- Devel - libvirt List Archives

[libvirt] [PATCH] Convert HAVE_SELINUX to WITH_SELINUX
by Guido Günther 11 Mar '13

11 Mar '13

these were missed by 63f18f378693cac6b6e33f4a8c15c20cb191c7c8 --- src/lxc/lxc_container.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/lxc/lxc_container.c b/src/lxc/lxc_container.c index 7ec67f4..4d09791 100644 --- a/src/lxc/lxc_container.c +++ b/src/lxc/lxc_container.c @@ -1927,7 +1927,7 @@ static int lxcContainerSetupPivotRoot(virDomainDefPtr vmDef, if (lxcContainerPivotRoot(root) < 0) goto cleanup; -#if HAVE_SELINUX +#if WITH_SELINUX /* Some versions of Linux kernel don't let you overmount * the selinux filesystem, so make sure we kill it first */ @@ -2039,7 +2039,7 @@ static int lxcContainerSetupExtraMounts(virDomainDefPtr vmDef, if (lxcContainerIdentifyCGroups(&mounts, &nmounts, &cgroupRoot) < 0) goto cleanup; -#if HAVE_SELINUX +#if WITH_SELINUX /* Some versions of Linux kernel don't let you overmount * the selinux filesystem, so make sure we kill it first */ -- 1.7.10.4

2 2

[libvirt] [PATCH] bandwidth: Require network QoS if interface uses 'floor'
by Michal Privoznik 11 Mar '13

11 Mar '13

By current implementation, network inbound is required in order to use 'floor' for guaranteeing minimal throughput. This is so, because we want user to tell us the maximal throughput of the network instead of finding out ourselves (and detect bogus values in case of virtual interfaces). However, we are nowadays requiring this only on documentation level. So if user starts a domain with 'floor' set on one its interfaces, we silently ignore the setting. We should error out instead. --- src/network/bridge_driver.c | 17 ++++++++++++++--- 1 file changed, 14 insertions(+), 3 deletions(-) diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c index 31c8585..330c147 100644 --- a/src/network/bridge_driver.c +++ b/src/network/bridge_driver.c @@ -4535,11 +4535,22 @@ networkCheckBandwidth(virNetworkObjPtr net, unsigned long long tmp_new_rate = 0; char ifmac[VIR_MAC_STRING_BUFLEN]; + virMacAddrFormat(&iface->mac, ifmac); + + if (ifaceBand && ifaceBand->in && ifaceBand->in->floor && + !(netBand && netBand->in)) { + virReportError(VIR_ERR_OPERATION_UNSUPPORTED, + _("Cannot use 'floor' on %s because network '%s' " + "does not have any inbound QoS set"), + ifmac, net->def->name); + return -1; + } + if (!ifaceBand || !ifaceBand->in || !ifaceBand->in->floor || - !netBand || !netBand->in) + !netBand || !netBand->in) { + /* no QoS required, claim success */ return 1; - - virMacAddrFormat(&iface->mac, ifmac); + } tmp_new_rate = netBand->in->average; tmp_floor_sum += ifaceBand->in->floor; -- 1.8.1.5

2 2

[libvirt] [PATCH 0/2] Proper fix for xmlSaveToBuffer missing on RHEL4
by Peter Krempa 10 Mar '13

10 Mar '13

The original fix for one of the issues was done by defining a missing symbol from libxml2. The fix is not needed as the code using the function is never used. Revert the fix and remove the unused offending function instead. Peter Krempa (2): Revert "Ensure xmlSaveToBuffer is always defined" virsh-domain: Remove unused vshCompleteXMLFromDomain tools/virsh-domain.c | 139 --------------------------------------------------- 1 file changed, 139 deletions(-) -- 1.8.1.1

2 4

[libvirt] [PATCH] Add NUMA memory information to virsh capabilities output.
by Dusty Mabe 08 Mar '13

08 Mar '13

Hi, This is a follow up to a previous proposed patch that added NUMA memory and CPU thread siblings information to virsh capabilities outout (https://www.redhat.com/archives/libvir-list/2012-November/msg00449.html) Since in a recent commit the thread siblings information was added (http://libvirt.org/git/?p=libvirt.git;a=commit;h=828820e2d) I have modified my patch to just add the NUMA memory information. An example of the output from virsh capabilities looks like the following text: <topology> <cells num='2'> <cell id='0'> <memory unit='KiB'>67073420</memory> <cpus num='16'> <cpu id='0' socket_id='0' core_id='0' siblings='0,16'/> <cpu id='1' socket_id='0' core_id='1' siblings='1,17'/> ... ... </cpus> </cell> Dusty Mabe Dusty Mabe (1): Add NUMA memory information to virsh capabilities output. docs/schemas/capability.rng | 10 ++++ src/conf/capabilities.c | 8 +++ src/conf/capabilities.h | 2 + src/nodeinfo.c | 64 +++++++++++++++++++++- src/test/test_driver.c | 2 +- src/xen/xend_internal.c | 2 +- tests/capabilityschemadata/caps-test3.xml | 88 +++++++++++++++++++++++++++++++ 7 files changed, 173 insertions(+), 3 deletions(-) create mode 100644 tests/capabilityschemadata/caps-test3.xml -- 1.7.11.7

