July 2019 - Devel - libvirt List Archives

[libvirt] Checking the expected behavior on dynamic image file ownership when live migrating
by Christian Ehrhardt 02 Jul '19

02 Jul '19

Hi, today I was debugging an issue that I found with qemu 4.0 and ended up puzzled about file ownership. I'm almost EOD now, but wanted to reach out here for a sanity check before I debug further. The case sumamrized is this: 1. start guest 1.1 image files are changed to libvirt-qemu:kvm (which matches Ubuntus user/group config) 2. live migrate the guest to a different node 2.1 image files go back to root:root which they initially had (ok) 3. migrate guest back to the original node x. image files stay root:root and are not changed back to libvirt-qemu:kvm That is odd/unexpected, but it seems the same applies to older versions and there it was never a problem so far. In my case I stumbled over it because the newer qemu --copy-storage-* options now want to re-open the file at some point and that fails with: error: internal error: unable to execute QEMU command 'drive-mirror': Could not reopen file: Permission denied If I e.g. shut down and start the guest on the source node (which fixes up the ownership at runtime) then migration with copy-storage works fine again. I'm mostly looking for hints if this is a known issue or being worked on to not waste time tomorrow. So hints are welcome. -- Christian Ehrhardt Software Engineer, Ubuntu Server Canonical Ltd

2 4

[libvirt] [PATCH v2] test_driver: implement virDomainGetMemoryParameters
by Ilias Stamatis 02 Jul '19

02 Jul '19

Signed-off-by: Ilias Stamatis <stamatis.iliass(a)gmail.com> --- Things changed since v1: - virDomainObjGetOneDef is used in order to get the appropriate def - the TEST_ASSIGN_MEM_PARAM macro got removed src/test/test_driver.c | 52 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 52 insertions(+) diff --git a/src/test/test_driver.c b/src/test/test_driver.c index 4b1f2724a0..01b3ed88f3 100755 --- a/src/test/test_driver.c +++ b/src/test/test_driver.c @@ -2514,6 +2514,57 @@ testDomainGetMaxVcpus(virDomainPtr domain) VIR_DOMAIN_VCPU_MAXIMUM)); } + +static int +testDomainGetMemoryParameters(virDomainPtr dom, + virTypedParameterPtr params, + int *nparams, + unsigned int flags) +{ + int ret = -1; + virDomainObjPtr vm = NULL; + virDomainDefPtr def = NULL; + + virCheckFlags(VIR_DOMAIN_AFFECT_LIVE | + VIR_DOMAIN_AFFECT_CONFIG | + VIR_TYPED_PARAM_STRING_OKAY, -1); + + if ((*nparams) == 0) { + *nparams = 3; + return 0; + } + + if (!(vm = testDomObjFromDomain(dom))) + goto cleanup; + + if (!(def = virDomainObjGetOneDef(vm, flags))) + goto cleanup; + + if (*nparams > 0 && + virTypedParameterAssign(&params[0], VIR_DOMAIN_MEMORY_HARD_LIMIT, + VIR_TYPED_PARAM_ULLONG, def->mem.hard_limit) < 0) + goto cleanup; + + if (*nparams > 1 && + virTypedParameterAssign(&params[1], VIR_DOMAIN_MEMORY_SOFT_LIMIT, + VIR_TYPED_PARAM_ULLONG, def->mem.soft_limit) < 0) + goto cleanup; + + if (*nparams > 2 && + virTypedParameterAssign(&params[2], VIR_DOMAIN_MEMORY_SWAP_HARD_LIMIT, + VIR_TYPED_PARAM_ULLONG, def->mem.swap_hard_limit) < 0) + goto cleanup; + + if (*nparams > 3) + *nparams = 3; + + ret = 0; + cleanup: + virDomainObjEndAPI(&vm); + return ret; +} + + static int testDomainSetVcpusFlags(virDomainPtr domain, unsigned int nrCpus, unsigned int flags) @@ -7275,6 +7326,7 @@ static virHypervisorDriver testHypervisorDriver = { .domainGetVcpus = testDomainGetVcpus, /* 0.7.3 */ .domainGetVcpuPinInfo = testDomainGetVcpuPinInfo, /* 1.2.18 */ .domainGetMaxVcpus = testDomainGetMaxVcpus, /* 0.7.3 */ + .domainGetMemoryParameters = testDomainGetMemoryParameters, /* 5.6.0 */ .domainGetXMLDesc = testDomainGetXMLDesc, /* 0.1.4 */ .connectListDefinedDomains = testConnectListDefinedDomains, /* 0.1.11 */ .connectNumOfDefinedDomains = testConnectNumOfDefinedDomains, /* 0.1.11 */ -- 2.22.0

