May 2020 - Devel - libvirt List Archives

[PATCH 0/2] Introduce GitLab CI and merge requests
by Daniel P. Berrangé 18 May '20

18 May '20

In v2: - Use ubuntu for publishing pages build Daniel P. Berrangé (2): gitlab: introduce CI jobs for building content gitlab: add CONTRIBUTING.rst file to indicate use of merge requests .gitlab-ci.yml | 126 +++++++++++++++++++++++++++ .gitpublish | 4 - CONTRIBUTING.rst | 28 ++++++ Makefile | 7 +- ci/libvirt-centos-7.Dockerfile | 83 ++++++++++++++++++ ci/libvirt-debian-10.Dockerfile | 53 +++++++++++ ci/libvirt-debian-9.Dockerfile | 56 ++++++++++++ ci/libvirt-debian-sid.Dockerfile | 53 +++++++++++ ci/libvirt-fedora-31.Dockerfile | 50 +++++++++++ ci/libvirt-fedora-32.Dockerfile | 50 +++++++++++ ci/libvirt-fedora-rawhide.Dockerfile | 51 +++++++++++ ci/libvirt-ubuntu-1804.Dockerfile | 56 ++++++++++++ ci/libvirt-ubuntu-2004.Dockerfile | 53 +++++++++++ ci/refresh | 22 +++++ 14 files changed, 685 insertions(+), 7 deletions(-) delete mode 100644 .gitpublish create mode 100644 CONTRIBUTING.rst create mode 100644 ci/libvirt-centos-7.Dockerfile create mode 100644 ci/libvirt-debian-10.Dockerfile create mode 100644 ci/libvirt-debian-9.Dockerfile create mode 100644 ci/libvirt-debian-sid.Dockerfile create mode 100644 ci/libvirt-fedora-31.Dockerfile create mode 100644 ci/libvirt-fedora-32.Dockerfile create mode 100644 ci/libvirt-fedora-rawhide.Dockerfile create mode 100644 ci/libvirt-ubuntu-1804.Dockerfile create mode 100644 ci/libvirt-ubuntu-2004.Dockerfile create mode 100755 ci/refresh -- 2.26.2

2 5

[PATCH 00/21] Fix compatibility with netdev-add in qemu-5.0
by Peter Krempa 18 May '20

