- Devel - libvirt List Archives

[libvirt] [PATCH 00/24] Move all NUMA related configuration into one structure
by Peter Krempa 20 Feb '15

20 Feb '15

Due to historical madne^Wreasons the NUMA configuration is split between /domain/cpu and /domain/numatune. Internally the data was also split into two places. We can't do anything about the external representation but we certainly can store all the definitions in one place internally. This series does that. Peter Krempa (24): conf: Move numatune_conf to numa_conf conf: Move NUMA cell parsing code from cpu conf to numa conf conf: Refactor virDomainNumaDefCPUParseXML conf: numa: Don't duplicate NUMA cell cpumask conf: Move NUMA cell formatter to numa_conf conf: numa: Rename virDomainNumatune to virDomainNuma conf: Move enum virMemAccess to the NUMA code and rename it conf: numa: Recalculate rather than remember total NUMA cpu count conf: numa: Improve error message in case a numa node doesn't have cpus conf: numa: Reformat virDomainNumatuneParseXML conf: numa: Refactor logic in virDomainNumatuneParseXML conf: numa: Format <numatune> XML only if necessary conf: Separate helper for creating domain objects conf: Allocate domain definition with the new helper conf: numa: Always allocate the NUMA config conf: numa: Avoid re-allocation of the NUMA conf numa: conf: Tweak parameters of virDomainNumatuneSet conf: numa: Don't pass double pointer to virDomainNumatuneParseXML qemu: command: Unify retrieval of NUMA cell count in qemuBuildNumaArgStr conf: numa: Add helper to get guest NUMA node count and refactor users conf: numa: Add accesor for the NUMA node cpu mask conf: numa: Add accessor to NUMA node's memory access mode conf: numa: Add setter/getter for NUMA node memory size conf: Move all NUMA configuration to virDomainNuma po/POTFILES.in | 2 +- src/Makefile.am | 2 +- src/conf/cpu_conf.c | 151 +------- src/conf/cpu_conf.h | 25 +- src/conf/domain_conf.c | 59 ++- src/conf/domain_conf.h | 11 +- src/conf/{numatune_conf.c => numa_conf.c} | 429 +++++++++++++++------ src/conf/{numatune_conf.h => numa_conf.h} | 77 ++-- src/cpu/cpu.c | 2 +- src/libvirt_private.syms | 13 +- src/lxc/lxc_cgroup.c | 4 +- src/lxc/lxc_controller.c | 6 +- src/lxc/lxc_native.c | 4 +- src/openvz/openvz_conf.c | 2 +- src/parallels/parallels_sdk.c | 4 +- src/phyp/phyp_driver.c | 2 +- src/qemu/qemu_cgroup.c | 12 +- src/qemu/qemu_command.c | 52 +-- src/qemu/qemu_driver.c | 20 +- src/qemu/qemu_process.c | 4 +- src/vbox/vbox_common.c | 8 +- src/vmx/vmx.c | 2 +- src/xen/xen_hypervisor.c | 8 +- src/xen/xend_internal.c | 4 +- src/xen/xm_internal.c | 4 +- src/xenconfig/xen_sxpr.c | 2 +- src/xenconfig/xen_xl.c | 2 +- src/xenconfig/xen_xm.c | 2 +- tests/cputest.c | 2 +- tests/openvzutilstest.c | 2 +- .../qemuxml2argv-numatune-memnode.xml | 2 +- .../qemuxml2xmlout-numatune-memnode.xml | 2 +- tests/securityselinuxtest.c | 2 +- tests/testutilsqemu.c | 4 - 34 files changed, 503 insertions(+), 424 deletions(-) rename src/conf/{numatune_conf.c => numa_conf.c} (60%) rename src/conf/{numatune_conf.h => numa_conf.h} (50%) -- 2.2.2

2 58

[libvirt] [PATCH v2] libvirt-guests: Allow time sync on guests resume
by Michal Privoznik 20 Feb '15

20 Feb '15

Well, imagine domains were running, and as the host went down, they were managesaved. Later, after some time, the host went up again and domains got restored. But without correct time. And depending on how long was the host shut off, it may take some time for ntp to sync the time too. But hey, wait a minute. We have an API just for that! So: 1) Introduce SYNC_TIME variable in libvirt-guests.sysconf to allow users control over the new functionality 2) Call 'virsh domtime --sync $dom' in the libvirt-guests script. Unfortunately, this is all-or-nothing approach (just like anything else with the script). Domains are required to have configured and running qemu-ga inside. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- diff to v1: -ignore the return value of domtime command tools/libvirt-guests.sh.in | 5 +++++ tools/libvirt-guests.sysconf | 7 +++++++ 2 files changed, 12 insertions(+) diff --git a/tools/libvirt-guests.sh.in b/tools/libvirt-guests.sh.in index 1b17bbe..9aa06fa 100644 --- a/tools/libvirt-guests.sh.in +++ b/tools/libvirt-guests.sh.in @@ -171,7 +171,9 @@ start() { isfirst=true bypass= + sync_time=false test "x$BYPASS_CACHE" = x0 || bypass=--bypass-cache + test "x$SYNC_TIME" = x0 || sync_time=true while read uri list; do configured=false set -f @@ -206,6 +208,9 @@ start() { retval run_virsh "$uri" start $bypass "$name" \ >/dev/null && \ gettext "done"; echo + if "$sync_time"; then + run_virsh "$uri" domtime --sync "$name" >/dev/null + fi fi fi done diff --git a/tools/libvirt-guests.sysconf b/tools/libvirt-guests.sysconf index d1f2051..03e732f 100644 --- a/tools/libvirt-guests.sysconf +++ b/tools/libvirt-guests.sysconf @@ -39,3 +39,10 @@ # restoring guests, even though this may give slower operation for # some file systems. #BYPASS_CACHE=0 + +# If non-zero, try to sync guest time on a domain resume. Be aware, that +# this requires guest agent, which, moreover, has to run under supported +# system. For instance, qemu-ga doesn't support guest time synchronization +# on Windows guests, but Linux ones. By default, this piece of +# functionality is turned off. +#SYNC_TIME=1 -- 2.0.5

