- Devel - libvirt List Archives

[libvirt] [PATCH] udevEnumerateDevices: remove dead code
by Jim Meyering 01 Mar '10

01 Mar '10

Clang reported the dead store. >From 0557e23ca2f5eb78ed9f260040895011785f670f Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 1 Mar 2010 16:14:35 +0100 Subject: [PATCH] udevEnumerateDevices: remove dead code * src/node_device/node_device_udev.c (udevEnumerateDevices): Remove unnecessary call to udev_list_entry_get_name. --- src/node_device/node_device_udev.c | 4 +--- 1 files changed, 1 insertions(+), 3 deletions(-) diff --git a/src/node_device/node_device_udev.c b/src/node_device/node_device_udev.c index 68509f4..eee44c4 100644 --- a/src/node_device/node_device_udev.c +++ b/src/node_device/node_device_udev.c @@ -1,11 +1,11 @@ /* * node_device_udev.c: node device enumeration - libudev implementation * - * Copyright (C) 2009 Red Hat + * Copyright (C) 2009-2010 Red Hat * * This library is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public * License as published by the Free Software Foundation; either * version 2.1 of the License, or (at your option) any later version. * * This library is distributed in the hope that it will be useful, @@ -1333,30 +1333,28 @@ static int udevProcessDeviceListEntry(struct udev *udev, } static int udevEnumerateDevices(struct udev *udev) { struct udev_enumerate *udev_enumerate = NULL; struct udev_list_entry *list_entry = NULL; - const char *name = NULL; int ret = 0; udev_enumerate = udev_enumerate_new(udev); ret = udev_enumerate_scan_devices(udev_enumerate); if (0 != ret) { VIR_ERROR("udev scan devices returned %d", ret); goto out; } udev_list_entry_foreach(list_entry, udev_enumerate_get_list_entry(udev_enumerate)) { udevProcessDeviceListEntry(udev, list_entry); - name = udev_list_entry_get_name(list_entry); } out: udev_enumerate_unref(udev_enumerate); return ret; } -- 1.7.0.1.414.g89213d

2 1

[libvirt] [PATCH] cmdPoolDiscoverSources: initialize earlier to avoid FP from clang
by Jim Meyering 01 Mar '10

01 Mar '10

This change hoists the initialization of srcSpec to the top, to keep clang from reporting a false-positive used-uninitialized problem. >From f487655124c887310be3bc82d7beae3cd47e7d21 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 1 Mar 2010 15:41:15 +0100 Subject: [PATCH] cmdPoolDiscoverSources: initialize earlier to avoid FP from clang * tools/virsh.c (cmdPoolDiscoverSources): Always initialize srcSpec. Otherwise, clang would report that srcSpec could be used uninitialized in the call to virConnectFindStoragePoolSources. --- tools/virsh.c | 7 +++---- 1 files changed, 3 insertions(+), 4 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 3b1011c..191e7ba 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -4431,17 +4431,16 @@ static const vshCmdOptDef opts_find_storage_pool_sources[] = { static int cmdPoolDiscoverSources(vshControl * ctl, const vshCmd * cmd ATTRIBUTE_UNUSED) { - char *type, *srcSpec, *srcSpecFile, *srcList; + char *type, *srcSpecFile, *srcList; + char *srcSpec = NULL; int found; type = vshCommandOptString(cmd, "type", &found); if (!found) return FALSE; srcSpecFile = vshCommandOptString(cmd, "srcSpec", &found); - if (!found) { + if (!found) srcSpecFile = NULL; - srcSpec = NULL; - } if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) return FALSE; -- 1.7.0.1.414.g89213d

2 1

[libvirt] [PATCH] qemu: avoid null dereference on failed migration
by Eric Blake 01 Mar '10

01 Mar '10

From: Eric Blake <eblake(a)redhat.com> * src/qemu/qemu_monitor_text.c (qemuMonitorTextGetMigrationStatus): Check for failed strchr, to silence a coverity warning. Signed-off-by: Eric Blake <eblake(a)redhat.com> --- src/qemu/qemu_monitor_text.c | 5 +++++ 1 files changed, 5 insertions(+), 0 deletions(-) diff --git a/src/qemu/qemu_monitor_text.c b/src/qemu/qemu_monitor_text.c index 62ffcc6..e7b4b1f 100644 --- a/src/qemu/qemu_monitor_text.c +++ b/src/qemu/qemu_monitor_text.c @@ -989,6 +989,11 @@ int qemuMonitorTextGetMigrationStatus(qemuMonitorPtr mon, if ((tmp = strstr(reply, MIGRATION_PREFIX)) != NULL) { tmp += strlen(MIGRATION_PREFIX); end = strchr(tmp, '\r'); + if (end == NULL) { + qemuReportError(VIR_ERR_INTERNAL_ERROR, + _("unexpected migration status in %s"), reply); + goto cleanup; + } *end = '\0'; if ((*status = qemuMonitorMigrationStatusTypeFromString(tmp)) < 0) { -- 1.6.6.1

3 2

[libvirt] [PATCH] Do not search xenstore for disk device IDs
by Jim Fehlig 01 Mar '10

01 Mar '10

Disk devices can be referenced by name in Xen, e.g. when modifying their configuration or remvoving them. As such, don't search xenstore for a device ID corresponding to the disk device. Instead, search the disks contained in the domain definition and use the disk's target name if found. This approach allows removing a disk when domain is inactive. We obviously can't search xenstore when the domain is inactive. --- src/xen/xend_internal.c | 38 +++++++++++++++++++++++--------------- 1 files changed, 23 insertions(+), 15 deletions(-) diff --git a/src/xen/xend_internal.c b/src/xen/xend_internal.c index 9d95291..d0d32e2 100644 --- a/src/xen/xend_internal.c +++ b/src/xen/xend_internal.c @@ -93,6 +93,7 @@ xenDaemonFormatSxprOnePCI(virConnectPtr conn, static int virDomainXMLDevID(virDomainPtr domain, + virDomainDefPtr domDef, virDomainDeviceDefPtr dev, char *class, char *ref, @@ -4212,7 +4213,7 @@ xenDaemonAttachDeviceFlags(virDomainPtr domain, const char *xml, sexpr = virBufferContentAndReset(&buf); - if (virDomainXMLDevID(domain, dev, class, ref, sizeof(ref))) { + if (virDomainXMLDevID(domain, def, dev, class, ref, sizeof(ref))) { /* device doesn't exist, define it */ ret = xend_op(domain->conn, domain->name, "op", "device_create", "config", sexpr, NULL); @@ -4306,7 +4307,7 @@ xenDaemonDetachDeviceFlags(virDomainPtr domain, const char *xml, def, xml, VIR_DOMAIN_XML_INACTIVE))) goto cleanup; - if (virDomainXMLDevID(domain, dev, class, ref, sizeof(ref))) + if (virDomainXMLDevID(domain, def, dev, class, ref, sizeof(ref))) goto cleanup; if (dev->type == VIR_DOMAIN_DEVICE_HOSTDEV) { @@ -6048,6 +6049,7 @@ error: */ static int virDomainXMLDevID(virDomainPtr domain, + virDomainDefPtr domDef, virDomainDeviceDefPtr dev, char *class, char *ref, @@ -6056,27 +6058,33 @@ virDomainXMLDevID(virDomainPtr domain, xenUnifiedPrivatePtr priv = domain->conn->privateData; char *xref; char *tmp; + unsigned int i; + virDomainDiskDefPtr disk; if (dev->type == VIR_DOMAIN_DEVICE_DISK) { + if (dev->data.disk->dst == NULL) + return -1; if (dev->data.disk->driverName && STREQ(dev->data.disk->driverName, "tap")) strcpy(class, "tap"); else strcpy(class, "vbd"); - if (dev->data.disk->dst == NULL) - return -1; - xenUnifiedLock(priv); - xref = xenStoreDomainGetDiskID(domain->conn, domain->id, - dev->data.disk->dst); - xenUnifiedUnlock(priv); - if (xref == NULL) - return -1; - - tmp = virStrcpy(ref, xref, ref_len); - VIR_FREE(xref); - if (tmp == NULL) - return -1; + /* For disks, the device name can be used directly. + * If disk device exists in domain definintion, + * copy it to ref and return success. + */ + for (i = 0; i < domDef->ndisks; i++) { + disk = domDef->disks[i]; + if (STREQ(dev->data.disk->dst, disk->dst)) { + tmp = virStrcpy(ref, disk->dst, ref_len); + if (tmp == NULL) + return -1; + else + return 0; + } + } + return -1; } else if (dev->type == VIR_DOMAIN_DEVICE_NET) { char mac[30]; virDomainNetDefPtr def = dev->data.net; -- 1.6.0.2

3 4

[libvirt] running libvirt on Xen Cloud Platform 0.7.1
by Gadi Naor 01 Mar '10

01 Mar '10

Hi, I am trying to run libvirt daemon in Dom0 of Xen Cloud Platform 0.7.1. For that purpose I downloaded the libvirt source files 0.7.6 and trying to build the libvirt rpm with XEN driver using the XCP DDK. I am running into to many problems getting it done - so I was wondering whether this is supported or if anyone is using it in such way. Thank You, Gadi _________________________________________________________________ Hotmail: Free, trusted and rich email service. https://signup.live.com/signup.aspx?id=60969

1 0

[libvirt] and for LXC?
by Mihamina Rakotomandimby 01 Mar '10

01 Mar '10

Manao ahoana, Hello, Bonjour, Is there some plan/schedule to make libvirt LXC friendly? Or is it already? Misaotra, Thanks, Merci. -- Architecte Informatique chez Blueline/Gulfsat: Administration Systeme, Recherche & Developpement +261 34 29 155 34 / +261 33 11 207 36

3 2

[libvirt] (no subject)
by Daniel P. Berrange 01 Mar '10

01 Mar '10

Repost of http://www.redhat.com/archives/libvir-list/2010-February/msg00177.html The virsh output is now scaled to show KB/MB as needed, instead of always bytes. The public API gained extra status VIR_DOMAIN_JOB_COMPLETED = 3, /* Job has finished, but isn't cleaned up */ VIR_DOMAIN_JOB_FAILED = 4, /* Job hit error, but isn't cleaned up */ VIR_DOMAIN_JOB_CANCELLED = 6, /* Job was aborted, but isn't cleaned up */ to allow to more easily detecting end of a job, once we allow for truely async usage

2 21

[libvirt] [PATCH] [macvtap] interface type direct web docs addition
by Stefan Berger 01 Mar '10

01 Mar '10

This adds a description about the 'direct' type of interface recently added for macvtap device type support on the host.