18 May '20

This is based on top of tests: qemu: Detect deprecation in the QMP schema (deprecation part 1) https://www.redhat.com/archives/libvir-list/2020-April/msg01444.html Both can be fetched at: git fetch https://gitlab.com/pipo.sk/libvirt.git netdev_add-validate QEMU-5.0 released a fully QAPIfied 'netdev_add' command. Since we didn't format the arguments in a way qemu now thinks we should do them network device hotplug broke. Fix it by modifying our internals slightly and add some testing. This series also adds schema testing of -blockdev arguments and prepares for a simple addition of schema testing for arguments of -device and -object if qemu ever finishes the qapification of those. Theoretically we can even do it pre-emtptively but it's almost guaranteed to fail at least for -device. I just hope that qemu let's us know in advance as -device will be harder. Note that if the loading of the schema in the tests I've added will be deemed to be slow I'll add caching. Peter Krempa (21): qemu: domain: Forbid unsupported 'tftp' protocol and handle tests qemuMonitorJSONParseKeywords: remove constant argument qemuBuildChannelsCommandLine: Use typecasted switch for channel type qemuBuildChannelsCommandLine: Extract common formatting of 'chardev' qemuBuildChannelChrDeviceStr: Remove formatting of properties for -netdev qemuBuildHostNetStr: Stop using 'ipv6-net' convenience argument virQEMUBuildCommandLineJSON: Allow skipping certain keys virQEMUBuildCommandLineJSON: Add possibility for using 'on/off' instead of 'yes/no' virCommand: Introduce virCommandGetArgList util: json: Introduce virJSONValueObjectAppendStringPrintf testCompareXMLToArgv: Split out preparation and command formatting qemuMonitorJSON(Add|Remove)Netdev: Refactor cleanup util: virqemu: Introduce virQEMUBuildNetdevCommandlineFromJSON qemu: command: Generate -netdev command line via JSON->cmdline conversion virQEMUBuildNetdevCommandlineFromJSON: Prepare for quirky 'guestfwd' qemuBuildChannelGuestfwdNetdevProps: Convert to generating JSON props qemuMonitorAddNetdev: Convert to the native JSON props object qemu: Prepare for testing of 'netdev_add' props via qemuxml2argvtest testQEMUSchemaLoad: Rename to testQEMUSchemaLoadLatest testutilsqemuschema: Allow loading non-latest schema qemuxml2argvtest: Add QAPI/QMP schema validation for -blockdev and -netdev src/libvirt_private.syms | 3 + src/qemu/qemu_command.c | 289 +++++++++++------- src/qemu/qemu_command.h | 22 +- src/qemu/qemu_domain.c | 10 + src/qemu/qemu_driver.c | 2 +- src/qemu/qemu_hotplug.c | 31 +- src/qemu/qemu_monitor.c | 8 +- src/qemu/qemu_monitor.h | 2 +- src/qemu/qemu_monitor_json.c | 76 ++--- src/qemu/qemu_monitor_json.h | 2 +- src/qemu/qemu_process.c | 11 +- src/qemu/qemu_process.h | 1 + src/util/vircommand.c | 23 ++ src/util/vircommand.h | 1 + src/util/virjson.c | 17 ++ src/util/virjson.h | 2 + src/util/virqemu.c | 119 +++++++- src/util/virqemu.h | 18 +- tests/Makefile.am | 2 +- tests/qemublocktest.c | 2 +- tests/qemucommandutiltest.c | 2 +- tests/qemuhotplugtest.c | 2 +- tests/qemumonitorjsontest.c | 6 +- .../qemuxml2argvdata/disk-cdrom-network.args | 3 - .../disk-cdrom-network.x86_64-2.12.0.args | 3 - .../disk-cdrom-network.x86_64-latest.args | 17 +- tests/qemuxml2argvdata/disk-cdrom-network.xml | 9 - tests/qemuxml2argvdata/net-user-addr.args | 3 +- tests/qemuxml2argvtest.c | 236 +++++++++----- tests/testutilsqemu.c | 5 + tests/testutilsqemu.h | 1 + tests/testutilsqemuschema.c | 66 ++-- tests/testutilsqemuschema.h | 5 +- 33 files changed, 667 insertions(+), 332 deletions(-) -- 2.26.2

2 43

[libvirt PATCH 0/2] fix a leak in virQEMUDriverConfigLoadMemoryEntry
by Ján Tomko 18 May '20

18 May '20

Thanks to valgrind Ján Tomko (2): qemu: do not leak the directory in virQEMUDriverConfigLoadMemoryEntry qemu: only return two values in virQEMUDriverConfigLoadMemoryEntry src/qemu/qemu_conf.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) -- 2.25.4

2 3

[PATCH v2] qemuProcessStop: Reattach NVMe disks a domain is mirroring into
by Michal Privoznik 18 May '20

18 May '20

In v5.10.0-rc1~42 (which was later fixed in v6.0.0-rc1~487) I am removing XATTRs for a file that QEMU is mirroring a disk into but it is killed meanwhile. Well, we can call qemuSecurityRestoreImageLabel() which will not only remove XATTRs but also use them to restore the original owner of the file. Moreover, if the mirror destination is not a file but a NVMe disk, then call qemuHostdevReAttachOneNVMeDisk() to reattach the NVMe back to the host. All of this would be done by blockjob code when the job finishes, but in this case the job won't finish - QEMU is killed meanwhile. Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1825785 Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- v2 of: https://www.redhat.com/archives/libvir-list/2020-May/msg00210.html diff to v1: - per Peter's review, instead of calling qemuDomainStorageSourceAccessRevoke() call only some functions it would call. src/qemu/qemu_process.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 8ea470f75f..65d51e58c9 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -7604,8 +7604,13 @@ void qemuProcessStop(virQEMUDriverPtr driver, for (i = 0; i < def->ndisks; i++) { virDomainDiskDefPtr disk = def->disks[i]; - if (disk->mirror) - qemuBlockRemoveImageMetadata(driver, vm, disk->dst, disk->mirror); + if (disk->mirror) { + if (qemuSecurityRestoreImageLabel(driver, vm, disk->mirror, false) < 0) + VIR_WARN("Unable to restore security label on %s", disk->dst); + + if (virStorageSourceChainHasNVMe(disk->mirror)) + qemuHostdevReAttachOneNVMeDisk(driver, vm->def->name, disk->mirror); + } qemuBlockRemoveImageMetadata(driver, vm, disk->dst, disk->src); } -- 2.26.2