2 1

[libvirt] [PATCH 0/3] Introduce machine type based capabilities filtering
by Michal Privoznik 20 Feb '15

20 Feb '15

Well, here's an alternative approach to my question here [1]. 1: https://www.redhat.com/archives/libvir-list/2015-February/msg00369.html Michal Privoznik (3): virQEMUCapsCacheLookupCopy: Pass machine type virQEMUCapsCacheLookupCopy: Filter qemuCaps based on machineType qemuCaps: Disable memdev for rhel6.5.0 machine type src/qemu/qemu_capabilities.c | 45 +++++++++++++++++++++- src/qemu/qemu_capabilities.h | 5 ++- src/qemu/qemu_migration.c | 3 +- src/qemu/qemu_process.c | 9 +++-- .../qemuxml2argv-numatune-memnode-rhel650.args | 7 ++++ .../qemuxml2argv-numatune-memnode-rhel650.xml | 31 +++++++++++++++ tests/qemuxml2argvtest.c | 6 +++ 7 files changed, 100 insertions(+), 6 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-numatune-memnode-rhel650.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-numatune-memnode-rhel650.xml -- 2.0.5

3 10

[libvirt] [PATCH] virsh: fix vcpupin info
by Pavel Hrdina 20 Feb '15

20 Feb '15

The "virDomainGetInfo" will get for running domain only live info and for offline domain only config info. There was no way how to get config info for running domain. We will use "vshCPUCountCollect" instead to get the correct cpu count that we need to pass to "virDomainGetVcpuPinInfo". Also cleanup some unnecessary variables and checks that are done by drivers. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1160559 Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com> --- tests/vcpupin | 2 +- tools/virsh-domain.c | 75 ++++++++++++++++++++++++---------------------------- 2 files changed, 35 insertions(+), 42 deletions(-) diff --git a/tests/vcpupin b/tests/vcpupin index 9f34ec0..cd09145 100755 --- a/tests/vcpupin +++ b/tests/vcpupin @@ -43,7 +43,7 @@ compare exp out || fail=1 $abs_top_builddir/tools/virsh --connect test:///default vcpupin test 100 0,1 > out 2>&1 test $? = 1 || fail=1 cat <<\EOF > exp || fail=1 -error: vcpupin: vCPU index out of range. +error: invalid argument: requested vcpu is higher than allocated vcpus EOF compare exp out || fail=1 diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index 2506b89..61a3385 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -6450,20 +6450,17 @@ vshParseCPUList(vshControl *ctl, const char *cpulist, static bool cmdVcpuPin(vshControl *ctl, const vshCmd *cmd) { - virDomainInfo info; virDomainPtr dom; unsigned int vcpu = 0; const char *cpulist = NULL; bool ret = false; unsigned char *cpumap = NULL; - unsigned char *cpumaps = NULL; size_t cpumaplen; int maxcpu, ncpus; size_t i; bool config = vshCommandOptBool(cmd, "config"); bool live = vshCommandOptBool(cmd, "live"); bool current = vshCommandOptBool(cmd, "current"); - bool query = false; /* Query mode if no cpulist */ int got_vcpu; unsigned int flags = VIR_DOMAIN_AFFECT_CURRENT; @@ -6481,48 +6478,47 @@ cmdVcpuPin(vshControl *ctl, const vshCmd *cmd) if (vshCommandOptStringReq(ctl, cmd, "cpulist", &cpulist) < 0) return false; - if (!