3 2

[libvirt] [PATCH] Free the macvtap mode string
by Stefan Berger 01 Mar '10

01 Mar '10

I forgot to free the macvtap mode string. This fixes it. Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com>

3 2

[libvirt] Hypervisor raw options
by Daniel Dehennin 01 Mar '10

01 Mar '10

Hello, I wonder if it's possible to add hypervisor pass-through options, I add some usage of kvm -startdate and need to use a script which call kvm with the desired option. Something like: <emulator>/usr/bin/kvm</emulator> <emulator-options> <option>-option1</option> <option>-option2</option> </emulator-options> The several <option/> will be passed to the <emulator/> command. Regards. -- Daniel Dehennin Récupérer ma clef GPG: gpg --keyserver pgp.mit.edu --recv-keys 0x6A2540D1

2 1

[libvirt] reload libvirtd.conf has a little problem.
by Osier Yang 01 Mar '10

01 Mar '10

hi When libvirtd is running, I changed "unix_sock_ro_perms=0777" into ""unix_sock_ro_perms=0755", and then reload the configuration with: kill -SIGHUP `pidof libvirtd` or "service libvirtd reload" the permission of /var/lib/libvirt/libvirt-sock-ro still is 0777, but not 0755. and after "service libvirtd restart", the permission is changed into 0755. think it's a little problem. libvirt version: libvirt-0.7.6-1.fc12.x86_64 Thanks and Regards osier

2 2

[libvirt] [PATCH] build: avoid warning about unused variables
by Jim Meyering 01 Mar '10

01 Mar '10

>From 8ddff3e6cdccc2b4289509c6941b43f2ddf8e643 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Thu, 25 Feb 2010 14:19:33 +0100 Subject: [PATCH] build: avoid warning about unused variables * tools/virsh.c (cmdCPUBaseline): Remove declarations of unused variables, p and cur. --- tools/virsh.c | 3 +-- 1 files changed, 1 insertions(+), 2 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 5fdbbe5..89eefcf 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -7048,12 +7048,11 @@ cmdCPUBaseline(vshControl *ctl, const vshCmd *cmd) int found; int ret = TRUE; char *buffer; - char *p; char *result = NULL; const char **list = NULL; unsigned int count = 0; xmlDocPtr doc = NULL; - xmlNodePtr node_list, cur; + xmlNodePtr node_list; xmlXPathContextPtr ctxt = NULL; xmlSaveCtxtPtr sctxt = NULL; xmlBufferPtr buf = NULL; -- 1.7.0.401.g84adb

2 1

[libvirt] [PATCH] libvirtd: do not ignore failure to set group ID in privileged mode
by Jim Meyering 26 Feb '10

26 Feb '10

Coverity spotted these unchecked setgid calls: >From 7eddaf457d54974859a8c22cced6f43e2d267746 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Fri, 26 Feb 2010 10:42:14 +0100 Subject: [PATCH] libvirtd: do not ignore failure to set group ID in privileged mode * daemon/libvirtd.c (qemudListenUnix): Diagnose and fail upon failure to set or restore group-ID. --- daemon/libvirtd.c | 12 ++++++++---- 1 files changed, 8 insertions(+), 4 deletions(-) diff --git a/daemon/libvirtd.c b/daemon/libvirtd.c index cee6f13..9bdbecb 100644 --- a/daemon/libvirtd.c +++ b/daemon/libvirtd.c @@ -560,8 +560,10 @@ static int qemudListenUnix(struct qemud_server *server, oldgrp = getgid(); oldmask = umask(readonly ? ~unix_sock_ro_mask : ~unix_sock_rw_mask); - if (server->privileged) - setgid(unix_sock_gid); + if (server->privileged && setgid(unix_sock_gid)) { + VIR_ERROR(_("Failed to set group ID to %d"), unix_sock_gid); + goto cleanup; + } if (bind(sock->fd, (struct sockaddr *)&addr, sizeof(addr)) < 0) { VIR_ERROR(_("Failed to bind socket to '%s': %s"), @@ -569,8 +571,10 @@ static int qemudListenUnix(struct qemud_server *server, goto cleanup; } umask(oldmask); - if (server->privileged) - setgid(oldgrp); + if (server->privileged && setgid(oldgrp)) { + VIR_ERROR(_("Failed to restore group ID to %d"), oldgrp); + goto cleanup; + } if (listen(sock->fd, 30) < 0) { VIR_ERROR(_("Failed to listen for connections on '%s': %s"), -- 1.7.0.442.g55ad1

2 2

[libvirt] [PATCH] build: avoid warning about return-with-value in void function
by Jim Meyering 26 Feb '10

26 Feb '10

I saw this: virsh.c: In function 'virshReportError': virsh.c:363: warning: 'return' with a value, in function returning void >From fc83f08ac204cec5fba474a23bd19386d1a415d7 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Fri, 26 Feb 2010 11:50:34 +0100 Subject: [PATCH] build: avoid warning about return-with-value in void function * tools/virsh.c: Just "return;", not "return NULL;". --- tools/virsh.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index f7fd7d4..3b1011c 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -360,7 +360,7 @@ virshReportError(vshControl *ctl) * no error was ever raised, so just ignore */ last_error = virSaveLastError(); if (!last_error || last_error->code == VIR_ERR_OK) - return NULL; + return; } if (last_error->code == VIR_ERR_OK) { -- 1.7.0.442.g55ad1

2 1

[libvirt] build: update gnulib submodule to latest
by Jim Meyering 26 Feb '10

26 Feb '10

Thanks, Eric. >From bd3ff6b34892fe75cfcc67f1aeab43e8a273c1b6 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Fri, 26 Feb 2010 17:14:01 +0100 Subject: [PATCH] build: update gnulib submodule to latest * .gnulib: Update to latest. Commit 89bdf84bcd9c6032e37 inadvertently rewound the .gnulib submodule by 51 commits. This corrects it. Spotted by Eric Blake. --- .gnulib | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/.gnulib b/.gnulib index 11fbc57..80cd995 160000 --- a/.gnulib +++ b/.gnulib @@ -1 +1 @@ -Subproject commit 11fbc57405a118e6ec9a3ebc19bbf5ececdae4d6 +Subproject commit 80cd995cdcbf4b9ded895a43621a11f11806ad8d -- 1.7.0.442.g55ad1

2 1

[libvirt] [PATCH] Only build virDomainObjFormat if not building proxy.
by Chris Lalancette 26 Feb '10

26 Feb '10

While building under RHEL-5, I got a compile warning because virDomainObjFormat was defined but not used. That came about because in RHEL-5 we build with "#define PROXY", and virDomainObjFormat is only used with !PROXY. Move the define. Signed-off-by: Chris Lalancette <clalance(a)redhat.com> --- src/conf/domain_conf.c | 5 ++--- 1 files changed, 2 insertions(+), 3 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index b6d2c61..1d7f458 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -5668,6 +5668,8 @@ char *virDomainDefFormat(virDomainDefPtr def, return NULL; } +#ifndef PROXY + static char *virDomainObjFormat(virCapsPtr caps, virDomainObjPtr obj, int flags) @@ -5703,9 +5705,6 @@ error: return NULL; } - -#ifndef PROXY - int virDomainSaveXML(const char *configDir, virDomainDefPtr def, const char *xml) -- 1.6.6.1

3 3

[libvirt] Odd implementation of libvirt qemudDomainSetMaxMemory
by Chris Lalancette 26 Feb '10

26 Feb '10

Hey Cole, In commit 09a33fd8a9cc85e5e1064168472bff7aecff15fb, you made it so that qemudDomainSetMaxMemory() was hotplug only. However, with the changes you pushed, this function no longer makes a lot of sense. It looks up the domain, checks to make sure it is active, and then if it is active, it fails. Assuming that qemu doesn't support changing the maximum amount of memory on the fly, wouldn't it be better to just remove this function entirely and let the higher libvirt layers report "function not supported"? -- Chris Lalancette

2 1

[libvirt] [PATCH] openvzGetVEID: don't leak (memory + file descriptor)
by Jim Meyering 25 Feb '10

25 Feb '10

Coverity spotted the leaks. Here's a proposed fix. I didn't bother with a separate diagnostic for the unlikely event that we read a negative number from the pipe. Seems far-fetched enough not to bother, but it's easy to add, if anyone cares. >From f3439c7eae46681eacf5b469a6f0e22cb8fec1b4 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Thu, 25 Feb 2010 19:24:50 +0100 Subject: [PATCH] openvzGetVEID: don't leak (memory + file descriptor) * src/openvz/openvz_conf.c (openvzGetVEID): Always call fclose. Diagnose parse failure also when vzlist output is empty. If somehow we read a -1, diagnose that (albeit as a parse failure). --- src/openvz/openvz_conf.c | 19 +++++++------------ 1 files changed, 7 insertions(+), 12 deletions(-) diff --git a/src/openvz/openvz_conf.c b/src/openvz/openvz_conf.c index ce0c4d3..3713a45 100644 --- a/src/openvz/openvz_conf.c +++ b/src/openvz/openvz_conf.c @@ -964,6 +964,7 @@ int openvzGetVEID(const char *name) { char *cmd; int veid; FILE *fp; + bool ok; if (virAsprintf(&cmd, "%s %s -ovpsid -H", VZLIST, name) < 0) { openvzError(NULL, VIR_ERR_INTERNAL_ERROR, "%s", @@ -979,18 +980,12 @@ int openvzGetVEID(const char *name) { return -1; } - if (fscanf(fp, "%d\n", &veid ) != 1) { - if (feof(fp)) - return -1; - - openvzError(NULL, VIR_ERR_INTERNAL_ERROR, - "%s", _("Failed to parse vzlist output")); - goto cleanup; - } - - return veid; - - cleanup: + ok = fscanf(fp, "%d\n", &veid ) == 1; fclose(fp); + if (ok && veid >= 0) + return veid; + + openvzError(NULL, VIR_ERR_INTERNAL_ERROR, + "%s", _("Failed to parse vzlist output")); return -1; } -- 1.7.0.401.g84adb