2 1

[libvirt PATCH v2 00/10] qemu: block: basic implementation of deletion of external snapshots
by Pavel Mores 18 May '20

18 May '20

Deleting external snapshots has been unimplemented so far. This series aims to enable limited functionality in that direction, handling just the common cases for now. The intention is to subsequently build upon this to eventually cover the more complex/exotic cases as well. It works by blockcommitting the snapshot to be deleted into its parent. Handles both deleting children as well and deleting children only (by committing the children into the snapshot referred to by snapshot-delete argument). If the necessary block commit operations turns out to be active (the snapshot referred to by snapshot-delete argument is current, or deleting children is requested) pivot is performed as well. This implemetation is limited to the straightforward case which assumes no branching snapshot chains below the snapshot to be deleted, and the current snapshot has to be the leaf of the (linear) chain starting at the snapshot to be deleted. These requirements are checked and enforced at the top of qemuDomainSnapshotDeleteExternal(). They might even be too restrictive for the case where deletion of children is not requested as in that case, requiring that the parent of the snapshot to be deleted not be a branching point in snapshot tree should be enough. The above limitations do not appear too severe under the current state of matters where a major source of branching snapshot structures, snapshot-revert, is not even implemented for external snapshots yet. The only other known cause of branching in external snapshots thus seems to be if a user creates branching by manually creating snapshot images and metadata and applies them using snapshot-create --redefine. Also, the snapshot images involved basically have to be files. This does not seem overly limiting either, at least for the time being, because it's a limitation shared with the underlying blockcommit code. Just as the existing code, the new code ultimately relies on virStorageFileChainLookup() as its image comparison and look-up engine and is affected by its limitaions. At any rate, this work should be understood as just a first step to a full support of deleting external snapshots. The first 5 commits of this series are just refactors used by the last 5 commits which actually implement snapshot deletion. However, the qemuDomainBlockCommit() refactor (the initial 2 commits of this series) was affected by a rather hairy merge on rebasing to the current master and although I tried to do my best to get the merge right I still feel a careful review is in order for these two commits. Pavel Mores (10): qemu: block: factor implementation out of qemuDomainBlockCommit() qemu: block: refactor blockcommit so that it's callable with storage sources qemu: block: factor implementation out of qemuDomainBlockJobAbort() conf: rename virDomainMomentFindLeaf() to virDomainMomentObjListFindLeaf() conf: add virDomainMomentFindLeaf() which operates on virDomainMomentObjPtr qemu: block: add function to collect blockcommit params and verify them qemu: block: add function to launch all prepared blockcommits qemu: block: add function to wait for blockcommits and collect results qemu: block: add external snapshot-delete top-level algorithm qemu: block: add actual invocation of external snapshot-delete src/conf/virdomaincheckpointobjlist.c | 2 +- src/conf/virdomainmomentobjlist.c | 27 +- src/conf/virdomainmomentobjlist.h | 3 +- src/libvirt_private.syms | 1 + src/qemu/qemu_driver.c | 552 ++++++++++++++++++++++---- 5 files changed, 485 insertions(+), 100 deletions(-) -- 2.24.1