2 4

[libvirt] [PATCH v2] test_driver: implement virDomainGetInterfaceParameters
by Ilias Stamatis 02 Jul '19

02 Jul '19

Signed-off-by: Ilias Stamatis <stamatis.iliass(a)gmail.com> --- The only thing changed since v1 is calling virDomainObjGetOneDef to get the appropriate def pointer. src/test/test_driver.c | 104 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 104 insertions(+) diff --git a/src/test/test_driver.c b/src/test/test_driver.c index 4b1f2724a0..dd756ed8bd 100755 --- a/src/test/test_driver.c +++ b/src/test/test_driver.c @@ -1964,6 +1964,109 @@ static int testDomainGetInfo(virDomainPtr domain, return ret; } + +static int +testDomainGetInterfaceParameters(virDomainPtr dom, + const char *device, + virTypedParameterPtr params, + int *nparams, + unsigned int flags) +{ + size_t i; + virDomainObjPtr vm = NULL; + virDomainDefPtr def = NULL; + virDomainNetDefPtr net = NULL; + int ret = -1; + + virCheckFlags(VIR_DOMAIN_AFFECT_LIVE | + VIR_DOMAIN_AFFECT_CONFIG | + VIR_TYPED_PARAM_STRING_OKAY, -1); + + if ((*nparams) == 0) { + *nparams = 7; + return 0; + } + + if (!(vm = testDomObjFromDomain(dom))) + return -1; + + if (!(def = virDomainObjGetOneDef(vm, flags))) + goto cleanup; + + if (!(net = virDomainNetFind(def, device))) + goto cleanup; + + for (i = 0; i < *nparams && i < 7; i++) { + switch (i) { + case 0: /* inbound.average */ + if (virTypedParameterAssign(&params[i], + VIR_DOMAIN_BANDWIDTH_IN_AVERAGE, + VIR_TYPED_PARAM_UINT, 0) < 0) + goto cleanup; + if (net->bandwidth && net->bandwidth->in) + params[i].value.ui = net->bandwidth->in->average; + break; + case 1: /* inbound.peak */ + if (virTypedParameterAssign(&params[i], + VIR_DOMAIN_BANDWIDTH_IN_PEAK, + VIR_TYPED_PARAM_UINT, 0) < 0) + goto cleanup; + if (net->bandwidth && net->bandwidth->in) + params[i].value.ui = net->bandwidth->in->peak; + break; + case 2: /* inbound.burst */ + if (virTypedParameterAssign(&params[i], + VIR_DOMAIN_BANDWIDTH_IN_BURST, + VIR_TYPED_PARAM_UINT, 0) < 0) + goto cleanup; + if (net->bandwidth && net->bandwidth->in) + params[i].value.ui = net->bandwidth->in->burst; + break; + case 3: /* inbound.floor */ + if (virTypedParameterAssign(&params[i], + VIR_DOMAIN_BANDWIDTH_IN_FLOOR, + VIR_TYPED_PARAM_UINT, 0) < 0) + goto cleanup; + if (net->bandwidth && net->bandwidth->in) + params[i].value.ui = net->bandwidth->in->floor; + break; + case 4: /* outbound.average */ + if (virTypedParameterAssign(&params[i], + VIR_DOMAIN_BANDWIDTH_OUT_AVERAGE, + VIR_TYPED_PARAM_UINT, 0) < 0) + goto cleanup; + if (net->bandwidth && net->bandwidth->out) + params[i].value.ui = net->bandwidth->out->average; + break; + case 5: /* outbound.peak */ + if (virTypedParameterAssign(&params[i], + VIR_DOMAIN_BANDWIDTH_OUT_PEAK, + VIR_TYPED_PARAM_UINT, 0) < 0) + goto cleanup; + if (net->bandwidth && net->bandwidth->out) + params[i].value.ui = net->bandwidth->out->peak; + break; + case 6: /* outbound.burst */ + if (virTypedParameterAssign(&params[i], + VIR_DOMAIN_BANDWIDTH_OUT_BURST, + VIR_TYPED_PARAM_UINT, 0) < 0) + goto cleanup; + if (net->bandwidth && net->bandwidth->out) + params[i].value.ui = net->bandwidth->out->burst; + break; + } + } + + if (*nparams > 7) + *nparams = 7; + + ret = 0; + cleanup: + virDomainObjEndAPI(&vm); + return ret; +} + + static int testDomainGetState(virDomainPtr domain, int *state, @@ -7258,6 +7361,7 @@ static virHypervisorDriver testHypervisorDriver = { .domainSetMemory = testDomainSetMemory, /* 0.1.4 */ .domainGetHostname = testDomainGetHostname, /* 5.5.0 */ .domainGetInfo = testDomainGetInfo, /* 0.1.1 */ + .domainGetInterfaceParameters = testDomainGetInterfaceParameters, /* 5.6.0 */ .domainGetState = testDomainGetState, /* 0.9.2 */ .domainGetTime = testDomainGetTime, /* 5.4.0 */ .domainSave = testDomainSave, /* 0.3.2 */ -- 2.22.0