2 3

[libvirt] avoid "make rpm" failure in doc/
by Jim Meyering 25 Feb '10

25 Feb '10

"make rpm" fails like this: Rebuilding the HTML pages from the XML API make[3]: *** No rule to make target `html/libvirt-libvirt.html', needed by `distdir'. Stop. make[3]: *** Waiting for unfinished jobs.... Validating the resulting XHTML pages These two patches fix that: (though "make rpm" still fails for me: extracting debug info from /j/rpmbuild/BUILDROOT/libvirt-0.7.6-1.fc12.x86_6h.. *** ERROR: No build ID note found in /j/rpmbuild/BUILDROOT/libvirt-0.7.6-1.h.. The first (1/2) change is not officially required to solve this problem, but it overlapped, and does fix a non-distdir build failure. >From ace4dca5dfbc022b057ce1d9f3d2770ad3a34716 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Wed, 24 Feb 2010 22:51:47 +0100 Subject: [PATCH 1/2] build: teach apibuild.py to work in a non-srcdir build * docs/Makefile.am (libvirt-api.xml libvirt-refs.xml): Generalize apibuild.py to work in a non-srcdir build. Pass "srcdir" to it. * docs/apibuild.py (rebuild): Honor the $srcdir envvar. --- docs/Makefile.am | 2 +- docs/apibuild.py | 10 +++++++--- 2 files changed, 8 insertions(+), 4 deletions(-) diff --git a/docs/Makefile.am b/docs/Makefile.am index e47b668..dbbd1ef 100644 --- a/docs/Makefile.am +++ b/docs/Makefile.am @@ -132,7 +132,7 @@ devhelp/index.html devhelp/libvirt.devhelp: libvirt-api.xml $(devhelpxsl) libvirt-api.xml libvirt-refs.xml: apibuild.py \ $(srcdir)/../include/libvirt/*.h \ $(srcdir)/../src/libvirt.c $(srcdir)/../src/util/virterror.c - -$(srcdir)/apibuild.py + -srcdir=$(srcdir) $(srcdir)/apibuild.py clean-local: rm -f *~ *.bak *.hierarchy *.signals *-unused.txt diff --git a/docs/apibuild.py b/docs/apibuild.py index e233c8b..0ab5db2 100755 --- a/docs/apibuild.py +++ b/docs/apibuild.py @@ -2102,10 +2102,14 @@ class docBuilder: def rebuild(): builder = None - if glob.glob("../src/libvirt.c") != [] : + srcdir = os.environ["srcdir"] + if glob.glob(srcdir + "/../src/libvirt.c") != [] : print "Rebuilding API description for libvirt" - builder = docBuilder("libvirt", ["../src", "../src/util", "../include/libvirt"], - []) + builder = docBuilder("libvirt", + [srcdir + "/../src", + srcdir + "/../src/util", + srcdir + "/../include/libvirt"], + []) elif glob.glob("src/libvirt.c") != [] : print "Rebuilding API description for libvir" builder = docBuilder("libvirt", ["src", "src/util", "include/libvirt"], -- 1.7.0.401.g84adb >From a79fb1c491197a91094cd2345c3d7bbba40da2fd Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Thu, 25 Feb 2010 10:35:20 +0100 Subject: [PATCH 2/2] build: avoid "make rpm" failure in docs/ Add missing rule to build html/libvirt-libvirt.html. Use a GNU Make pattern rule to avoid running apibuild.py once for each out-of-date target, in a parallel build. * docs/Makefile.am --- docs/Makefile.am | 15 +++++++++------ 1 files changed, 9 insertions(+), 6 deletions(-) diff --git a/docs/Makefile.am b/docs/Makefile.am index dbbd1ef..edf6fc8 100644 --- a/docs/Makefile.am +++ b/docs/Makefile.am @@ -123,17 +123,20 @@ html/index.html: libvirt-api.xml newapi.xsl page.xsl sitemap.html.in $(XMLLINT) --nonet --valid --noout html/*.html ; \ else echo "missing XHTML1 DTD" ; fi ; fi ); -devhelp/index.html devhelp/libvirt.devhelp: libvirt-api.xml $(devhelpxsl) +$(devhelphtml): libvirt-api.xml $(devhelpxsl) -@(echo Rebuilding devhelp files) -@(if [ -x $(XSLTPROC) ] ; then \ - $(XSLTPROC) --nonet -o devhelp/libvirt.devhelp $(top_srcdir)/docs/devhelp/devhelp.xsl $(top_srcdir)/docs/libvirt-api.xml ; fi ); + $(XSLTPROC) --nonet -o devhelp/libvirt.devhelp \ + $(top_srcdir)/docs/devhelp/devhelp.xsl libvirt-api.xml ; fi ); - -libvirt-api.xml libvirt-refs.xml: apibuild.py \ - $(srcdir)/../include/libvirt/*.h \ - $(srcdir)/../src/libvirt.c $(srcdir)/../src/util/virterror.c +html/%-%.html html/%-virterror.html %-api.xml %-refs.xml: $(srcdir)/apibuild.py -srcdir=$(srcdir) $(srcdir)/apibuild.py +html/%-%.html html/%-virterror.html %-api.xml %-refs.xml: \ + $(srcdir)/../include/%/*.h \ + $(srcdir)/../src/%.c \ + $(srcdir)/../src/util/virterror.c + clean-local: rm -f *~ *.bak *.hierarchy *.signals *-unused.txt -- 1.7.0.401.g84adb

2 3

[libvirt] [PATCH] build: avoid non-srcdir "make distcheck" failures (CLEANFILES)
by Jim Meyering 25 Feb '10

25 Feb '10

A quick review won't hurt, but this doesn't really need one. >From 9ac893a1342b31ef7df6fc57c03767c49ef98d5b Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Thu, 25 Feb 2010 09:28:51 +0100 Subject: [PATCH 1/4] build: avoid non-srcdir "make distcheck" failures (CLEANFILES) * docs/Makefile.am (MAINTAINERCLEANFILES): Use this variable for generated-and-distributed files, not "CLEANFILES". Besides, "make clean" and "make distclean" should not delete distributed files. --- docs/Makefile.am | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/Makefile.am b/docs/Makefile.am index 07ea9bf..dbbd1ef 100644 --- a/docs/Makefile.am +++ b/docs/Makefile.am @@ -17,7 +17,7 @@ apipng = \ html/home.png \ html/right.png -devhelphtml = \ +devhelphtml = \ devhelp/libvirt.devhelp \ devhelp/index.html \ devhelp/general.html \ @@ -79,7 +79,7 @@ EXTRA_DIST= \ sitemap.html.in \ ChangeLog.awk -CLEANFILES = $(dot_html) $(apihtml) $(devhelphtml) +MAINTAINERCLEANFILES = $(dot_html) $(apihtml) $(devhelphtml) all: web -- 1.7.0.401.g84adb

2 1

[libvirt] [PATCH] qemu: Report binary path if error parsing -help
by Cole Robinson 25 Feb '10

25 Feb '10

Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- src/qemu/qemu_conf.c | 14 ++++++++------ src/qemu/qemu_conf.h | 3 ++- tests/qemuhelptest.c | 3 ++- 3 files changed, 12 insertions(+), 8 deletions(-) diff --git a/src/qemu/qemu_conf.c b/src/qemu/qemu_conf.c index e3da08a..1f8a9c3 100644 --- a/src/qemu/qemu_conf.c +++ b/src/qemu/qemu_conf.c @@ -1237,7 +1237,8 @@ static unsigned long long qemudComputeCmdFlags(const char *help, #define SKIP_BLANKS(p) do { while ((*(p) == ' ') || (*(p) == '\t')) (p)++; } while (0) -int qemudParseHelpStr(const char *help, +int qemudParseHelpStr(const char *qemu, + const char *help, unsigned long long *flags, unsigned int *version, unsigned int *is_kvm, @@ -1308,8 +1309,8 @@ fail: p = strndup(help, p - help); qemuReportError(VIR_ERR_INTERNAL_ERROR, - _("cannot parse QEMU version number in '%s'"), - p ? p : help); + _("cannot parse %s version number in '%s'"), + qemu, p ? p : help); VIR_FREE(p); @@ -1340,12 +1341,13 @@ int qemudExtractVersionInfo(const char *qemu, enum { MAX_HELP_OUTPUT_SIZE = 1024*64 }; int len = virFileReadLimFD(newstdout, MAX_HELP_OUTPUT_SIZE, &help); if (len < 0) { - virReportSystemError(errno, "%s", - _("Unable to read QEMU help output")); + virReportSystemError(errno, + _("Unable to read %s help output"), qemu); goto cleanup2; } - if (qemudParseHelpStr(help, &flags, &version, &is_kvm, &kvm_version) == -1) + if (qemudParseHelpStr(qemu, help, &flags, + &version, &is_kvm, &kvm_version) == -1) goto cleanup2; if (retversion) diff --git a/src/qemu/qemu_conf.h b/src/qemu/qemu_conf.h index ae187dc..c8757c5 100644 --- a/src/qemu/qemu_conf.h +++ b/src/qemu/qemu_conf.h @@ -175,7 +175,8 @@ int qemudExtractVersionInfo (const char *qemu, unsigned int *version, unsigned long long *qemuCmdFlags); -int qemudParseHelpStr (const char *str, +int qemudParseHelpStr (const char *qemu, + const char *str, unsigned long long *qemuCmdFlags, unsigned int *version, unsigned int *is_kvm, diff --git a/tests/qemuhelptest.c b/tests/qemuhelptest.c index 0f2b509..ad355d7 100644 --- a/tests/qemuhelptest.c +++ b/tests/qemuhelptest.c @@ -49,7 +49,8 @@ static int testHelpStrParsing(const void *data) if (virtTestLoadFile(path, &help, MAX_HELP_OUTPUT_SIZE) < 0) return -1; - if (qemudParseHelpStr(help, &flags, &version, &is_kvm, &kvm_version) == -1) + if (qemudParseHelpStr("QEMU", help, &flags, + &version, &is_kvm, &kvm_version) == -1) return -1; if (flags != info->flags) { -- 1.6.5.2

2 2

[libvirt] [PATCH] remote: Improve daemon startup error reporting
by Cole Robinson 25 Feb '10

25 Feb '10

If I toggle enable_tcp in libvirtd.conf and add --listen in /etc/init.d/libvirtd, I get the unhelpful error: Starting libvirtd daemon: error: Unable to initialize network sockets. Running without --daemon provides much more useful info: sudo libvirtd --listen 11:29:26.117: error : remoteCheckCertFile:270 : Cannot access CA certificate '/etc/pki/CA/cacert.pem': No such file or directory The daemon architecture makes it difficult to report this useful info if daemonized, so point users to /var/log/messages and dropping the --daemon flag if they want more info. Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- daemon/libvirtd.c | 5 ++++- 1 files changed, 4 insertions(+), 1 deletions(-) diff --git a/daemon/libvirtd.c b/daemon/libvirtd.c index fefa7aa..cee6f13 100644 --- a/daemon/libvirtd.c +++ b/daemon/libvirtd.c @@ -474,7 +474,10 @@ static int daemonForkIntoBackground(void) { goto again; if (ret == 1 && status != 0) { - fprintf(stderr, "error: %s\n", virDaemonErrTypeToString(status)); + fprintf(stderr, + "error: %s. Check /var/log/messages or run without " + "--daemon for more info.\n", + virDaemonErrTypeToString(status)); } _exit(ret == 1 && status == 0 ? 0 : 1); } -- 1.6.5.2