2 15

[PATCH] RFC: Support network interface downscript
by Chen Hanxiao 18 May '20

18 May '20

https://gitlab.com/libvirt/libvirt/-/issues/13 Add support for downscript: <interface type='ethernet'> <mac address='00:11:22:33:44:55'/> <script path='/etc/qemu-ifup'/> <downscript path='/path/to/my/downscript'/> </interface> But -net ...,script=/etc/qemu-ifup has some limitation: https://github.com/qemu/qemu/blob/035b448b84f3557206abc44d786c5d3db2638f7d/… Maybe virNetDevRunEthernetScript is another choice. Signed-off-by: Chen Hanxiao <chen_han_xiao(a)126.com> --- src/conf/domain_conf.c | 7 +++++++ src/conf/domain_conf.h | 1 + src/qemu/qemu_extdevice.c | 2 ++ 3 files changed, 10 insertions(+) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index c201fc9..5cb39a3 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -2520,6 +2520,7 @@ virDomainNetDefClear(virDomainNetDefPtr def) VIR_FREE(def->teaming.persistent); VIR_FREE(def->virtPortProfile); VIR_FREE(def->script); + VIR_FREE(def->downscript); VIR_FREE(def->domain_name); VIR_FREE(def->ifname); VIR_FREE(def->ifname_guest); @@ -11977,6 +11978,7 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, g_autofree char *ifname_guest = NULL; g_autofree char *ifname_guest_actual = NULL; g_autofree char *script = NULL; + g_autofree char *downscript = NULL; g_autofree char *address = NULL; g_autofree char *port = NULL; g_autofree char *localaddr = NULL; @@ -12149,6 +12151,9 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, } else if (!script && virXMLNodeNameEqual(cur, "script")) { script = virXMLPropString(cur, "path"); + } else if (!downscript && + virXMLNodeNameEqual(cur, "downscript")) { + downscript = virXMLPropString(cur, "path"); } else if (!domain_name && virXMLNodeNameEqual(cur, "backenddomain")) { domain_name = virXMLPropString(cur, "name"); @@ -12482,6 +12487,8 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, if (script != NULL) def->script = g_steal_pointer(&script); + if (downscript != NULL) + def->downscript = g_steal_pointer(&downscript); if (domain_name != NULL) def->domain_name = g_steal_pointer(&domain_name); if (ifname != NULL) diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index ddc75d8..e152c59 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -1055,6 +1055,7 @@ struct _virDomainNetDef { unsigned long sndbuf; } tune; char *script; + char *downscript; char *domain_name; /* backend domain name */ char *ifname; /* interface name on the host (<target dev='x'/>) */ int managed_tap; /* enum virTristateBool - ABSENT == YES */ diff --git a/src/qemu/qemu_extdevice.c b/src/qemu/qemu_extdevice.c index 2096272..f8e440a 100644 --- a/src/qemu/qemu_extdevice.c +++ b/src/qemu/qemu_extdevice.c @@ -234,6 +234,8 @@ qemuExtDevicesStop(virQEMUDriverPtr driver, if (slirp) qemuSlirpStop(slirp, vm, driver, net); + if (net->downscript) + virNetDevRunEthernetScript(net->ifname, net->downscript); } for (i = 0; i < def->nfss; i++) { -- 1.8.3.1

2 1

[PATCH] Unlock the storage pool objects after looking it up
by Yi Li 18 May '20

18 May '20

The lock should be released. Signed-off-by: Yi Li <yili(a)winhong.com> --- src/conf/virstorageobj.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/src/conf/virstorageobj.c b/src/conf/virstorageobj.c index 13b75b648d..9f24ae67ca 100644 --- a/src/conf/virstorageobj.c +++ b/src/conf/virstorageobj.c @@ -2077,8 +2077,10 @@ virStoragePoolObjListExport(virConnectPtr conn, virObjectRWLockRead(poolobjs); - if (pools && VIR_ALLOC_N(data.pools, virHashSize(poolobjs->objs) + 1) < 0) - goto error; + if (pools && VIR_ALLOC_N(data.pools, virHashSize(poolobjs->objs) + 1) < 0) { + virObjectRWUnlock(poolobjs); + return -1; + } virHashForEach(poolobjs->objs, virStoragePoolObjListExportCallback, &data); virObjectRWUnlock(poolobjs); -- 2.25.3