3 8

[libvirt] Mentors and project ideas wanted for GSoC 2013
by Stefan Hajnoczi 08 Mar '13

08 Mar '13

QEMU will apply to Google Summer of Code 2013 on Monday 18th March. We need the have project ideas ready as part of QEMU's application as a GSoC 2013 organization. The wiki is still sparse so I'm asking that everyone posts their project ideas: http://wiki.qemu.org/Google_Summer_of_Code_2013 Quick recap: Google Summer of Code is a 12-week program where university students are funded to contribute to open source projects. QEMU will apply and we need project ideas that students can work on this summer. The experience can be very rewarding for students and mentors, and QEMU has participated in past years. Each project idea needs a QEMU contributor who is willing to mentor a student. At the bottom of the wiki page there is information if you want to find out what mentoring is all about. If you have questions about posting project ideas or mentoring, please respond or ping me on IRC! CCed a subset of folks who might be interested. Thanks, Stefan

1 0

[libvirt] [PATCH] qemuDomainBlockStatsFlags: Guard disk lookup with a domain job
by Michal Privoznik 08 Mar '13

08 Mar '13

When there are two concurrent threads, we may dereference a NULL pointer, even though it has been checked before: 1. Thread1: starts executing qemuDomainBlockStatsFlags() with nparams != 0. It finds given disk and successfully pass check for disk->info.alias not being NULL. 2. Thread2: starts executing qemuDomainDetachDeviceFlags() on the very same disk as Thread1 is working on. 3. Thread1: gets to qemuDomainObjBeginJob() where it sets a job on a domain. 4. Thread2: also tries to set a job. However, we are not guaranteed which thread wins. So assume it's Thread2 who can continue. 5. Thread2: does the actual detach and frees disk->info.alias 6. Thread2: quits the job 7. Thread1: now successfully acquires the job, and accesses a NULL pointer. --- src/qemu/qemu_driver.c | 18 ++++++------------ 1 file changed, 6 insertions(+), 12 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 32b0522..f4bbd74 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -8496,17 +8496,20 @@ qemuDomainBlockStatsFlags(virDomainPtr dom, if (!(vm = qemuDomObjFromDomain(dom))) goto cleanup; + if (qemuDomainObjBeginJob(driver, vm, QEMU_JOB_QUERY) < 0) + goto cleanup; + if (!virDomainObjIsActive(vm)) { virReportError(VIR_ERR_OPERATION_INVALID, "%s", _("domain is not running")); - goto cleanup; + goto endjob; } if (*nparams != 0) { if ((i = virDomainDiskIndexByName(vm->def, path, false)) < 0) { virReportError(VIR_ERR_INVALID_ARG, _("invalid path: %s"), path); - goto cleanup; + goto endjob; } disk = vm->def->disks[i]; @@ -8514,22 +8517,13 @@ qemuDomainBlockStatsFlags(virDomainPtr dom, virReportError(VIR_ERR_INTERNAL_ERROR, _("missing disk device alias name for %s"), disk->dst); - goto cleanup; + goto endjob; } } priv = vm->privateData; VIR_DEBUG("priv=%p, params=%p, flags=%x", priv, params, flags); - if (qemuDomainObjBeginJob(driver, vm, QEMU_JOB_QUERY) < 0) - goto cleanup; - - if (!virDomainObjIsActive(vm)) { - virReportError(VIR_ERR_OPERATION_INVALID, - "%s", _("domain is not running")); - goto endjob; - } - qemuDomainObjEnterMonitor(driver, vm); tmp = *nparams; ret = qemuMonitorGetBlockStatsParamsNumber(priv->mon, nparams); -- 1.8.1.5

2 2

[libvirt] how virbr0-nic is created
by yue 08 Mar '13

08 Mar '13

libvirt has a default NAT network, it create a bridge based on virbr0-nic, how virbr0-nic is created? thanks

2 1

[libvirt] [PATCH 0/7] Make a minimal client only build work on RHEL4
by Daniel P. Berrange 08 Mar '13

08 Mar '13

This series makes a few changes to let a minimal client only build of libvirt work on RHEL4. Well aside from the last patch which is nothing todo with RHEL4, but happens to be in my queue

3 22

[libvirt] [PATCH] Fix starting qemu instances when apparmor driver is enabled
by Jim Fehlig 08 Mar '13