2 1

[libvirt] [PATCH] virsh: Show errors reported by nonAPI functions
by Cole Robinson 25 Feb '10

25 Feb '10

Only API calls trigger the error callback, which is required for proper virsh error reporting. Since we use non API functions from util/, make sure we properly report these errors. Fixes lack of error message from 'virsh create idontexit.xml' Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- tools/virsh.c | 12 ++++++++++-- 1 files changed, 10 insertions(+), 2 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 5fdbbe5..74e3fad 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -352,8 +352,16 @@ virshErrorHandler(void *unused ATTRIBUTE_UNUSED, virErrorPtr error) static void virshReportError(vshControl *ctl) { - if (last_error == NULL) - return; + if (last_error == NULL) { + /* Calling directly into libvirt util functions won't trigger the + * error callback (which sets last_error), so check it ourselves. + * + * If the returned error has CODE_OK, this most likely means that + * no error was ever raised, so just ignore */ + last_error = virSaveLastError(); + if (!last_error || last_error->code == VIR_ERR_OK) + return NULL; + } if (last_error->code == VIR_ERR_OK) { vshError(ctl, "%s", _("unknown error")); -- 1.6.5.2

2 1

[libvirt] [PATCH] Use standard spacing for user/pass prompt
by Cole Robinson 24 Feb '10

24 Feb '10

Kind of minor, but it annoys me that the default auth callback doesn't put a space between the prompt and the input, like a typical terminal, ssh, etc. This patch changes the current prompt: Please enter your authentication name:myuser to Please enter your authentication name: myuser Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- src/libvirt.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/src/libvirt.c b/src/libvirt.c index be17668..9d50c92 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -159,7 +159,7 @@ static int virConnectAuthCallbackDefault(virConnectCredentialPtr cred, case VIR_CRED_AUTHNAME: case VIR_CRED_ECHOPROMPT: case VIR_CRED_REALM: - if (printf("%s:", cred[i].prompt) < 0) + if (printf("%s: ", cred[i].prompt) < 0) return -1; if (fflush(stdout) != 0) return -1; @@ -178,7 +178,7 @@ static int virConnectAuthCallbackDefault(virConnectCredentialPtr cred, case VIR_CRED_PASSPHRASE: case VIR_CRED_NOECHOPROMPT: - if (printf("%s:", cred[i].prompt) < 0) + if (printf("%s: ", cred[i].prompt) < 0) return -1; if (fflush(stdout) != 0) return -1; -- 1.6.5.2

2 1

[libvirt] the price we pay for using gettext-0.14
by Jim Meyering 24 Feb '10

24 Feb '10

Make distcheck currently fails like this: make[2]: Entering directory `/t/libvirt-0.7.6/_build/po' /bin/sh @MKINSTALLDIRS@ /t/libvirt-0.7.6/_inst/share /bin/sh: @MKINSTALLDIRS@: No such file or directory make[2]: *** [install-data-yes] Error 127 That is due to our use of a version of gettext that is way too old: $ grep GETT configure.ac AM_GNU_GETTEXT_VERSION([0.14.1]) The work-around is to define (via AC_SUBST) the now-obsolete symbol, MKINSTALLDIRS, so that @MKINSTALLDIRS@ in the autopoint-provided Makefile.in.in (from the old version of gettext) ends up being substituted. As I recall, people insist on using this old version of gettext, because nothing newer is supplied by RHEL5.x. IMHO, that is RHEL5's problem, and should not be allowed to cause continuing problems in upstream projects. It is easy and safe to install a newer version of gettext, especially since any project using autopoint will automatically continue to get the same bit-rotting old files: a gettext installation includes an archive of all older versions of its files, to provide just this type of backwards compatibility. We had a similar problem in libguestfs, and it was not trivial to diagnose there, either, even though I'd seen it at least once or twice before then. Here's the work-around: [just realized this requires a comment. adding that now] >From 16b9418427101a495b7fdd6917f518f254d22720 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Wed, 24 Feb 2010 10:53:44 +0100 Subject: [PATCH] build: ensure that MKINSTALLDIRS is AC_SUBST-defined since we're using gettext-0.14.1, which uses that now-obsolete automake symbol. Otherwise, make distcheck would fails like this: make[2]: Entering directory `/t/libvirt-0.7.6/_build/po' /bin/sh @MKINSTALLDIRS@ /t/libvirt-0.7.6/_inst/share /bin/sh: @MKINSTALLDIRS@: No such file or directory make[2]: *** [install-data-yes] Error 127 * configure.ac (MKINSTALLDIRS): Define. --- configure.ac | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/configure.ac b/configure.ac index 117cb20..2f79970 100644 --- a/configure.ac +++ b/configure.ac @@ -49,6 +49,8 @@ AC_PROG_CPP AC_OBJEXT +AC_SUBST([MKINSTALLDIRS], [$ac_aux_dir/mkinstalldirs]) + dnl gl_INIT uses m4_foreach_w, yet that is not defined in autoconf-2.59. dnl In order to accommodate developers with such old tools, here's a dnl replacement definition. -- 1.7.0.367.g566c3

3 4

[libvirt] [PATCH] maint: relax git minimum version
by Eric Blake 24 Feb '10

24 Feb '10

From: Eric Blake <eblake(a)redhat.com> Requiring git 1.6.4, just for the optional GNULIB_SRCDIR support, was too harsh. Resynchronize from gnulib. * .gnulib: Import from latest gnulib. * bootstrap: Re-synchronize from .gnulib/build-aux. * bootstrap.conf: Drop git to 1.5.5. * README-hacking: Document use of GNULIB_SRCDIR. --- Chris is currently testing that RHEL 5 with git 1.5.5 can actually grok this patch. We don't have a README-prereq (but README-hacking was borrowed from coreutils, which does). Should we add one? The README-prereq in coreutils is not quite relevant for a straight-across copy. .gnulib | 2 +- README-hacking | 9 +++++++-- bootstrap | 41 +++++++++++++++++++++++++++++++++++------ bootstrap.conf | 2 +- 4 files changed, 44 insertions(+), 10 deletions(-) diff --git a/.gnulib b/.gnulib index 11fbc57..108cfb8 160000 --- a/.gnulib +++ b/.gnulib @@ -1 +1 @@ -Subproject commit 11fbc57405a118e6ec9a3ebc19bbf5ececdae4d6 +Subproject commit 108cfb803174abf17eb101e9999b9d1a642ce1ba diff --git a/README-hacking b/README-hacking index 4105a3e..80b022c 100644 --- a/README-hacking +++ b/README-hacking @@ -12,8 +12,7 @@ requirements on anyone wishing to build from the just-checked-out sources. Note the requirements to build the released archive are much less and are just the requirements of the standard ./configure && make procedure. Specific development tools and versions will be checked for and listed by -the bootstrap script. See README-prereq for specific notes on obtaining -these prerequisite tools. +the bootstrap script. Valgrind <http://valgrind.org/> is also highly recommended, if Valgrind supports your architecture. See also README-valgrind. @@ -28,6 +27,12 @@ You can get a copy of the source repository like this: $ git clone git://libvirt.org/libvirt $ cd libvirt +As an optional step, if you already have a copy of the gnulib git +repository on your hard drive, then you can use it as a reference to +reduce download time and disk space requirements: + + $ export GNULIB_SRCDIR=/path/to/gnulib + The next step is to get all required pieces from gnulib, to run autoreconf, and to invoke ./configure: diff --git a/bootstrap b/bootstrap index be96444..365a3d9 100755 --- a/bootstrap +++ b/bootstrap @@ -1,4 +1,6 @@ #! /bin/sh +# Print a version string. +scriptversion=2010-02-24.17; # UTC # Bootstrap this package from checked-out sources. @@ -17,7 +19,15 @@ # You should have received a copy of the GNU General Public License # along with this program. If not, see <http://www.gnu.org/licenses/>. -# Written by Paul Eggert. +# Originally written by Paul Eggert. The canonical version of this +# script is maintained as build-aux/bootstrap in gnulib, however, to +# be useful to your project, you should place a copy of it under +# version control in the top-level directory of your project. The +# intent is that all customization can be done with a bootstrap.conf +# file also maintained in your version control; gnulib comes with a +# template build-aux/bootstrap.conf to get you started. + +# Please report bugs or propose patches to bug-gnulib(a)gnu.org. nl=' ' @@ -161,7 +171,6 @@ vc_ignore=auto # die otherwise. find_tool () { - # Find sha1sum, named gsha1sum on MacPorts. find_tool_envvar=$1 shift find_tool_names=$@ @@ -424,12 +433,28 @@ case ${GNULIB_SRCDIR--} in GNULIB_SRCDIR=$gnulib_path ;; *) - # Use GNULIB_SRCDIR as a reference. Assumes git 1.6.4 or newer. + # Use GNULIB_SRCDIR as a reference. if test -d "$GNULIB_SRCDIR"/.git && \ git_modules_config submodule.gnulib.url >/dev/null; then echo "$0: getting gnulib files..." - git submodule update --init --reference "$GNULIB_SRCDIR" \ + if git submodule --help|grep reference > /dev/null; then + # Prefer the one-liner available in git 1.6.4 or newer. + git submodule update --init --reference "$GNULIB_SRCDIR" \ "$gnulib_path" || exit $? + else + # This fallback allows at least git 1.5.5. + if test -f "$gnulib_path"/gnulib-tool; then + # Since file already exists, assume submodule init already complete. + git submodule update || exit $? + else + # Older git can't clone into an empty directory. + rmdir "$gnulib_path" 2>/dev/null + git clone --reference "$GNULIB_SRCDIR" \ + "$(git_modules_config submodule.gnulib.url)" "$gnulib_path" \ + && git submodule init && git submodule update \ + || exit $? + fi + fi GNULIB_SRCDIR=$gnulib_path fi ;; @@ -819,6 +844,10 @@ bootstrap_epilogue echo "$0: done. Now you can run './configure'." -# Local Variables: -# indent-tabs-mode: nil +# Local variables: +# eval: (add-hook 'write-file-hooks 'time-stamp) +# time-stamp-start: "scriptversion=" +# time-stamp-format: "%:y-%02m-%02d.%02H" +# time-stamp-time-zone: "UTC" +# time-stamp-end: "; # UTC" # End: diff --git a/bootstrap.conf b/bootstrap.conf index d315e04..58ef2ab 100644 --- a/bootstrap.conf +++ b/bootstrap.conf @@ -124,7 +124,7 @@ autoconf 2.59 automake 1.9.6 autopoint - gettext - -git 1.6.4 +git 1.5.5 gzip - libtool - perl 5.5 -- 1.6.6.1

2 2

[libvirt] [PATCH v2] libvirtd: Better initscript error reporting
by Cole Robinson 24 Feb '10