4 5

[PATCH trivial 1/1] qemu_alias.c: fix qemuAssingDeviceMemballoonAlias() typo
by Daniel Henrique Barboza 18 May '20

18 May '20

Signed-off-by: Daniel Henrique Barboza <danielhb413(a)gmail.com> --- src/qemu/qemu_alias.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_alias.c b/src/qemu/qemu_alias.c index b0ea62af39..d6527cb046 100644 --- a/src/qemu/qemu_alias.c +++ b/src/qemu/qemu_alias.c @@ -390,7 +390,7 @@ qemuAssignDeviceSmartcardAlias(virDomainSmartcardDefPtr smartcard, static int -qemuAssingDeviceMemballoonAlias(virDomainMemballoonDefPtr memballoon, +qemuAssignDeviceMemballoonAlias(virDomainMemballoonDefPtr memballoon, int idx) { if (memballoon->info.alias) @@ -662,7 +662,7 @@ qemuAssignDeviceAliases(virDomainDefPtr def, virQEMUCapsPtr qemuCaps) } if (def->memballoon && def->memballoon->model != VIR_DOMAIN_MEMBALLOON_MODEL_NONE) { - if (qemuAssingDeviceMemballoonAlias(def->memballoon, 0) < 0) + if (qemuAssignDeviceMemballoonAlias(def->memballoon, 0) < 0) return -1; } for (i = 0; i < def->nrngs; i++) { -- 2.26.2

2 1

[PATCH v3 0/9] Introducing TPM Proxy device support for PPC64
by Daniel Henrique Barboza 15 May '20

15 May '20

changes in v3 after reviews from Stefan Berger: - new patch 4: trivial change to allow g_autoptr() with virDomainTPMDefPtr - patch 5 (former 4): added g_autoptr(), removed VIR_FREE() calls - changed wording in patches 1 and 9 (former 8) to make it clear for users that this new device brings no benefits unless the guest is running on specialized PPC64 hardware - fixed identation in patch 6 (former 5). Added Stefan's r-b v2 link: https://www.redhat.com/archives/libvir-list/2020-May/msg00604.html v1 link: https://www.redhat.com/archives/libvir-list/2020-May/msg00604.html Daniel Henrique Barboza (9): docs: documentation and schema for the new TPM Proxy model qemu: Extend QEMU capabilities with 'spapr-tpm-proxy' conf, qemu: adding 'tpmproxy' in domain definition conf, domain: register AUTOPTR_CLEANUP_FUNC for virDomainTPMDef domain_conf.c: XML parsing for VIR_DOMAIN_TPM_MODEL_SPAPR_PROXY tests: add XML schema tests for the TPM Proxy device qemu: build command line for the TPM Proxy device tests/qemuxml2argvtest.c: add TPM Proxy command line tests docs/news.xml: update for the new TPM Proxy device docs/formatdomain.html.in | 18 +++++- docs/news.xml | 17 +++++ docs/schemas/domaincommon.rng | 1 + src/conf/domain_audit.c | 3 + src/conf/domain_conf.c | 63 +++++++++++++++++-- src/conf/domain_conf.h | 3 + src/qemu/qemu_alias.c | 16 +++++ src/qemu/qemu_capabilities.c | 4 ++ src/qemu/qemu_capabilities.h | 3 + src/qemu/qemu_cgroup.c | 12 +++- src/qemu/qemu_command.c | 21 +++++++ src/qemu/qemu_domain.c | 9 +-- src/qemu/qemu_validate.c | 12 ++++ src/security/security_dac.c | 14 +++++ src/security/security_selinux.c | 11 ++++ .../qemucapabilitiesdata/caps_4.2.0.ppc64.xml | 1 + .../qemucapabilitiesdata/caps_5.0.0.ppc64.xml | 1 + tests/qemuxml2argvdata/ppc64-tpm-double.xml | 34 ++++++++++ .../ppc64-tpmproxy-double.xml | 38 +++++++++++ .../ppc64-tpmproxy-single.ppc64-latest.args | 34 ++++++++++ .../ppc64-tpmproxy-single.xml | 33 ++++++++++ .../ppc64-tpmproxy-with-tpm.ppc64-latest.args | 37 +++++++++++ .../ppc64-tpmproxy-with-tpm.xml | 36 +++++++++++ tests/qemuxml2argvtest.c | 15 +++++ .../ppc64-tpmproxy-single.ppc64-latest.xml | 42 +++++++++++++ .../ppc64-tpmproxy-with-tpm.ppc64-latest.xml | 46 ++++++++++++++ tests/qemuxml2xmltest.c | 2 + 27 files changed, 514 insertions(+), 12 deletions(-) create mode 100644 tests/qemuxml2argvdata/ppc64-tpm-double.xml create mode 100644 tests/qemuxml2argvdata/ppc64-tpmproxy-double.xml create mode 100644 tests/qemuxml2argvdata/ppc64-tpmproxy-single.ppc64-latest.args create mode 100644 tests/qemuxml2argvdata/ppc64-tpmproxy-single.xml create mode 100644 tests/qemuxml2argvdata/ppc64-tpmproxy-with-tpm.ppc64-latest.args create mode 100644 tests/qemuxml2argvdata/ppc64-tpmproxy-with-tpm.xml create mode 100644 tests/qemuxml2xmloutdata/ppc64-tpmproxy-single.ppc64-latest.xml create mode 100644 tests/qemuxml2xmloutdata/ppc64-tpmproxy-with-tpm.ppc64-latest.xml -- 2.26.2