08 Mar '13

With the apparmor security driver enabled, qemu instances fail to start # grep ^security_driver /etc/libvirt/qemu.conf security_driver = "apparmor" # virsh start test-kvm error: Failed to start domain test-kvm error: internal error security label already defined for VM The model field of virSecurityLabelDef object is always populated by virDomainDefGetSecurityLabelDef(), so remove the check for a NULL model when verifying if a label is already defined for the instance. Checking for a NULL model and populating it later in AppArmorGenSecurityLabel() has been left in the code to be consistent with virSecuritySELinuxGenSecurityLabel(). --- src/security/security_apparmor.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/src/security/security_apparmor.c b/src/security/security_apparmor.c index ddc1fe4..2e6a57f 100644 --- a/src/security/security_apparmor.c +++ b/src/security/security_apparmor.c @@ -436,8 +436,7 @@ AppArmorGenSecurityLabel(virSecurityManagerPtr mgr ATTRIBUTE_UNUSED, return rc; } - if ((secdef->label) || - (secdef->model) || (secdef->imagelabel)) { + if (secdef->label || secdef->imagelabel) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("security label already defined for VM")); @@ -461,8 +460,7 @@ AppArmorGenSecurityLabel(virSecurityManagerPtr mgr ATTRIBUTE_UNUSED, goto err; } - secdef->model = strdup(SECURITY_APPARMOR_NAME); - if (!secdef->model) { + if (!secdef->model && !(secdef->model = strdup(SECURITY_APPARMOR_NAME))) { virReportOOMError(); goto err; } -- 1.8.0.1

3 9

[libvirt] [PATCH] util: distinguish get{pw|gr}nam_r error from the "not found" cases
by Guannan Ren 08 Mar '13

08 Mar '13

manpage: "If no matching record was found, these functions return 0 and store NULL in *result. In case of error, an error number is returned, and NULL is stored in *result." So we can distinguish function error from "no found" cases based on the return value. --- src/util/virutil.c | 46 ++++++++++++++++++++++++---------------------- 1 file changed, 24 insertions(+), 22 deletions(-) diff --git a/src/util/virutil.c b/src/util/virutil.c index 4605c78..9a4682e 100644 --- a/src/util/virutil.c +++ b/src/util/virutil.c @@ -2532,6 +2532,7 @@ virGetUserIDByName(const char *name, uid_t *uid) goto cleanup; } + errno = 0; while ((rc = getpwnam_r(name, &pwbuf, strbuf, strbuflen, &pw)) == ERANGE) { if (VIR_RESIZE_N(strbuf, strbuflen, strbuflen, strbuflen) < 0) { virReportOOMError(); @@ -2540,16 +2541,11 @@ virGetUserIDByName(const char *name, uid_t *uid) } if (!pw) { - if (rc != 0) { - char buf[1024]; - /* log the possible error from getpwnam_r. Unfortunately error - * reporting from this function is bad and we can't really - * rely on it, so we just report that the user wasn't found */ - VIR_WARN("User record for user '%s' was not found: %s", - name, virStrerror(rc, buf, sizeof(buf))); - } - - ret = 1; + if (rc == 0) + ret = 1; + else + virReportSystemError(errno = rc, + _("cannot getpwnam_r(%s)"), name); goto cleanup; } @@ -2577,8 +2573,13 @@ virGetUserID(const char *user, uid_t *uid) user++; } else { int rc = virGetUserIDByName(user, uid); - if (rc <= 0) + if (rc > 0) { + virReportError(VIR_ERR_INVALID_ARG, + _("User '%s' was not found"), user); + return -1; + } else { return rc; + } } if (virStrToLong_ui(user, NULL, 10, &uint_uid) < 0 || @@ -2616,6 +2617,7 @@ virGetGroupIDByName(const char *name, gid_t *gid) goto cleanup; } + errno = 0; while ((rc = getgrnam_r(name, &grbuf, strbuf, strbuflen, &gr)) == ERANGE) { if (VIR_RESIZE_N(strbuf, strbuflen, strbuflen, strbuflen) < 0) { virReportOOMError(); @@ -2624,16 +2626,11 @@ virGetGroupIDByName(const char *name, gid_t *gid) } if (!gr) { - if (rc != 0) { - char buf[1024]; - /* log the possible error from getgrnam_r. Unfortunately error - * reporting from this function is bad and we can't really - * rely on it, so we just report that the user wasn't found */ - VIR_WARN("Group record for user '%s' was not found: %s", - name, virStrerror(rc, buf, sizeof(buf))); - } - - ret = 1; + if (rc == 0) + ret = 1; + else + virReportSystemError(errno = rc, + _("cannot getgrnam_r(%s)"), name); goto cleanup; } @@ -2661,8 +2658,13 @@ virGetGroupID(const char *group, gid_t *gid) group++; } else { int rc = virGetGroupIDByName(group, gid); - if (rc <= 0) + if (rc > 0) { + virReportError(VIR_ERR_INVALID_ARG, + _("Group '%s' was not found"), group); + return -1; + } else { return rc; + } } if (virStrToLong_ui(group, NULL, 10, &uint_gid) < 0 || -- 1.7.11.2

2 2