24 Feb '10

>From time to time I bork my install, and hate it when the initscript returns no info. This patch removes the sanity check, which lets the shell give us 'command not found' or 'permission denied' errors. Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- daemon/libvirtd.init.in | 3 --- 1 files changed, 0 insertions(+), 3 deletions(-) diff --git a/daemon/libvirtd.init.in b/daemon/libvirtd.init.in index 071fe2c..b808ab3 100644 --- a/daemon/libvirtd.init.in +++ b/daemon/libvirtd.init.in @@ -32,9 +32,6 @@ # pidfile: @localstatedir@/run/libvirtd.pid # -# Sanity checks. -[ -x @sbindir@/libvirtd ] || exit 0 - # Source function library. . @sysconfdir@/rc.d/init.d/functions -- 1.6.5.2

1 0

[libvirt] [PATCH] remote: Improve error message when libvirtd isn't running
by Cole Robinson 24 Feb '10

24 Feb '10

While this could lead people in the wrong direction, people most commonly hit this error when libvirtd isn't running, so let's mention a possible fix. Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- src/remote/remote_driver.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/src/remote/remote_driver.c b/src/remote/remote_driver.c index 8914c39..e10312c 100644 --- a/src/remote/remote_driver.c +++ b/src/remote/remote_driver.c @@ -723,8 +723,8 @@ doRemoteOpen (virConnectPtr conn, } } virReportSystemError(errno, - _("unable to connect to '%s'"), - sockname); + _("unable to connect to '%s', libvirtd may need to be started"), + sockname); goto failed; } -- 1.6.5.2

2 1

[libvirt] [PATCH] libvirtd: Better initscript error reporting
by Cole Robinson 24 Feb '10

24 Feb '10

>From time to time I bork my install, and hate it when the initscript returns no info. This patch makes things a bit more clear. Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- daemon/libvirtd.init.in | 4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/daemon/libvirtd.init.in b/daemon/libvirtd.init.in index 071fe2c..c5d4eee 100644 --- a/daemon/libvirtd.init.in +++ b/daemon/libvirtd.init.in @@ -32,8 +32,10 @@ # pidfile: @localstatedir@/run/libvirtd.pid # +LIBVIRTD_BIN="@sbindir@/libvirtd" + # Sanity checks. -[ -x @sbindir@/libvirtd ] || exit 0 +[ ! -e $LIBVIRTD_BIN ] && echo "$LIBVIRTD_BIN not found." && exit 1 # Source function library. . @sysconfdir@/rc.d/init.d/functions -- 1.6.5.2

2 2

[libvirt] [PATCH] Ignore SIGWINCH in remote client call to poll(2) (RHBZ#567931).
by Richard W.M. Jones 24 Feb '10

24 Feb '10

-- Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v

3 6

[libvirt] [PATCH] conf: storage: Remove iSCSI <auth> parsing
by Cole Robinson 24 Feb '10

24 Feb '10

This was never wired up, and even generated broken XML until 0.7.2, so clearly no one was trying to use it. Dan recommended its removal, so lets drop it. Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- docs/schemas/storagepool.rng | 19 --------- src/conf/storage_conf.c | 48 ----------------------- src/conf/storage_conf.h | 18 -------- tests/storagepoolxml2xmlin/pool-iscsi-auth.xml | 17 -------- tests/storagepoolxml2xmlout/pool-iscsi-auth.xml | 20 --------- tests/storagepoolxml2xmltest.c | 1 - 6 files changed, 0 insertions(+), 123 deletions(-) delete mode 100644 tests/storagepoolxml2xmlin/pool-iscsi-auth.xml delete mode 100644 tests/storagepoolxml2xmlout/pool-iscsi-auth.xml diff --git a/docs/schemas/storagepool.rng b/docs/schemas/storagepool.rng index 247664e..bcdca62 100644 --- a/docs/schemas/storagepool.rng +++ b/docs/schemas/storagepool.rng @@ -234,22 +234,6 @@ </element> </define> - <define name='sourceinfoauth'> - <element name='auth'> - <attribute name='type'> - <choice> - <value>chap</value> - </choice> - </attribute> - <attribute name='login'> - <text/> - </attribute> - <attribute name='passwd'> - <text/> - </attribute> - </element> - </define> - <define name='sourcefmtfs'> <optional> <element name='format'> @@ -374,9 +358,6 @@ <optional> <ref name='initiatorinfoiqn'/> </optional> - <optional> - <ref name='sourceinfoauth'/> - </optional> </element> </define> diff --git a/src/conf/storage_conf.c b/src/conf/storage_conf.c index 19a1db9..dd375b9 100644 --- a/src/conf/storage_conf.c +++ b/src/conf/storage_conf.c @@ -283,11 +283,6 @@ virStoragePoolSourceFree(virStoragePoolSourcePtr source) { VIR_FREE(source->name); VIR_FREE(source->adapter); VIR_FREE(source->initiator.iqn); - - if (source->authType == VIR_STORAGE_POOL_AUTH_CHAP) { - VIR_FREE(source->auth.chap.login); - VIR_FREE(source->auth.chap.passwd); - } } void @@ -363,26 +358,6 @@ virStoragePoolObjRemove(virStoragePoolObjListPtr pools, static int -virStoragePoolDefParseAuthChap(xmlXPathContextPtr ctxt, - virStoragePoolAuthChapPtr auth) { - auth->login = virXPathString("string(./auth/@login)", ctxt); - if (auth->login == NULL) { - virStorageReportError(VIR_ERR_XML_ERROR, - "%s", _("missing auth host attribute")); - return -1; - } - - auth->passwd = virXPathString("string(./auth/@passwd)", ctxt); - if (auth->passwd == NULL) { - virStorageReportError(VIR_ERR_XML_ERROR, - "%s", _("missing auth passwd attribute")); - return -1; - } - - return 0; -} - -static int virStoragePoolDefParseSource(xmlXPathContextPtr ctxt, virStoragePoolSourcePtr source, int pool_type, @@ -445,25 +420,6 @@ virStoragePoolDefParseSource(xmlXPathContextPtr ctxt, source->dir = virXPathString("string(./dir/@path)", ctxt); source->adapter = virXPathString("string(./adapter/@name)", ctxt); - authType = virXPathString("string(./auth/@type)", ctxt); - if (authType == NULL) { - source->authType = VIR_STORAGE_POOL_AUTH_NONE; - } else { - if (STREQ(authType, "chap")) { - source->authType = VIR_STORAGE_POOL_AUTH_CHAP; - } else { - virStorageReportError(VIR_ERR_XML_ERROR, - _("unknown auth type '%s'"), - (const char *)authType); - goto cleanup; - } - } - - if (source->authType == VIR_STORAGE_POOL_AUTH_CHAP) { - if (virStoragePoolDefParseAuthChap(ctxt, &source->auth.chap) < 0) - goto cleanup; - } - ret = 0; cleanup: ctxt->node = relnode; @@ -867,10 +823,6 @@ virStoragePoolSourceFormat(virBufferPtr buf, } - if (src->authType == VIR_STORAGE_POOL_AUTH_CHAP) - virBufferVSprintf(buf," <auth type='chap' login='%s' passwd='%s'/>\n", - src->auth.chap.login, - src->auth.chap.passwd); virBufferAddLit(buf," </source>\n"); return 0; diff --git a/src/conf/storage_conf.h b/src/conf/storage_conf.h index c643984..1408128 100644 --- a/src/conf/storage_conf.h +++ b/src/conf/storage_conf.h @@ -134,19 +134,6 @@ enum virStoragePoolDeviceType { }; -enum virStoragePoolAuthType { - VIR_STORAGE_POOL_AUTH_NONE, - VIR_STORAGE_POOL_AUTH_CHAP, -}; - -typedef struct _virStoragePoolAuthChap virStoragePoolAuthChap; -typedef virStoragePoolAuthChap *virStoragePoolAuthChapPtr; -struct _virStoragePoolAuthChap { - char *login; - char *passwd; -}; - - /* * For remote pools, info on how to reach the host */ @@ -232,11 +219,6 @@ struct _virStoragePoolSource { /* Initiator IQN */ virStoragePoolSourceInitiatorAttr initiator; - int authType; /* virStoragePoolAuthType */ - union { - virStoragePoolAuthChap chap; - } auth; - int format; /* Pool type specific format such as filesystem type, or lvm version, etc */ }; diff --git a/tests/storagepoolxml2xmlin/pool-iscsi-auth.xml b/tests/storagepoolxml2xmlin/pool-iscsi-auth.xml deleted file mode 100644 index f7d4d52..0000000 --- a/tests/storagepoolxml2xmlin/pool-iscsi-auth.xml +++ /dev/null @@ -1,17 +0,0 @@ -<pool type='iscsi'> - <name>virtimages</name> - <uuid>e9392370-2917-565e-692b-d057f46512d6</uuid> - <source> - <host name="iscsi.example.com"/> - <device path="demo-target"/> - <auth type='chap' login='foobar' passwd='frobbar'/> - </source> - <target> - <path>/dev/disk/by-path</path> - <permissions> - <mode>0700</mode> - <owner>0</owner> - <group>0</group> - </permissions> - </target> -</pool> diff --git a/tests/storagepoolxml2xmlout/pool-iscsi-auth.xml b/tests/storagepoolxml2xmlout/pool-iscsi-auth.xml deleted file mode 100644 index 557295d..0000000 --- a/tests/storagepoolxml2xmlout/pool-iscsi-auth.xml +++ /dev/null @@ -1,20 +0,0 @@ -<pool type='iscsi'> - <name>virtimages</name> - <uuid>e9392370-2917-565e-692b-d057f46512d6</uuid> - <capacity>0</capacity> - <allocation>0</allocation> - <available>0</available> - <source> - <host name='iscsi.example.com'/> - <device path='demo-target'/> - <auth type='chap' login='foobar' passwd='frobbar'/> - </source> - <target> - <path>/dev/disk/by-path</path> - <permissions> - <mode>0700</mode> - <owner>0</owner> - <group>0</group> - </permissions> - </target> -</pool> diff --git a/tests/storagepoolxml2xmltest.c b/tests/storagepoolxml2xmltest.c index 4550407..1d7094b 100644 --- a/tests/storagepoolxml2xmltest.c +++ b/tests/storagepoolxml2xmltest.c @@ -91,7 +91,6 @@ mymain(int argc, char **argv) DO_TEST("pool-logical-create"); DO_TEST("pool-disk"); DO_TEST("pool-iscsi"); - DO_TEST("pool-iscsi-auth"); DO_TEST("pool-netfs"); DO_TEST("pool-scsi"); DO_TEST("pool-mpath"); -- 1.6.6