3 21

[libvirt PATCH] qemu: prevent attempts to detach a device on a controller with hotplug='off'
by Laine Stump 15 May '20

15 May '20

Although the original patches to support controllers with hotplug='off' were checking during hotplug/attach requests that the device was being plugged into a PCI controller that didn't have hotplug disabled, but I forgot to do the same for device detach (the main impetus for adding the feature was to prevent unplugs originating from within the guest, so it slipped my mind). So although the guest OS was ultimately unable to honor the unplug request, libvirt could still be used to make such a request, and since device attach/detach are asynchronous operations, the caller to libvirt would receive a success status back (the device would stubbornly/correctly remain in the domain status XML however) This patch remedies that, by looking at the controller for the device in the detach request, and immediately failing the operation if that controller has hotplug=off. r changes. Lines starting Signed-off-by: Laine Stump <laine(a)redhat.com> --- src/qemu/qemu_hotplug.c | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index ab5a7aef84..5fe125b1a7 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -5891,6 +5891,33 @@ qemuDomainDetachDeviceLive(virDomainObjPtr vm, return -1; } + if (info->type == VIR_DOMAIN_DEVICE_ADDRESS_TYPE_PCI) { + int controllerIdx = virDomainControllerFind(vm->def, + VIR_DOMAIN_CONTROLLER_TYPE_PCI, + info->addr.pci.bus); + if (controllerIdx < 0) { + virReportError(VIR_ERR_OPERATION_FAILED, + _("cannot hot unplug %s device with PCI guest address: " + VIR_PCI_DEVICE_ADDRESS_FMT + " - controller not found"), + virDomainDeviceTypeToString(detach.type), + info->addr.pci.domain, info->addr.pci.bus, + info->addr.pci.slot, info->addr.pci.function); + return -1; + } + + if (vm->def->controllers[controllerIdx]->opts.pciopts.hotplug + == VIR_TRISTATE_SWITCH_OFF) { + virReportError(VIR_ERR_OPERATION_FAILED, + _("cannot hot unplug %s device with PCI guest address: " + VIR_PCI_DEVICE_ADDRESS_FMT + " - not allowed by controller"), + virDomainDeviceTypeToString(detach.type), + info->addr.pci.domain, info->addr.pci.bus, + info->addr.pci.slot, info->addr.pci.function); + return -1; + } + } /* * Issue the qemu monitor command to delete the device (based on * its alias), and optionally wait a short time in case the -- 2.25.4

3 2