(dom = vshCommandOptDomain(ctl, cmd, NULL))) - return false; - - query = !cpulist; + if (!cpulist) + VSH_EXCLUSIVE_OPTIONS_VAR(live, config); if ((got_vcpu = vshCommandOptUInt(cmd, "vcpu", &vcpu)) < 0) { vshError(ctl, "%s", _("vcpupin: Invalid vCPU number.")); - goto cleanup; + return false; } /* In pin mode, "vcpu" is necessary */ - if (!query && got_vcpu == 0) { + if (cpulist && got_vcpu == 0) { vshError(ctl, "%s", _("vcpupin: Missing vCPU number in pin mode.")); - goto cleanup; - } - - if (virDomainGetInfo(dom, &info) != 0) { - vshError(ctl, "%s", _("vcpupin: failed to get domain information.")); - goto cleanup; - } - - if (vcpu >= info.nrVirtCpu) { - vshError(ctl, "%s", _("vcpupin: vCPU index out of range.")); - goto cleanup; + return false; } if ((maxcpu = vshNodeGetCPUCount(ctl->conn)) < 0) - goto cleanup; - + return false; cpumaplen = VIR_CPU_MAPLEN(maxcpu); + if (!(dom = vshCommandOptDomain(ctl, cmd, NULL))) + return false; + /* Query mode: show CPU affinity information then exit.*/ - if (query) { + if (!cpulist) { /* When query mode and neither "live", "config" nor "current" * is specified, set VIR_DOMAIN_AFFECT_CURRENT as flags */ if (flags == -1) flags = VIR_DOMAIN_AFFECT_CURRENT; - cpumaps = vshMalloc(ctl, info.nrVirtCpu * cpumaplen); - if ((ncpus = virDomainGetVcpuPinInfo(dom, info.nrVirtCpu, - cpumaps, cpumaplen, flags)) >= 0) { + if ((ncpus = vshCPUCountCollect(ctl, dom, flags, true)) < 0) { + if (ncpus == -1) { + if (flags & VIR_DOMAIN_AFFECT_LIVE) + vshError(ctl, "%s", _("cannot get vcpupin for offline domain")); + else + vshError(ctl, "%s", _("cannot get vcpupin for transient domain")); + } + goto cleanup; + } + cpumap = vshMalloc(ctl, ncpus * cpumaplen); + if ((ncpus = virDomainGetVcpuPinInfo(dom, ncpus, cpumap, + cpumaplen, flags)) >= 0) { vshPrintExtra(ctl, "%s %s\n", _("VCPU:"), _("CPU Affinity")); vshPrintExtra(ctl, "----------------------------------\n"); for (i = 0; i < ncpus; i++) { @@ -6530,30 +6526,27 @@ cmdVcpuPin(vshControl *ctl, const vshCmd *cmd) continue; vshPrint(ctl, "%4zu: ", i); - ret = vshPrintPinInfo(cpumaps, cpumaplen, maxcpu, i); + ret = vshPrintPinInfo(cpumap, cpumaplen, maxcpu, i); vshPrint(ctl, "\n"); if (!ret) break; } } - - VIR_FREE(cpumaps); - goto cleanup; - } - - /* Pin mode: pinning specified vcpu to specified physical cpus*/ - if (!(cpumap = vshParseCPUList(ctl, cpulist, maxcpu, cpumaplen))) - goto cleanup; - - if (flags == -1) { - if (virDomainPinVcpu(dom, vcpu, cpumap, cpumaplen) != 0) - goto cleanup; } else { - if (virDomainPinVcpuFlags(dom, vcpu, cpumap, cpumaplen, flags) != 0) + /* Pin mode: pinning specified vcpu to specified physical cpus*/ + if (!(cpumap = vshParseCPUList(ctl, cpulist, maxcpu, cpumaplen))) goto cleanup; + + if (flags == -1) { + if (virDomainPinVcpu(dom, vcpu, cpumap, cpumaplen) != 0) + goto cleanup; + } else { + if (virDomainPinVcpuFlags(dom, vcpu, cpumap, cpumaplen, flags) != 0) + goto cleanup; + } + ret = true; } - ret = true; cleanup: VIR_FREE(cpumap); virDomainFree(dom); -- 2.0.5

2 1

[libvirt] [PATCH v3] virsh-edit: Make force editing usable
by Martin Kletzander 20 Feb '15