3 3

[libvirt] [PATCH 1/2] build: avoid non-srcdir installation failure (apibuild.py)
by Jim Meyering 24 Feb '10

24 Feb '10

Running "make distcheck" from a non-srcdir build was failing like this: /bin/sh: ./apibuild.py: No such file or directory Fixed that with the first patch below (the apibuild.py script is not in ".", but in $(srcdir)). Then retrying, it got a little further but failed like this: (../../docs/apibuild.py) make[3]: *** No rule to make target `sitemap.html.in', needed by `html/index.html'. Stop. That was because sitemap.html.in was not being distributed. Adding it to EXTRA_DIST solved that: >From f0fb87ed1838a9a67d03985e35a4c7625da9d95e Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Wed, 24 Feb 2010 14:18:40 +0100 Subject: [PATCH 1/2] build: avoid non-srcdir installation failure (apibuild.py) * docs/Makefile.am (libvirt-api.xml): Insert missing "$(srcdir)/". --- docs/Makefile.am | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/docs/Makefile.am b/docs/Makefile.am index cce7a48..26b1d29 100644 --- a/docs/Makefile.am +++ b/docs/Makefile.am @@ -131,7 +131,7 @@ devhelp/index.html devhelp/libvirt.devhelp: libvirt-api.xml $(devhelpxsl) libvirt-api.xml libvirt-refs.xml: apibuild.py \ $(srcdir)/../include/libvirt/*.h \ $(srcdir)/../src/libvirt.c $(srcdir)/../src/util/virterror.c - -(./apibuild.py) + -($(srcdir)/apibuild.py) clean-local: rm -f *~ *.bak *.hierarchy *.signals *-unused.txt -- 1.7.0.367.g566c3 >From 8e948091c4c3b2346be8ede0309ce768136621e9 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Wed, 24 Feb 2010 14:15:26 +0100 Subject: [PATCH 2/2] build: avoid non-srcdir installation failure (sitemap.html.in) * docs/Makefile.am (EXTRA_DIST): Add sitemap.html.in. --- docs/Makefile.am | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/docs/Makefile.am b/docs/Makefile.am index 26b1d29..4828752 100644 --- a/docs/Makefile.am +++ b/docs/Makefile.am @@ -76,6 +76,7 @@ EXTRA_DIST= \ $(devhelphtml) $(devhelppng) $(devhelpcss) $(devhelpxsl) \ $(xml) $(fig) $(png) \ $(patches) \ + sitemap.html.in \ ChangeLog.awk CLEANFILES = $(dot_html) $(apihtml) $(devhelphtml) -- 1.7.0.367.g566c3

2 2

[libvirt] [PATCH] build: make git submodule checking more reliable
by Jim Meyering 24 Feb '10

24 Feb '10

After investigating some autogen.sh/bootstrap-vs-new-gnulib problems today, I realized that libvirt's SHA1-comparing code lacked improvements I'd made in libguestfs, which does the same thing. AFAIK, this is not related to the problems we've seen today, but it might well help someone down the road. Here's the result of that synchronization: >From fada075bcaf46d1bd0c51165475594544e18ecfd Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Wed, 24 Feb 2010 12:01:09 +0100 Subject: [PATCH] build: make git submodule checking more reliable * cfg.mk (_curr_status): Accommodate leading "+" or "-" in the output of git submodule status. * autogen.sh: Likewise. --- autogen.sh | 2 +- cfg.mk | 4 +++- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/autogen.sh b/autogen.sh index 833b1d1..9ae8d67 100755 --- a/autogen.sh +++ b/autogen.sh @@ -65,7 +65,7 @@ fi # Ensure that whenever we pull in a gnulib update or otherwise change to a # different version (i.e., when switching branches), we also rerun ./bootstrap. curr_status=.git-module-status -t=$(git submodule status) +t=$(git submodule status|sed 's/^[ +-]//;s/ .*//') if test "$t" = "$(cat $curr_status 2>/dev/null)"; then : # good, it's up to date, all we need is autoreconf autoreconf -if diff --git a/cfg.mk b/cfg.mk index 17bb341..8317ae2 100644 --- a/cfg.mk +++ b/cfg.mk @@ -262,7 +262,9 @@ ifeq (0,$(MAKELEVEL)) # submodule name. Contrast these: # -b653eda3ac4864de205419d9f41eec267cb89eeb .gnulib # b653eda3ac4864de205419d9f41eec267cb89eeb .gnulib (v0.0-2286-gb653eda) - _submodule_hash = sed 's/.//;s/ .*//' + # $ cat .git-module-status + # b653eda3ac4864de205419d9f41eec267cb89eeb + _submodule_hash = sed 's/^[ +-]//;s/ .*//' _update_required := $(shell \ actual=$$(git submodule status | $(_submodule_hash)); \ stamp="$$($(_submodule_hash) $(_curr_status) 2>/dev/null)"; \ -- 1.7.0.377.g90adb

2 1

[libvirt] [PATCH] build: avoid non-srcdir "make distcheck" failures (XML_EXAMPLES)
by Jim Meyering 24 Feb '10

24 Feb '10

This avoids two of the 3 remaining failures, -error: loading host definition file '/t/libvirt-0.7.6/examples/xml/test/testnode.xml': No such file or directory -error: failed to connect to the hypervisor +error: failed to get domain '4294967298' >From b2318b64abc8ac23b4fa43dc52d3c2a32226b8cb Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Wed, 24 Feb 2010 15:52:44 +0100 Subject: [PATCH] build: avoid non-srcdir "make distcheck" failures (XML_EXAMPLES) * Makefile.am (XML_EXAMPLES): Use $(wildcard in $(srcdir)-aware manner. --- Makefile.am | 4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/Makefile.am b/Makefile.am index a02cd26..09439c8 100644 --- a/Makefile.am +++ b/Makefile.am @@ -9,7 +9,9 @@ SUBDIRS = gnulib/lib include src daemon tools proxy docs gnulib/tests \ ACLOCAL_AMFLAGS = -I m4 -I gnulib/m4 -XML_EXAMPLES = $(wildcard examples/xml/test/*.xml) $(wildcard examples/xml/storage/*.xml) +XML_EXAMPLES = \ + $(patsubst $(srcdir)/%,%,$(wildcard $(addprefix $(srcdir)/examples/xml/, \ + test/*.xml storage/*.xml))) EXTRA_DIST = \ ChangeLog-old \ -- 1.7.0.367.g566c3

2 2

[libvirt] [PATCH 1/4] Avoid searching for windres when not building for cygwin or mingw.
by Diego Elio Pettenò 24 Feb '10

24 Feb '10

Just checking for a windres tool might hit even on Linux systems when building for Linux (e.g.: when using Gentoo and having built binutils with multitarget support), and will then fail to link properly at the end of the build. Check the host string before deciding whether to look for windres or not. --- configure.ac | 9 +++++++-- 1 files changed, 7 insertions(+), 2 deletions(-) diff --git a/configure.ac b/configure.ac index 117cb20..5b1eb5f 100644 --- a/configure.ac +++ b/configure.ac @@ -1682,8 +1682,13 @@ AC_SUBST([CYGWIN_EXTRA_PYTHON_LIBADD]) AC_SUBST([MINGW_EXTRA_LDFLAGS]) dnl Look for windres to build a Windows icon resource. -AC_CHECK_TOOL([WINDRES], [windres], [no]) -AM_CONDITIONAL([WITH_WIN_ICON], [test "$WINDRES" != "no"]) +case "$host" in + *cygwin* | *mingw* ) + AC_CHECK_TOOL([WINDRES], [windres], []) + ;; +esac + +AM_CONDITIONAL([WITH_WIN_ICON], [test "$WINDRES" != ""]) -- 1.7.0

5 18

[libvirt] [PATCH] build: fix typos in configure-set makefile variable names
by Jim Meyering 24 Feb '10

24 Feb '10

Without this change, the ..._MODULES_... variables are set in configure.ac, but never used. The variables that are actually used in src/Makefile.am are named _MODULE_... >From a3361a6ce032f71273d6f7a6b4b4f54438b1dd73 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Wed, 24 Feb 2010 12:04:27 +0100 Subject: [PATCH] build: fix typos in makefile variable names * configure.ac: Fix typos: s/DRIVER_MODULES_CFLAGS/DRIVER_MODULE_CFLAGS/ s/DRIVER_MODULES_LIBS/DRIVER_MODULE_LIBS/ --- configure.ac | 14 +++++++------- 1 files changed, 7 insertions(+), 7 deletions(-) diff --git a/configure.ac b/configure.ac index 2f79970..43ea771 100644 --- a/configure.ac +++ b/configure.ac @@ -1695,8 +1695,8 @@ AC_ARG_WITH([driver-modules], [], [with_driver_modules=no]) -DRIVER_MODULES_CFLAGS= -DRIVER_MODULES_LIBS= +DRIVER_MODULE_CFLAGS= +DRIVER_MODULE_LIBS= if test "x$with_driver_modules" = "xyes" ; then old_cflags="$CFLAGS" old_libs="$LIBS" @@ -1710,13 +1710,13 @@ if test "x$with_driver_modules" = "xyes" ; then LIBS="$old_libs" fi if test "$with_driver_modules" = "yes"; then - DRIVER_MODULES_CFLAGS="-export-dynamic" - DRIVER_MODULES_LIBS="-ldl" + DRIVER_MODULE_CFLAGS="-export-dynamic" + DRIVER_MODULE_LIBS="-ldl" AC_DEFINE_UNQUOTED([WITH_DRIVER_MODULES], 1, [whether to build drivers as modules]) fi AM_CONDITIONAL([WITH_DRIVER_MODULES], [test "$with_driver_modules" != "no"]) -AC_SUBST([DRIVER_MODULES_CFLAGS]) -AC_SUBST([DRIVER_MODULES_LIBS]) +AC_SUBST([DRIVER_MODULE_CFLAGS]) +AC_SUBST([DRIVER_MODULE_LIBS]) # Set LV_LIBTOOL_OBJDIR to "." or $lt_cv_objdir, depending on whether @@ -1931,7 +1931,7 @@ AC_MSG_NOTICE([]) AC_MSG_NOTICE([Driver Loadable Modules]) AC_MSG_NOTICE([]) if test "$with_driver_modules" != "no" ; then -AC_MSG_NOTICE([ dlopen: $DRIVER_MODULES_CFLAGS $DRIVER_MODULES_LIBS]) +AC_MSG_NOTICE([ dlopen: $DRIVER_MODULE_CFLAGS $DRIVER_MODULE_LIBS]) else AC_MSG_NOTICE([ dlopen: no]) fi -- 1.7.0.367.g566c3

3 6

[libvirt] [PATCH] build: avoid non-srcdir installation failure
by Jim Meyering 24 Feb '10

