf, "ifname=%s", dev_name_ve); + virBufferVSprintf(configBuf, ",mac=%s", macaddr); /* Guest dev mac */ + virBufferVSprintf(configBuf, ",host_ifname=%s", net->ifname); /* Host dev */ + virBufferVSprintf(configBuf, ",host_mac=%s", macaddr); /* Host dev mac */ + virBufferVSprintf(configBuf, ",bridge=%s", net->data.bridge.brname); /* Host bridge */ + } + + VIR_FREE(dev_name_ve); + + if (!(opt = virBufferContentAndReset(&buf))) + goto no_memory; + ADD_ARG_LIT(opt) ; - }else if (net->type == VIR_DOMAIN_NET_TYPE_ETHERNET && + VIR_FREE(opt); + } else if (net->type == VIR_DOMAIN_NET_TYPE_ETHERNET && net->data.ethernet.ipaddr != NULL) { //--ipadd ip ADD_ARG_LIT("--ipadd") ; @@ -402,18 +486,66 @@ openvzDomainSetNetwork(virConnectPtr con exit: cmdExecFree(prog); - VIR_FREE(mac); return rc; no_memory: + VIR_FREE(opt); openvzError(conn, VIR_ERR_INTERNAL_ERROR, _("Could not put argument to %s"), VZCTL); cmdExecFree(prog); - VIR_FREE(mac); return -1; #undef ADD_ARG_LIT } + + +static int +openvzDomainSetNetworkConfig(virConnectPtr conn, + virDomainDefPtr def) +{ + unsigned int i; + virBuffer buf = VIR_BUFFER_INITIALIZER; + char *param; + int first = 1; + struct openvz_driver *driver = (struct openvz_driver *) conn->privateData; + + for (i = 0 ; i < def->nnets ; i++) { + if (driver->version < VZCTL_BRIDGE_MIN_VERSION && + def->nets[i]->type == VIR_DOMAIN_NET_TYPE_BRIDGE) { + if (first) + first = 0; + else + virBufferAddLit(&buf, ";"); + } + + if (openvzDomainSetNetwork(conn, def->name, def->nets[i], &buf) < 0) { + openvzError(conn, VIR_ERR_INTERNAL_ERROR, + "%s", _("Could not configure network")); + goto exit; + } + } + + if (driver->version < VZCTL_BRIDGE_MIN_VERSION && def->nnets) { + param = virBufferContentAndReset(&buf); + if (param) { + if (openvzWriteConfigParam(strtoI(def->name), "NETIF", param) < 0) { + VIR_FREE(param); + openvzError(conn, VIR_ERR_INTERNAL_ERROR, + "%s", _("cannot replace NETIF config")); + return -1; + } + VIR_FREE(param); + } + } + + return 0; + +exit: + param = virBufferContentAndReset(&buf); + VIR_FREE(param); + return -1; +} + static virDomainPtr openvzDomainDefineXML(virConnectPtr conn, const char *xml) @@ -422,7 +554,6 @@ openvzDomainDefineXML(virConnectPtr conn virDomainDefPtr vmdef = NULL; virDomainObjPtr vm = NULL; virDomainPtr dom = NULL; - int i; const char *prog[OPENVZ_MAX_ARG]; prog[0] = NULL; @@ -468,17 +599,12 @@ openvzDomainDefineXML(virConnectPtr conn goto exit; } + if (openvzDomainSetNetworkConfig(conn, vmdef) < 0) + goto exit; + dom = virGetDomain(conn, vm->def->name, vm->def->uuid); if (dom) dom->id = -1; - - for (i = 0 ; i < vmdef->nnets ; i++) { - if (openvzDomainSetNetwork(conn, vmdef->name, vmdef->nets[i]) < 0) { - openvzError(conn, VIR_ERR_INTERNAL_ERROR, - "%s", _("Could not configure network")); - goto exit; - } - } if (vmdef->vcpus > 0) { if (openvzDomainSetVcpus(dom, vmdef->vcpus) < 0) { @@ -500,7 +626,6 @@ openvzDomainCreateXML(virConnectPtr conn virDomainDefPtr vmdef = NULL; virDomainObjPtr vm = NULL; virDomainPtr dom = NULL; - int i; struct openvz_driver *driver = (struct openvz_driver *) conn->privateData; const char *progstart[] = {VZCTL, "--quiet", "start", NULL, NULL}; const char *progcreate[OPENVZ_MAX_ARG]; @@ -546,13 +671,8 @@ openvzDomainCreateXML(virConnectPtr conn goto exit; } - for (i = 0 ; i < vmdef->nnets ; i++) { - if (openvzDomainSetNetwork(conn, vmdef->name, vmdef->nets[i]) < 0) { - openvzError(conn, VIR_ERR_INTERNAL_ERROR, - "%s", _("Could not configure network")); - goto exit; - } - } + if (openvzDomainSetNetworkConfig(conn, vmdef) < 0) + goto exit; progstart[3] = vmdef->name; -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