20 Feb '15

When editing a domain with 'virsh edit' and failing validation, the usual message pops up: Failed. Try again? [y,n,f,?]: Turning off validation can be useful, mainly for testing (but other purposes too), so this patch adds support for relaxing definition in virsh-edit and makes 'virsh edit <domain>' more usable. Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- v3: - turn 'v' into an 'i' (as in "Ignore validation") - fix translation v2: - make a special case 'v' for turning off validation tools/virsh-domain.c | 6 ++++++ tools/virsh-edit.c | 21 +++++++++++++++++++-- tools/virsh.c | 28 ++++++++++++++++++++-------- tools/virsh.h | 4 ++-- 4 files changed, 47 insertions(+), 12 deletions(-) diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index 2506b89..1b8d473 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -11265,7 +11265,13 @@ cmdEdit(vshControl *ctl, const vshCmd *cmd) } while (0) #define EDIT_DEFINE \ (dom_edited = vshDomainDefine(ctl->conn, doc_edited, define_flags)) +#define EDIT_RELAX \ + do { \ + define_flags &= ~VIR_DOMAIN_DEFINE_VALIDATE; \ + } while (0); + #include "virsh-edit.c" +#undef EDIT_RELAX vshPrint(ctl, _("Domain %s XML configuration edited.\n"), virDomainGetName(dom_edited)); diff --git a/tools/virsh-edit.c b/tools/virsh-edit.c index 41a6d53..1b39cb7 100644 --- a/tools/virsh-edit.c +++ b/tools/virsh-edit.c @@ -1,7 +1,7 @@ /* * virsh-edit.c: Implementation of generic virsh *-edit intelligence * - * Copyright (C) 2012 Red Hat, Inc. + * Copyright (C) 2012, 2015 Red Hat, Inc. * * This library is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public @@ -62,6 +62,7 @@ do { char *doc_reread = NULL; const char *msg = NULL; bool edit_success = false; + bool relax_avail = false; /* Get the XML configuration of the object. */ doc = (EDIT_GET_XML); @@ -74,6 +75,11 @@ do { goto edit_cleanup; reedit: + +#ifdef EDIT_RELAX + relax_avail = true; +#endif + /* Start the editor. */ if (vshEditFile(ctl, tmp) == -1) goto edit_cleanup; @@ -112,7 +118,7 @@ do { msg = _("Failed."); if (msg) { - int c = vshAskReedit(ctl, msg); + int c = vshAskReedit(ctl, msg, relax_avail); switch (c) { case 'y': goto reedit; @@ -126,6 +132,17 @@ do { goto edit_cleanup; break; +#ifdef EDIT_RELAX + case 'i': + if (relax_avail) { + EDIT_RELAX; + relax_avail = false; + goto redefine; + break; + } + /* fall-through */ +#endif + default: vshError(ctl, "%s", msg); break; diff --git a/tools/virsh.c b/tools/virsh.c index aba34ce..9ecddf3 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -1,7 +1,7 @@ /* * virsh.c: a shell to exercise the libvirt API * - * Copyright (C) 2005, 2007-2014 Red Hat, Inc. + * Copyright (C) 2005, 2007-2015 Red Hat, Inc. * * This library is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public @@ -512,13 +512,14 @@ vshPrintRaw(vshControl *ctl, ...) * edited file. * * Returns 'y' if he wants to - * 'f' if he forcibly wants to * 'n' if he doesn't want to + * 'i' if he wants to try defining it again while ignoring validation + * 'f' if he forcibly wants to * -1 on error * 0 otherwise */ int -vshAskReedit(vshControl *ctl, const char *msg) +vshAskReedit(vshControl *ctl, const char *msg, bool relax_avail) { int c = -1; @@ -531,9 +532,8 @@ vshAskReedit(vshControl *ctl, const char *msg) return -1; while (true) { - /* TRANSLATORS: For now, we aren't using LC_MESSAGES, and the user - * choices really are limited to just 'y', 'n', 'f' and '?' */ - vshPrint(ctl, "\r%s %s", msg, _("Try again? [y,n,f,?]:")); + vshPrint(ctl, "\r%s %s %s: ", msg, _("Try again?"), + relax_avail ? "[y,n,i,f,?]" : "[y,n,f,?]"); c = c_tolower(getchar()); if (c == '?') { @@ -541,11 +541,21 @@ vshAskReedit(vshControl *ctl, const char *msg) "", _("y - yes, start editor again"), _("n - no, throw away my changes"), + NULL); + + if (relax_avail) { + vshPrintRaw(ctl, + _("i - turn off validation and try to redefine again"), + NULL); + } + + vshPrintRaw(ctl, _("f - force, try to redefine again"), _("? - print this help"), NULL); continue; - } else if (c == 'y' || c == 'n' || c == 'f') { + } else if (c == 'y' || c == 'n' || c == 'f' || + (relax_avail && c == 'i')) { break; } } @@ -557,7 +567,9 @@ vshAskReedit(vshControl *ctl, const char *msg) } #else /* WIN32 */ int -vshAskReedit(vshControl *ctl, const char *msg ATTRIBUTE_UNUSED) +vshAskReedit(vshControl *ctl, + const char *msg ATTRIBUTE_UNUSED, + bool relax_avail ATTRIBUTE_UNUSED) { vshDebug(ctl, VSH_ERR_WARNING, "%s", _("This function is not " "supported on WIN32 platform")); diff --git a/tools/virsh.h b/tools/virsh.h index 7d5d8a2..df2ea7f 100644 --- a/tools/virsh.h +++ b/tools/virsh.h @@ -1,7 +1,7 @@ /* * virsh.h: a shell to exercise the libvirt API * - * Copyright (C) 2005, 2007-2014 Red Hat, Inc. + * Copyright (C) 2005, 2007-2015 Red Hat, Inc. * * This library is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public @@ -359,7 +359,7 @@ char *vshGetTypedParamValue(vshControl *ctl, virTypedParameterPtr item) char *vshEditWriteToTempFile(vshControl *ctl, const char *doc); int vshEditFile(vshControl *ctl, const char *filename); char *vshEditReadBackFile(vshControl *ctl, const char *filename); -int vshAskReedit(vshControl *ctl, const char *msg); +int vshAskReedit(vshControl *ctl, const char *msg, bool relax_avail); int vshStreamSink(virStreamPtr st, const char *bytes, size_t nbytes, void *opaque); double vshPrettyCapacity(unsigned long long val, const char **unit); -- 2.3.0

2 1

[libvirt] [PATCH v2] virsh-edit: Make force editing usable
by Martin Kletzander 20 Feb '15