24 Feb '10

>From 28193acd8f1cab9b91cde77afef68781e5c740b2 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Wed, 24 Feb 2010 14:01:58 +0100 Subject: [PATCH] build: avoid non-srcdir installation failure * docs/Makefile.am (html/index.html): Insert missing $(srcdir)/ on argument to $(XSLTPROC). This would make "make distcheck" fail when run from a non-srcdir build. Reported by Amit Shah --- docs/Makefile.am | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/docs/Makefile.am b/docs/Makefile.am index 4df5154..cce7a48 100644 --- a/docs/Makefile.am +++ b/docs/Makefile.am @@ -115,7 +115,7 @@ ChangeLog.html.in: ChangeLog.xml ChangeLog.xsl html/index.html: libvirt-api.xml newapi.xsl page.xsl sitemap.html.in -@(if [ -x $(XSLTPROC) ] ; then \ echo "Rebuilding the HTML pages from the XML API" ; \ - $(XSLTPROC) --nonet $(srcdir)/newapi.xsl libvirt-api.xml ; fi ) + $(XSLTPROC) --nonet $(srcdir)/newapi.xsl $(srcdir)/libvirt-api.xml; fi) -@(if [ -x $(XMLLINT) -a -x $(XMLCATALOG) ] ; then \ if $(XMLCATALOG) /etc/xml/catalog "-//W3C//DTD XHTML 1.0 Strict//EN" > /dev/null ; then \ echo "Validating the resulting XHTML pages" ; \ -- 1.7.0.367.g566c3

2 1

[libvirt] libvirt/dnsmasq integration (formatting fixed).
by Simon Kelley 24 Feb '10

24 Feb '10

(apologies for first attempt: I was using a strange mailer.) As the principal maintainer of dnsmasq, I'm seeing increasing reports of problems on systems which run both dnsmasq and libvirt. I'm fairly sure I understand what's going on in these cases, and I have a few proposals for changes in libvir and dnsmasq that should fix things. The problem is that libvirt runs a private instance of dnsmasq: on machines which are also running a "system" dnsmasq daemon, this can cause problems. Some background: dnsmasq can run in two modes. Default mode: dnsmasq binds the wildcard address and does network magic to determine which interface request packets actually come from, so that the results can be sent back with the correct source address. This has the advantage that network interfaces can come and go and change IP address and dnsmasq will keep working. It's possible to restrict dnsmasq to only reply to requests on some interfaces; requests from other interfaces will be read by dnsmasq and then silently dropped. Telling dnsmasq to use an interface which doesn't exist but might in the future will result in a logged warning, but dnsmasq will still start and when the interface comes up it will work. Bind-interfaces mode: This is the traditional way to do UDP servers. At startup dnsmasq enumerates all the extant interfaces and then opens a socket for each one, listening on the interfaces's IP address. Interfaces may be skipped if excluded by the --interface and --except-interface flags, and any interface specified in --interface which doesn't exist at start-up will generate a fatal error. In almost all cases, default mode is better: --bind-interfaces is only there to cope with old platforms which don't support enough socket options to do default mode. The only time when --bind-interfaces works better is when it's desirable to run more than one instance of dnsmasq or have dnsmasq co-exist with another DNS server. This is not possible in default mode, but it does work in bind-interfaces mode, providing than _all_ instances of dnsmasq are in bind-interfaces mode, and that they listen on a disjoint set of interfaces. Therefore, to allow multiple dnsmasq instances libvirt's private dnsmasq instance is started in bind-interfaces mode: that forces one of the dnsmasq instances to do bind-interfaces. Many of the Linux distibution dnsmasq packages have now implemented an /etc/dnsmasq.d directory where configuration fragments can be dropped. Their libvirt packages are putting a file there which contains a bind-interfaces command, so that the "system" dnsmasq is automatically forced into the same mode, and the two can co-exist. This works, sort-of, but there some disadvantages. Installing libvirt drops the configuration change for the system dnsmasq, but the packages frequently don't restart the system daemon, so that things transiently fail until everything has rebooted. Much worse, the system dnsmasq is forced into bind-interfaces mode and then service to transient interfaces (usb, ad-hoc wifi) no longer works, or, because those interfaces are mentioned in the dnsmasq configuration, dnsmasq now fails at start-up when the interfaces don't exist. I don't think there is a solution to this problem wholly within dnsmasq: the nature of the BSD-sockets universe means that there never be two instances of dnsmasq without bind-interfaces, and it's, at best, difficult to get round the limitations of that WRT transient network interfaces. There is, an alternative solution, which would involve changes to both dnsmasq, libvirt and their packaging. Hence this mail. My proposal is to get rid of the necessity for two dnsmasq instances. Libvirt should check for the existance of a "system" dnsmasq and, if the system daemon exists, libvirt should drop the required configuration into /etc/dnsmasq.d and then restart it. If the system daemon is not installed or enabled, libvirt can start a private instance as now. The difficulty with this scheme is that libvirt needs to create some configuration which enables the services it needs on the virtual network without disturbing, or being disturbed by, whatever configuration exists for the system daemon. That's not currently possible, but it can be made possible. I'm assuming that libvirt needs to provide a set of IP address / MAC address mappings, and range of IP addresses on a virtual network. It needs DHCP and DNS service on the virtual network. The dhcp-host IP/MAC mappings are a non-problem: they will be ignored for any other subnet where the IP addresses don't fit, and any other dhcp-hosts in the system configuration will be similarly ignored for DHCP on the virtual network subnet. The dhcp-range is more of a problem. Service to particular networks in dnsmasq is controlled by interface=<interface name"> lines in the configuration. If there are none of these, service is provided to all interfaces. If they exist, service is limited to the interfaces specified. The existence of any dhcp-range line in dnsmasq's configuration enables the DHCP server for any subnet unless explicitly limited to particular interfaces. So a default dnsmasq installation, (with no interface=<interface>) which provides DNS everywhere but DHCP nowhere would be turned into one which provided DHCP on every interface by libvirt adding a dhcp-range. Since there wouldn't be a suitable DHCP range for most subnets, this would only result in logged errors, but it is still not good. Worse, there's no good answer to the question 'should libvirt include interface=virt0"' in the configuration it supplies? If it does, then the "enable DHCP on all interfaces" problem is solved, but a default system configuration with no interface declaration is transformed from one which provides DNS everywhere to one which provides DNS only to the virtual interface. If libvirt doesn't provide "interface=virt0" and the system configuration includes interface declarations, then there will be no DNS or DHCP service to the virtual network. To solve this, I propose to add an optional interface name to the dhcp-range declaration. The semantics of this would be rather odd, but solve the problem perfectly. 1) for DHCP, if any other dhcp-range exists _without_ an interface name, them the interface name is ignored and and things behave as before, otherwise DHCP is only provided to interfaces mentioned in dhcp-range declarations. 2) for DNS, if there are no interface declarations, things work as before. If there are interface declarations, the interfaces mentioned in dhcp-ranges are added to the set which get DNS service. With these rules, it should be possible for libvirt to drop eg dhcp-range=interface:virt0,192.168.0.1,192.168.0.240 into the configuration of the system dnsmasq and get DHCP and DNS service for virt0, irrespective of any other configuration in the system dnsmasq, and doing so shouldn't affect the services supplied elsewhere. The code in libvirt to make this work looks like this: echo dhcp-range=interface:virt0,<ip range> >>/etc/dnsmasq.d/libvirt if <system dnsmasq is not installed or not enabled> dnsmasq --interface=virt0\ --bind-interfaces --conf-file=/etc/dnsmasq.d/libvirt else /etc/init.d/dnsmasq restart (The --bind-interfaces in the private-dnsmasq instance keeps dnsmasq from clashing with other nameservers eg BIND which may be running.) The system dnsmasq package has to ensure that /etc/dnsmasq.d is read for configuration fragments, and the dnsmasq package and the libvirt package will have to co-operate to manage transitions between private and system dnsmasq mode caused by package installation or removal. Does that make sense? It's a long and involved explanation to come to cold. I fear I may have over-simplified what libvirt is doing with dnsmasq, in which case please enlighten me and I'll modify my scheme to take that into account. If this looks good I can easily have the necessary dnsmasq changes in the next release. Cheers, Simon.

1 0

[libvirt] [PATCH] build: avoid non-srcdir "make distcheck" failures (srcdir vs wildcard)
by Jim Meyering 24 Feb '10

24 Feb '10

With this, I'm down to 3 of 43 test failures. >From dd6f73de8afe2bcac80cc0a5b7f60bc83e0e30fc Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Wed, 24 Feb 2010 15:18:51 +0100 Subject: [PATCH] build: avoid non-srcdir "make distcheck" failures (srcdir vs wildcard) * tests/xencapsdata/Makefile.am: Use $(wildcard in $(srcdir)-aware manner * tests/xmconfigdata/Makefile.am: Likewise. * tests/xml2sexprdata/Makefile.am: Likewise. --- tests/xencapsdata/Makefile.am | 5 +++-- tests/xmconfigdata/Makefile.am | 4 ++-- tests/xml2sexprdata/Makefile.am | 4 ++-- 3 files changed, 7 insertions(+), 6 deletions(-) diff --git a/tests/xencapsdata/Makefile.am b/tests/xencapsdata/Makefile.am index 7d77575..ed08bc4 100644 --- a/tests/xencapsdata/Makefile.am +++ b/tests/xencapsdata/Makefile.am @@ -1,2 +1,3 @@ - -EXTRA_DIST = $(wildcard *.xml) $(wildcard *.cpuinfo) $(wildcard *.caps) +EXTRA_DIST = \ + $(patsubst $(srcdir)/%,%,$(wildcard \ + $(srcdir)/*.xml $(srcdir)/*.cpuinfo $(srcdir)/*.caps)) diff --git a/tests/xmconfigdata/Makefile.am b/tests/xmconfigdata/Makefile.am index 05201a0..af0cb09 100644 --- a/tests/xmconfigdata/Makefile.am +++ b/tests/xmconfigdata/Makefile.am @@ -1,2 +1,2 @@ - -EXTRA_DIST = $(wildcard *.xml) $(wildcard *.cfg) +EXTRA_DIST = \ + $(patsubst $(srcdir)/%,%,$(wildcard $(srcdir)/*.xml $(srcdir)/*.cfg)) diff --git a/tests/xml2sexprdata/Makefile.am b/tests/xml2sexprdata/Makefile.am index 43117d5..20028bb 100644 --- a/tests/xml2sexprdata/Makefile.am +++ b/tests/xml2sexprdata/Makefile.am @@ -1,2 +1,2 @@ - -EXTRA_DIST = $(wildcard *.xml) $(wildcard *.sexpr) +EXTRA_DIST = \ + $(patsubst $(srcdir)/%,%,$(wildcard $(srcdir)/*.xml $(srcdir)/*.sexpr)) -- 1.7.0.367.g566c3