2 3

[libvirt] [PATCH 0/2] Couple of test_driver fixes
by Michal Privoznik 02 Jul '19

02 Jul '19

I was too eager to minimize patch 89320788ac4 and did some mistakes. Here are the fixes. Michal Prívozník (2): test_driver: Don't access @vm after it was set to NULL test_driver: Don't report VIR_DOMAIN_DISK_ERROR_NONE src/test/test_driver.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) -- 2.21.0

2 5

[libvirt] [PATCH v2 8/9] docs: Deprecate CPU model runnability guarantees
by Eduardo Habkost 02 Jul '19

02 Jul '19

Document that CPU model runnability guarantees won't apply to unversioned CPU models anymore. Signed-off-by: Eduardo Habkost <ehabkost(a)redhat.com> --- Changes v1 -> v2: * (none) Cc: libvir-list(a)redhat.com --- qemu-deprecated.texi | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/qemu-deprecated.texi b/qemu-deprecated.texi index 9cba82d5ec..18f85f70e1 100644 --- a/qemu-deprecated.texi +++ b/qemu-deprecated.texi @@ -276,3 +276,22 @@ Note that if you are exposing the export via /dev/nbd0, it is easier to just export the entire image and then mount only /dev/nbd0p1 than it is to reinvoke @command{qemu-nbd -c /dev/nbd0} limited to just a subset of the image. + +@section Backwards compatibility + +@subsection Runnability guarantee of CPU models (since 4.1.0) + +Previous versions of QEMU never changed existing CPU models in +ways that introduced additional host software or hardware +requirements to the VM. This allowed management software to +safely change the machine type of an existing VM without +introducing new requirements ("runnability guarantee"). This +prevented CPU models from being updated to include CPU +vulnerability mitigations, leaving guests vulnerable in the +default configuration. + +The CPU model runnability guarantee won't apply anymore to +existing CPU models. Management software that needs runnability +guarantees must resolve the CPU model aliases using te +``alias-of'' field returned by the ``query-cpu-definitions'' QMP +command. -- 2.18.0.rc1.1.g3f1ff2140

2 1

[libvirt] [PATCH] qemu: remove a redundant call from qemuDomainGetPerfEvents
by Ilias Stamatis 02 Jul '19

02 Jul '19

Calling virDomainObjUpdateModificationImpact directly inside the function body is redundant, since the same function call is embedded into virDomainObjGetOneDef. Signed-off-by: Ilias Stamatis <stamatis.iliass(a)gmail.com> --- src/qemu/qemu_driver.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index ef2e980216..6bfbc31c1c 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -10339,9 +10339,6 @@ qemuDomainGetPerfEvents(virDomainPtr dom, if (qemuDomainObjBeginJob(driver, vm, QEMU_JOB_QUERY) < 0) goto cleanup; - if (virDomainObjUpdateModificationImpact(vm, &flags) < 0) - goto endjob; - if (!(def = virDomainObjGetOneDef(vm, flags))) goto endjob; -- 2.22.0

2 1

[libvirt] mdevctl: A shoestring mediated device management and persistence utility
by Alex Williamson 02 Jul '19