20 Feb '15

When editing a domain with 'virsh edit' and failing validation, the usual message pops up: Failed. Try again? [y,n,f,?]: Turning of validation can be ussable, mainly for testing (but other purposes too), so this patch adds support for relaxing definition in virsh-edit and makes 'virsh edit <domain>' more usable. Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- v2: - make a special case 'v' for turning off validation tools/virsh-domain.c | 6 ++++++ tools/virsh-edit.c | 21 +++++++++++++++++++-- tools/virsh.c | 28 +++++++++++++++++++++------- tools/virsh.h | 4 ++-- 4 files changed, 48 insertions(+), 11 deletions(-) diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index 2506b89..b4761c6 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -11265,7 +11265,13 @@ cmdEdit(vshControl *ctl, const vshCmd *cmd) } while (0) #define EDIT_DEFINE \ (dom_edited = vshDomainDefine(ctl->conn, doc_edited, define_flags)) +#define EDIT_RELAX \ + do { \ + define_flags &= ~VIR_DOMAIN_DEFINE_VALIDATE; \ + } while (0); + #include "virsh-edit.c" +#undefine EDIT_RELAX vshPrint(ctl, _("Domain %s XML configuration edited.\n"), virDomainGetName(dom_edited)); diff --git a/tools/virsh-edit.c b/tools/virsh-edit.c index 41a6d53..2a13a04 100644 --- a/tools/virsh-edit.c +++ b/tools/virsh-edit.c @@ -1,7 +1,7 @@ /* * virsh-edit.c: Implementation of generic virsh *-edit intelligence * - * Copyright (C) 2012 Red Hat, Inc. + * Copyright (C) 2012, 2015 Red Hat, Inc. * * This library is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public @@ -62,6 +62,7 @@ do { char *doc_reread = NULL; const char *msg = NULL; bool edit_success = false; + bool relax_avail = false; /* Get the XML configuration of the object. */ doc = (EDIT_GET_XML); @@ -74,6 +75,11 @@ do { goto edit_cleanup; reedit: + +#ifdef EDIT_RELAX + relax_avail = true; +#endif + /* Start the editor. */ if (vshEditFile(ctl, tmp) == -1) goto edit_cleanup; @@ -112,7 +118,7 @@ do { msg = _("Failed."); if (msg) { - int c = vshAskReedit(ctl, msg); + int c = vshAskReedit(ctl, msg, relax_avail); switch (c) { case 'y': goto reedit; @@ -126,6 +132,17 @@ do { goto edit_cleanup; break; +#ifdef EDIT_RELAX + case 'v': + if (relax_avail) { + EDIT_RELAX; + relax_avail = false; + goto redefine; + break; + } + /* fall-through */ +#endif + default: vshError(ctl, "%s", msg); break; diff --git a/tools/virsh.c b/tools/virsh.c index aba34ce..464af3d 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -1,7 +1,7 @@ /* * virsh.c: a shell to exercise the libvirt API * - * Copyright (C) 2005, 2007-2014 Red Hat, Inc. + * Copyright (C) 2005, 2007-2015 Red Hat, Inc. * * This library is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public @@ -512,13 +512,14 @@ vshPrintRaw(vshControl *ctl, ...) * edited file. * * Returns 'y' if he wants to - * 'f' if he forcibly wants to * 'n' if he doesn't want to + * 'v' if he wants to try defining it again with validation turned off + * 'f' if he forcibly wants to * -1 on error * 0 otherwise */ int -vshAskReedit(vshControl *ctl, const char *msg) +vshAskReedit(vshControl *ctl, const char *msg, bool relax_avail) { int c = -1; @@ -532,8 +533,9 @@ vshAskReedit(vshControl *ctl, const char *msg) while (true) { /* TRANSLATORS: For now, we aren't using LC_MESSAGES, and the user - * choices really are limited to just 'y', 'n', 'f' and '?' */ - vshPrint(ctl, "\r%s %s", msg, _("Try again? [y,n,f,?]:")); + * choices really are limited to just 'y', 'n', 'v', 'f' and '?' */ + vshPrint(ctl, _("\r%s Try again? %s: "), msg, + relax_avail ? "[y,n,v,f,?]" : "[y,n,f,?]"); c = c_tolower(getchar()); if (c == '?') { @@ -541,11 +543,21 @@ vshAskReedit(vshControl *ctl, const char *msg) "", _("y - yes, start editor again"), _("n - no, throw away my changes"), + NULL); + + if (relax_avail) { + vshPrintRaw(ctl, + _("v - turn off validation and try to redefine again"), + NULL); + } + + vshPrintRaw(ctl, _("f - force, try to redefine again"), _("? - print this help"), NULL); continue; - } else if (c == 'y' || c == 'n' || c == 'f') { + } else if (c == 'y' || c == 'n' || c == 'f' || + (relax_avail && c == 'v')) { break; } } @@ -557,7 +569,9 @@ vshAskReedit(vshControl *ctl, const char *msg) } #else /* WIN32 */ int -vshAskReedit(vshControl *ctl, const char *msg ATTRIBUTE_UNUSED) +vshAskReedit(vshControl *ctl, + const char *msg ATTRIBUTE_UNUSED, + bool relax_avail ATTRIBUTE_UNUSED) { vshDebug(ctl, VSH_ERR_WARNING, "%s", _("This function is not " "supported on WIN32 platform")); diff --git a/tools/virsh.h b/tools/virsh.h index 7d5d8a2..df2ea7f 100644 --- a/tools/virsh.h +++ b/tools/virsh.h @@ -1,7 +1,7 @@ /* * virsh.h: a shell to exercise the libvirt API * - * Copyright (C) 2005, 2007-2014 Red Hat, Inc. + * Copyright (C) 2005, 2007-2015 Red Hat, Inc. * * This library is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public @@ -359,7 +359,7 @@ char *vshGetTypedParamValue(vshControl *ctl, virTypedParameterPtr item) char *vshEditWriteToTempFile(vshControl *ctl, const char *doc); int vshEditFile(vshControl *ctl, const char *filename); char *vshEditReadBackFile(vshControl *ctl, const char *filename); -int vshAskReedit(vshControl *ctl, const char *msg); +int vshAskReedit(vshControl *ctl, const char *msg, bool relax_avail); int vshStreamSink(virStreamPtr st, const char *bytes, size_t nbytes, void *opaque); double vshPrettyCapacity(unsigned long long val, const char **unit); -- 2.3.0