4 5

[libvirt] [PATCH] Java bindings for domain events
by David Lively 15 Nov '08

15 Nov '08

The attached patch (against libvirt-java) contains Java bindings for the new domain event code. It works (see EventTest.java), but there's a certain amount of hokiness regarding the EventImpl stuff that I'd like to discuss. Unlike the C and Python interfaces, the Java interface does not currently allow the client to supply an EventImpl. The problem is that Java really has no way to interact with unix file descriptors so there's no reasonable way to implement a fd-watching EventImpl in pure Java (java.io.FileDescriptor doesn't do the trick since there's no way of constructing one from raw (int) unix fd)[**]. So for now, I've had the Java bindings register an EventImpl when the Connect class is loaded. This EventImpl is a Java class, with native methods implementing it. In fact, I've simply stolen (verbatim) the EventImpl from libvirt/qemud/events.c and made the native methods call it. [If we stick with this solution, it would obviously be better to somehow share this code with libvirtd rather than copy it.] The other tricky subject is multi-threading. For now, I've avoided it by exposing Connect.eventImpl.run_once() and forcing the client to call it from their "event loop". But the normal Java way of doing things would simply run the EventImpl in a separate thread. In fact, this EventImpl does implement Runnable, which makes it trivial to run in a separate thread -- but I don't declare that it implements Runnable yet because it's not safe to run in a thread while another thread might be making libvirt calls. It shouldn't be hard to make this thread-safe using Java synchronized methods and statements, but I haven't done that yet. Should I?? ** java.nio.Channel and friends seem to be the "right" interface for exposing abstract "selectable channels" in Java. It's just complicated enough that I've avoided it for now. But I can look into going this way for allowing Java to provide an EventImpl in the future .. Cheers, Dave

4 9

Re: >>: [libvirt] OpenVZ capabilites in LibVIRT
by Anton Protopopov 15 Nov '08

15 Nov '08

2008/11/15 Ivan Vovk <vovk.is(a)gmail.com> > Hi Anton, > > do i understand correctly that patch for OpenVZ bridge support was > committed? If so can it be obtained from CVS repository with a new snapshot? > Yes, it is commited in libvirt-0.4.6-alt2 that appear in Sisyphus on monday, or so. If you are interesting, you can see my git: http://git.altlinux.org/people/aspsk/packages/?p=libvirt.git;a=summary > > Then can you help me a bit again? After I got a new checkout with > > cvs -d :pserver:anoncvs@libvirt.org:2401/data/cvs co libvirt > > and tried to do a build, i got the following: > > [root@alt-03 libvirt]# ./autogen.sh --prefix=$HOME/usr > > You must have autopoint installed to compile libvirt. > Download the appropriate package for your distribution, > But was is 'autopoint'? I was not able to find it neither in ALT's > repository, not somewhere else ... =( > # apt-get install gettext-tools # man autopoint

1 0

[libvirt] [PATCH] Fix lv scanning with encrypted volumes
by Cole Robinson 14 Nov '08

14 Nov '08

See https://bugzilla.redhat.com/show_bug.cgi?id=470693 Using ':' as a delimiter for the lvs command isn't reliable, since it looks like encrypted volume groups have a colon in the physical device name, which throws of the regex. The attached patch changes the delimiter to ','. Maintains existing behavior for me, just waiting on confirmation from the reporter that this indeed does the job. Thanks, Cole

2 2

[libvirt] [RFC] making (newly public) EventImpl interface more consistent
by David Lively 14 Nov '08

14 Nov '08

Hi Folks - Since virEventRegisterImpl is now public (in libvirt.h), a nagging concern of mine has become more urgent. Essentially this callback gives clients a way of registering their own "handle (fd) watcher" and "timer" functionality for use by libvirt. What bugs me is the inconsistency between the handle-watcher and timer interfaces: the timer "add" function returns a timer id, which is then used to identify the timer to the "update" and "remove" functions. But the handle-watcher add / update / remove functions identify the watcher by the handle (fd). The semantics of registering the same handle twice aren't specified (what happens when we pass that same fd in a subsequent "update" or "remove"?). Even worse, this doesn't allow one to manage multiple watches on the same handle reasonably. So why not make the handle "add" function return a "watch id" (analogous to the "timer id" returned by the timer "add" fn)? And then use this watch id to specify the handle-watcher in the "update" and "remove" functions. This conveniently handles multiple watches on the same handle, and also makes the "handle-watching" interface more consistent with the "timer" interface (which is registered at the same time). We'd pass both the watch id and the handle (fd) into the watch-handler callback. I'd like to implement and submit this (along with fixups to the event test code) if there are no objections. Thanks, Dave P.S. I'm currently working on Java bindings to the new event code ...

4 15