02 Jul '19

Hi, Currently mediated device management, much like SR-IOV VF management, is largely left as an exercise for the user. This is an attempt to provide something and see where it goes. I doubt we'll solve everyone's needs on the first pass, but maybe we'll solve enough and provide helpers for the rest. Without further ado, I'll point to what I have so far: https://github.com/awilliam/mdevctl This is inspired by driverctl, which is also a bash utility. mdevctl uses udev and systemd to record and recreate mdev devices for persistence and provides a command line utility for querying, listing, starting, stopping, adding, and removing mdev devices. Currently, for better or worse, it considers anything created to be persistent. I can imagine a global configuration option that might disable this and perhaps an autostart flag per mdev device, such that mdevctl might simply "know" about some mdevs but not attempt to create them automatically. Clearly command line usage help, man pages, and packaging are lacking as well, release early, release often, plus this is a discussion starter to see if perhaps this is sufficient to meet some needs. Originally I thought about making a utility to manage both mdev and SR-IOV VFs all in one, but it seemed more natural to start here (besides, I couldn't think of a good name for the combined utility). If this seems useful, maybe I'll start on a vfctl for SR-IOV and we'll see whether they have enough synergy to become one. It would be really useful if s390 folks could help me understand whether it's possible to glean all the information necessary to recreate a ccw or ap mdev device from sysfs. I expect the file where we currently only store the mdev_type to evolve into something that includes more information to facilitate more complicated devices. For now I make no claims to maintaining compatibility of recorded mdev devices, it will absolutely change, but I didn't want to get bogged down in making sure I don't accidentally source a root kit hidden in an mdev config file. I'm also curious how or if libvirt or openstack might use this. If nothing else, it makes libvirt hook scripts easier to write, especially if we add an option not to autostart mdevs, or if users don't mind persistent mdevs, maybe there's nothing more to do. BTW, feel free to clean up by bash, I'm a brute force and ignorance shell coder ;) Thanks, Alex

7 50

[libvirt] [PATCH] rpc: Fix build error for virNetServerNew ATTRIBUTE_NONNULL values
by John Ferlan 02 Jul '19

02 Jul '19

Commit 5a148ce84 altered the virNetServerNew to remove a parameter but neglected to update the ATTRIBUTE_NONNULL's which causes a build failure for when checking is enabled such as when lv_cv_static_analysis is enabled. Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- Pushing as build breaker/trivial src/rpc/virnetserver.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/rpc/virnetserver.h b/src/rpc/virnetserver.h index b47b71b4b2..6b2541588c 100644 --- a/src/rpc/virnetserver.h +++ b/src/rpc/virnetserver.h @@ -42,7 +42,7 @@ virNetServerPtr virNetServerNew(const char *name, virNetServerClientPrivPreExecRestart clientPrivPreExecRestart, virFreeCallback clientPrivFree, void *clientPrivOpaque) - ATTRIBUTE_NONNULL(1) ATTRIBUTE_NONNULL(11) ATTRIBUTE_NONNULL(13); + ATTRIBUTE_NONNULL(1) ATTRIBUTE_NONNULL(10) ATTRIBUTE_NONNULL(12); virNetServerPtr virNetServerNewPostExecRestart(virJSONValuePtr object, const char *name, -- 2.20.1

1 0

[libvirt] [PATCH] tools: Fix permissions for virt-pki-validate.in
by Andrea Bolognani 02 Jul '19

02 Jul '19

While the script ultimately needs to be executable, the source file really shouldn't be. Signed-off-by: Andrea Bolognani <abologna(a)redhat.com> --- Pushed as trivial. tools/virt-pki-validate.in | 0 1 file changed, 0 insertions(+), 0 deletions(-) mode change 100755 => 100644 tools/virt-pki-validate.in diff --git a/tools/virt-pki-validate.in b/tools/virt-pki-validate.in old mode 100755 new mode 100644 -- 2.21.0

1 0

[libvirt] [PATCH] test_driver: Fix permissions for test_driver.c
by Andrea Bolognani 02 Jul '19

02 Jul '19

Introduced in commit 4a6ee53581b3. Signed-off-by: Andrea Bolognani <abologna(a)redhat.com> --- Pushed as trivial. src/test/test_driver.c | 0 1 file changed, 0 insertions(+), 0 deletions(-) mode change 100755 => 100644 src/test/test_driver.c diff --git a/src/test/test_driver.c b/src/test/test_driver.c old mode 100755 new mode 100644 -- 2.21.0

1 0