4 5

[libvirt] [PATCH] network: allow <pf> together with <interface>/<address> in network status
by Laine Stump 20 Feb '15

20 Feb '15

The function that parses the <forward> subelement of a network used to fail/log an error if the network definition contained both a <pf> element as well as at least one <interface> or <address> element. That check was present because the configuration of a network should have either one <pf>, one or more <interface>, or one or more <address>, but never combinations of multiple kinds. This caused a problem when libvirtd was restarted with a network already active - when a network with a <pf> element is started, the referenced PF (Physical Function of an SRIOV-capable network card) is checked for VFs (Virtual Functions), and the <forward> is filled in with a list of all VFs for that PF either in the form of their PCI addresses (a list of <address>) or their netdev names (a list of <interface>); the <pf> element is not removed though. When libvirtd is restarted, it parses the network status and finds both the original <pf> from the config, as well as the list of either <address> or <interface>, fails the parse, and the network is not added to the active list. This failure is often obscured because the network is marked as autostart so libvirt immediately restarts it. It seems odd to me that <interface> and <address> are stored in the same array rather than keeping two separate arrays, and having separate arrays would have made the check much simpler. However, changing to use two separate arrays would have required changes in more places, potentially creating more conflicts and (more importantly) more possible regressions in the event of a backport, so I chose to keep the existing data structure in order to localize the change. It appears that this problem has been in the code ever since support for <pf> was added (0.9.10), but until commit 34cc3b2f106e296df5e64309620c79d16fd76c85 (first in libvirt 1.2.4) networks with interface pools were not properly marked as active on restart anyway, so there is no point in backporting this patch any further than that. --- src/conf/network_conf.c | 10 +--------- src/network/bridge_driver.c | 35 +++++++++++++++++++++++++++++++++++ 2 files changed, 36 insertions(+), 9 deletions(-) diff --git a/src/conf/network_conf.c b/src/conf/network_conf.c index f947d89..dce3360 100644 --- a/src/conf/network_conf.c +++ b/src/conf/network_conf.c @@ -1,7 +1,7 @@ /* * network_conf.c: network XML handling * - * Copyright (C) 2006-2014 Red Hat, Inc. + * Copyright (C) 2006-2015 Red Hat, Inc. * Copyright (C) 2006-2008 Daniel P. Berrange * * This library is free software; you can redistribute it and/or @@ -1638,14 +1638,6 @@ virNetworkForwardDefParseXML(const char *networkName, goto cleanup; } - if (((nForwardIfs > 0) + (nForwardAddrs > 0) + (nForwardPfs > 0)) > 1) { - virReportError(VIR_ERR_XML_ERROR, - _("<address>, <interface>, and <pf> elements in <forward> " - "of network %s are mutually exclusive"), - networkName); - goto cleanup; - } - forwardDev = virXPathString("string(./@dev)", ctxt); if (forwardDev && (nForwardAddrs > 0 || nForwardPfs > 0)) { virReportError(VIR_ERR_XML_ERROR, "%s", diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c index 2798010..af16c25 100644 --- a/src/network/bridge_driver.c +++ b/src/network/bridge_driver.c @@ -2734,6 +2734,7 @@ networkValidate(virNetworkDefPtr def, virNetworkIpDefPtr ipdef; bool ipv4def = false, ipv6def = false; bool bandwidthAllowed = true; + bool usesInterface = false, usesAddress = false; /* check for duplicate networks */ if (virNetworkObjIsDuplicate(&driver->networks, def, check_active) < 0) @@ -2798,6 +2799,40 @@ networkValidate(virNetworkDefPtr def, bandwidthAllowed = false; } + /* we support configs with a single PF defined: + * <pf dev='eth0'/> + * or with a list of netdev names: + * <interface dev='eth9'/> + * OR a list of PCI addresses + * <address type='pci' domain='0' bus='4' slot='0' function='1'/> + * but not any combination of those. + * + * Since <interface> and <address> are for some strange reason + * stored in the same array, we need to cycle through it and check + * the type of each. + */ + for (i = 0; i < def->forward.nifs; i++) { + switch ((virNetworkForwardHostdevDeviceType) + def->forward.ifs[i].type) { + case VIR_NETWORK_FORWARD_HOSTDEV_DEVICE_NETDEV: + usesInterface = true; + break; + case VIR_NETWORK_FORWARD_HOSTDEV_DEVICE_PCI: + usesAddress = true; + break; + case VIR_NETWORK_FORWARD_HOSTDEV_DEVICE_NONE: + case VIR_NETWORK_FORWARD_HOSTDEV_DEVICE_LAST: + break; + } + } + if ((def->forward.npfs > 0) + usesInterface + usesAddress > 1) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("<address>, <interface>, and <pf> elements of " + "<forward> in network %s are mutually exclusive"), + def->name); + return -1; + } + /* We only support dhcp on one IPv4 address and * on one IPv6 address per defined network */ -- 2.1.0

2 1

[libvirt] dmidecode dependency
by Jason Helfman 19 Feb '15

19 Feb '15

Hello All, I was wondering if it may be possible to call sysctl, instead of requiring dmidecode to be installed for getting the uuid of a system? According to the user that filed the bug below, dmidecode needs to be installed to get the uuid, or errors will be printed in logs to get the uuid. https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=196733 I was wondering if sysctl can be called to get this information for FreeBSD? I am not sure if Linux would give similar information, but would think it might. Beyond this, I am not sure if dmidecode is used for other aspects beyond uuid where it is required where sysctl doesn't give the information needed. $ sudo dmidecode |grep -i uuid UUID: 42355DE8-C55C-E487-8E16-509A78EB5EE1 $ sysctl kern.hostuuid kern.hostuuid: 42355de8-c55c-e487-8e16-509a78eb5ee1 Thanks, Jason -- Jason Helfman | FreeBSD Committer jgh(a)FreeBSD.org | http://people.freebsd.org/~jgh | The Power to Serve

2 2

[libvirt] [RFC PATCH] Use PAUSED state for domains that are starting up
by Jiri Denemark 19 Feb '15

19 Feb '15

When libvirt is starting a domain, it reports the state as SHUTOFF until it's RUNNING. This is not ideal because domain startup may take a long time (usually because of some configuration issues, firewalls blocking access to network disks, etc.) and domain lists provided by libvirt look awkward. One can see weird shutoff domains with IDs in a list of active domains or even shutoff transient domains. In any case, it looks more like a bug in libvirt than a normal state a domain goes through. I'm not quite sure what the best way to fix this is. In this patch, I tried to use PAUSED state with STARTING_UP reason. Alternatively, we could keep using SHUTOFF state and just set STARTING_UP reason instead of UNKNOWN but it just feels wrong and wouldn't really solve the confusion when looking at virsh list. I made the change to qemu driver only in this RFC patch, I will update all drivers once we agree on the best approach. Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- include/libvirt/libvirt-domain.h | 1 + src/conf/domain_conf.c | 3 ++- src/qemu/qemu_process.c | 22 ++++++++++++++-------- tools/virsh-domain-monitor.c | 3 ++- 4 files changed, 19 insertions(+), 10 deletions(-) diff --git a/include/libvirt/libvirt-domain.h b/include/libvirt/libvirt-domain.h index 4dbd7f5..90150f6 100644 --- a/include/libvirt/libvirt-domain.h +++ b/include/libvirt/libvirt-domain.h @@ -116,6 +116,7 @@ typedef enum { VIR_DOMAIN_PAUSED_SHUTTING_DOWN = 8, /* paused during shutdown process */ VIR_DOMAIN_PAUSED_SNAPSHOT = 9, /* paused while creating a snapshot */ VIR_DOMAIN_PAUSED_CRASHED = 10, /* paused due to a guest crash */ + VIR_DOMAIN_PAUSED_STARTING_UP = 11, /* the domain is being started */ # ifdef VIR_ENUM_SENTINELS VIR_DOMAIN_PAUSED_LAST diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index b3d63f8..2b7c5bf 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -661,7 +661,8 @@ VIR_ENUM_IMPL(virDomainPausedReason, VIR_DOMAIN_PAUSED_LAST, "from snapshot", "shutdown", "snapshot", - "panicked") + "panicked", + "starting up") VIR_ENUM_IMPL(virDomainShutdownReason, VIR_DOMAIN_SHUTDOWN_LAST, "unknown", diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 1d4e957..d317b19 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -3412,6 +3412,7 @@ qemuProcessUpdateState(virQEMUDriverPtr driver, virDomainObjPtr vm) virDomainState state; virDomainPausedReason reason; virDomainState newState = VIR_DOMAIN_NOSTATE; + int oldReason; int newReason; bool running; char *msg = NULL; @@ -3425,9 +3426,16 @@ qemuProcessUpdateState(virQEMUDriverPtr driver, virDomainObjPtr vm) if (ret < 0) return -1; - state = virDomainObjGetState(vm, NULL); + state = virDomainObjGetState(vm, &oldReason); - if (state == VIR_DOMAIN_PAUSED && running) { + if (running && + (state == VIR_DOMAIN_SHUTOFF || + (state == VIR_DOMAIN_PAUSED && + oldReason == VIR_DOMAIN_PAUSED_STARTING_UP))) { + newState = VIR_DOMAIN_RUNNING; + newReason = VIR_DOMAIN_RUNNING_BOOTED; + ignore_value(VIR_STRDUP_QUIET(msg, "finished booting")); + } else if (state == VIR_DOMAIN_PAUSED && running) { newState = VIR_DOMAIN_RUNNING; newReason = VIR_DOMAIN_RUNNING_UNPAUSED; ignore_value(VIR_STRDUP_QUIET(msg, "was unpaused")); @@ -3446,10 +3454,6 @@ qemuProcessUpdateState(virQEMUDriverPtr driver, virDomainObjPtr vm) ignore_value(virAsprintf(&msg, "was paused (%s)", virDomainPausedReasonTypeToString(reason))); } - } else if (state == VIR_DOMAIN_SHUTOFF && running) { - newState = VIR_DOMAIN_RUNNING; - newReason = VIR_DOMAIN_RUNNING_BOOTED; - ignore_value(VIR_STRDUP_QUIET(msg, "finished booting")); } if (newState != VIR_DOMAIN_NOSTATE) { @@ -3817,7 +3821,9 @@ qemuProcessReconnect(void *opaque) goto error; state = virDomainObjGetState(obj, &reason); - if (state == VIR_DOMAIN_SHUTOFF) { + if (state == VIR_DOMAIN_SHUTOFF || + (state == VIR_DOMAIN_PAUSED && + reason == VIR_DOMAIN_PAUSED_STARTING_UP)) { VIR_DEBUG("Domain '%s' wasn't fully started yet, killing it", obj->def->name); goto error; @@ -4435,7 +4441,7 @@ int qemuProcessStart(virConnectPtr conn, vm->def->id = qemuDriverAllocateID(driver); qemuDomainSetFakeReboot(driver, vm, false); - virDomainObjSetState(vm, VIR_DOMAIN_SHUTOFF, VIR_DOMAIN_SHUTOFF_UNKNOWN); + virDomainObjSetState(vm, VIR_DOMAIN_PAUSED, VIR_DOMAIN_PAUSED_STARTING_UP); if (virAtomicIntInc(&driver->nactive) == 1 && driver->inhibitCallback) driver->inhibitCallback(true, driver->inhibitOpaque); diff --git a/tools/virsh-domain-monitor.c b/tools/virsh-domain-monitor.c index 925eb1b..da23ace 100644 --- a/tools/virsh-domain-monitor.c +++ b/tools/virsh-domain-monitor.c @@ -184,7 +184,8 @@ VIR_ENUM_IMPL(vshDomainPausedReason, N_("from snapshot"), N_("shutting down"), N_("creating snapshot"), - N_("crashed")) + N_("crashed"), + N_("starting up")) VIR_ENUM_DECL(vshDomainShutdownReason) VIR_ENUM_IMPL(vshDomainShutdownReason, -- 2.3.0

2 7

[libvirt] [PATCH] virsh-edit: Make force editing usable
by Martin Kletzander 19 Feb '15

19 Feb '15

When editing a domain with 'virsh edit' and failing validation, the usual message pops up: Failed. Try again? [y,n,f,?]: The option 'f' (force) is pretty unusable. It tries to step into the same puddle again and again when it can at least try to force the definition. This patch adds support for relaxing definition in virsh-edit and makes 'virsh edit <domain>' more usable, mainly for testing. Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- tools/virsh-domain.c | 5 +++++ tools/virsh-edit.c | 5 ++++- tools/virsh.c | 4 ++-- 3 files changed, 11 insertions(+), 3 deletions(-) diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index 2506b89..3dd676c 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -11265,6 +11265,11 @@ cmdEdit(vshControl *ctl, const vshCmd *cmd) } while (0) #define EDIT_DEFINE \ (dom_edited = vshDomainDefine(ctl->conn, doc_edited, define_flags)) +#define EDIT_RELAX \ + do { \ + define_flags &= ~VIR_DOMAIN_DEFINE_VALIDATE; \ + } while (0); + #include "virsh-edit.c" vshPrint(ctl, _("Domain %s XML configuration edited.\n"), diff --git a/tools/virsh-edit.c b/tools/virsh-edit.c index 41a6d53..3186998 100644 --- a/tools/virsh-edit.c +++ b/tools/virsh-edit.c @@ -1,7 +1,7 @@ /* * virsh-edit.c: Implementation of generic virsh *-edit intelligence * - * Copyright (C) 2012 Red Hat, Inc. + * Copyright (C) 2012, 2015 Red Hat, Inc. * * This library is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public @@ -119,6 +119,9 @@ do { break; case 'f': +#ifdef EDIT_RELAX + EDIT_RELAX; +#endif goto redefine; break; diff --git a/tools/virsh.c b/tools/virsh.c index aba34ce..5b154e1 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -1,7 +1,7 @@ /* * virsh.c: a shell to exercise the libvirt API * - * Copyright (C) 2005, 2007-2014 Red Hat, Inc. + * Copyright (C) 2005, 2007-2015 Red Hat, Inc. * * This library is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public @@ -541,7 +541,7 @@ vshAskReedit(vshControl *ctl, const char *msg) "", _("y - yes, start editor again"), _("n - no, throw away my changes"), - _("f - force, try to redefine again"), + _("f - force, try to redefine again (without validation)"), _("? - print this help"), NULL); continue; -- 2.3.0

3 5