1 0

[libvirt] [PATCH] build: avoid non-srcdir "make distcheck" failure (test_conf.sh)
by Jim Meyering 24 Feb '10

24 Feb '10

"make distcheck" from a non-srcdir build dir was failing the "test_conf.sh" test TEST: test_conf.sh Failed to process /t/libvirt/.j/libvirt-0.7.6/tests/confdata/*.conf diff: /t/libvirt/.j/libvirt-0.7.6/tests/confdata/*.out: No such file or directory! 1 FAILED due to the fact that tests/confdata/Makefile.am was specifying its files via this: EXTRA_DIST = $(wildcard *.conf) $(wildcard *.out) That led to a distribution tarball that lacked those files, which are required by the test_conf.sh test. Automake normally warns us not to use $(wildcard ...), but we've suppressed the warning. This is one case in which using $(wildcard...) really should be avoided. In a non-srcdir build, each of those expands to the empty string, since the actual files are not in that directory, but in $(srcdir). My first reflex was to just hard-code the 4 file names: -EXTRA_DIST = $(wildcard *.conf) $(wildcard *.out) +EXTRA_DIST = \ + fc4.conf \ + fc4.out \ + libvirtd.conf \ + libvirtd.out But that'd make for more maintenance overhead if/when we add more files. Here it doesn't really matter, but in the next patch you'll see that it really does, in general. So here's a better way: Account for the fact that the files are available in $(srcdir): EXTRA_DIST = \ $(patsubst $(srcdir)/%,%,$(wildcard $(srcdir)/*.out $(srcdir)/*.conf)) >From ecce58c9432e83386f2a9ee275c873775577152b Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Wed, 24 Feb 2010 15:01:11 +0100 Subject: [PATCH] build: avoid non-srcdir "make distcheck" failure (test_conf.sh) * tests/confdata/Makefile.am (EXTRA_DIST): Apply $(wildcard... to $(srcdir)/..., and then remove the prefix. --- tests/confdata/Makefile.am | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/confdata/Makefile.am b/tests/confdata/Makefile.am index 5e97605..eaaadbc 100644 --- a/tests/confdata/Makefile.am +++ b/tests/confdata/Makefile.am @@ -1,2 +1,2 @@ - -EXTRA_DIST = $(wildcard *.conf) $(wildcard *.out) +EXTRA_DIST = \ + $(patsubst $(srcdir)/%,%,$(wildcard $(srcdir)/*.out $(srcdir)/*.conf)) -- 1.7.0.367.g566c3

1 0

[libvirt] [PATCH 0/4] maint: improve bootstrap
by Eric Blake 24 Feb '10

24 Feb '10

My biggest motivation in writing this patch series is the fact that I already have the complete gnulib.git on my hard drive, and repeatedly doing shallow clones of gnulib into multiple projects gets expensive in both network traffic and disk space. With this series, it is now possible to do: GNULIB_SRCDIR=/path/to/reference/gnulib.git ./autogen.sh so that the .gnulib subdirectory will exploit git's --reference capabilities rather than re-download all the files from scratch; it also results in less disk usage. The only caveat is that you should make sure that your reference gnulib.git is never rewound beyond contents referred to by a submodule, so that you never risk creating delta objects in the client repositories that would break if the reference module eventually prunes objects rendered unused by the rewind (in practice, this is only a problem if you are prone to pointing the .gnulib submodule to a non-published commit, in order to test out pending changes to gnulib.git, and can be avoided by doing your gnulib development in a different directory than in the reference gnulib.git that only tracks upstream). .gitignore | 20 +- autogen.sh | 21 +- bootstrap | 852 +++++++++++++++++++++++++++++++++++++++++++++----- bootstrap.conf | 145 +++++++++ build-aux/.gitignore | 12 +- po/.gitignore | 26 +- po/Makevars | 41 --- 7 files changed, 967 insertions(+), 150 deletions(-) Eric Blake (4): maint: fix quoting in autogen.sh maint: start factoring bootstrap maint: import modern bootstrap maint: sort .gitignore -- Eric Blake eblake(a)redhat.com +1-801-349-2682 Libvirt virtualization library http://libvirt.org

4 19

[libvirt] RFC: Libvirt FAQ
by Cole Robinson 24 Feb '10

24 Feb '10

Hi all, If you spend enough time on #virt and various virt mailing lists, there are quite a few libvirt questions that come up regularly. So, largely in an effort to save myself and everyone from repeating ourselves, I've put together a libvirt FAQ, found here: http://wiki.libvirt.org/page/FAQ Some of the really common questions that I tried to answer (paraphrased): Q) Where does libvirt store the VM XML? A) You shouldn't really care, use virsh edit/dumpxml/define/undefine Q) How do I access the qemu monitor with libvirt? A) You don't, monitor commands should be provided by libvirt APIs, take a look at this complete list... Q) Why is libvirt messing with iptables? A) Providing NAT based network access for your VMs. If you don't want it, here's how to get rid of it. I've also tried to answer some questions that I don't feel have good straightforward answers in a highly visible place, like: - What is libvirt? - What is some of the major functionality provided by libvirt? - What does it mean when someone says 'I use libvirt'? - Difference between qemu:///system and qemu:///session Feel free to make any additions or mention any feedback. At the very least, please take a glance, so when someone inevitably asks one of these questions, you know to point them to the FAQ, rather than type out an answer. If people think this is useful, we can link to it from the main libvirt.org page, and merge the current static FAQ content linked from the sidebar into the wiki page. Thanks, Cole

2 1

[libvirt] [PATCH 00/10] Various documentation updates
by Cole Robinson 24 Feb '10

24 Feb '10

Over the months I've noticed a few XML pieces that aren't documented, so I finally sat down to hammer these out. Patch #1 is actually a bug I noticed in the process of trying to document the affected attribute. Thanks, Cole Cole Robinson (10): storage: conf: Correctly calculate exabyte unit docs: storage: Document capacity/alloc 'unit' docs: storage: <volume><key> is always generated. docs: storage: Fix backingStore <format> docs docs: storage: Document <auth> element docs: storage: Document SCSI pools docs: storage: Document multipath pools docs: domain: Document <description> element docs: network: Document STP and delay attributes docs: network: Document <domain> element docs/formatdomain.html.in | 7 ++++ docs/formatnetwork.html.in | 12 ++++++- docs/formatstorage.html.in | 34 ++++++++++++++++--- docs/storage.html.in | 76 ++++++++++++++++++++++++++++++++++++++++++++ src/conf/storage_conf.c | 4 +- 5 files changed, 124 insertions(+), 9 deletions(-)

4 18

[libvirt] PATCH [0/1] Filesystem pool formatting
by David Allan 24 Feb '10

24 Feb '10

Here's an updated patch based on the feedback from DV and danpb. I haven't added support for any additional filesystems beyond what mke2fs supports. The addition of additional filesystems is easily done and better as a separate patch. Dave

4 7

[libvirt] [PATCH] virsh.c: avoid leak on OOM error path
by Jim Meyering 23 Feb '10

23 Feb '10

No one really cares if we leak memory while dying, but who knows... freeing that buffer may let us go down more gracefully. FYI, the leak is triggered when virFileReadAll succeeds (it allocates "buffer"), yet xmlNewDoc fails: if (virFileReadAll(from, VIRSH_MAX_XML_FILE, &buffer) < 0) return FALSE; doc = xmlNewDoc(NULL); if (doc == NULL) goto no_memory; >From 03c7a44e3d5b2e7c992bebc98fc8c6a7bf63881e Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Fri, 19 Feb 2010 18:03:41 +0100 Subject: [PATCH] virsh.c: avoid leak on OOM error path * tools/virsh.c (cmdCPUBaseline): Also free "buffer" upon OOM. --- tools/virsh.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index dd916f3..8756a7a 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -7139,6 +7139,7 @@ cleanup: return ret; no_memory: + VIR_FREE(buffer); vshError(ctl, "%s", _("Out of memory")); ret = FALSE; return ret; -- 1.7.0.233.g05e1a

3 6

[libvirt] [PATCH] Create raw storage files with O_DSYNC (again)
by Jiri Denemark 22 Feb '10

22 Feb '10

Recently we introduced O_DYSNC flag when creating raw storage files to avoid filling all disk cache with dirty pages. However, the patch got lost when virStorageBackendCreateRaw was reworked using virFileOperation. Let's use O_DYSNC again. Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/storage/storage_backend.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git a/src/storage/storage_backend.c b/src/storage/storage_backend.c index 8b9ed5d..3742493 100644 --- a/src/storage/storage_backend.c +++ b/src/storage/storage_backend.c @@ -356,7 +356,8 @@ virStorageBackendCreateRaw(virConnectPtr conn ATTRIBUTE_UNUSED, goto cleanup; } - if ((createstat = virFileOperation(vol->target.path, O_RDWR | O_CREAT | O_EXCL, + if ((createstat = virFileOperation(vol->target.path, + O_RDWR | O_CREAT | O_EXCL | O_DSYNC, vol->target.perms.mode, vol->target.perms.uid, vol->target.perms.gid, createRawFileOpHook, &hdata, -- 1.7.0

3 3

[libvirt] domain restore race condition
by Laine Stump 22 Feb '10

22 Feb '10

As noted in another message, the problem I was seeing is a race condition in qemudDomainRestore(), not with my modifications to qemudDmainSave(). Here's some discussion about that problem from IRC, with a question at the bottom: > <laine> Does anyone else see a failure of domain restore (immediately > after domain save? I'm very definitely seeing it on my machine with > F12+updates testing and libvirt built from unpatched sources. > <laine> It's very reproduceable - with virsh I do "save domain > filename", then "restore filename" and it pretty much always gives me > a black screen. Then I force shutdown the guest (with virt-manager) > and do "restore filename" again. Tada! It's restored and running! > [...] > <danpb> laine: possible race condition > <danpb> laine: try putting a sleep(10) before the qemuMonitorStartCPUs > in qemuDomainRestore() Dan's suggestion *did* eliminate the failures. > [...] > <danpb> laine: this sounds like the issue with libvirt prematurely > starting execution of the CPUs before QEMU has even started restoring > (or soemthing like that) > <danpb> laine: search the archives for a mail from Charles Duffy on > this subject some time ago > Here's the BZ filed by Charles Duffy https://bugzilla.redhat.com/show_bug.cgi?id=537938 It looks like he's dealing with a race condition earlier in the restore, since his solution was to wait for the migration process to terminate somewhere inside qemudStartVMDaemon(), rather than waiting until qemudStartVMDaemon() was finished (which is what it does now). Since this wait has already been done anyway by the time of Dan's sleep(10) in my test, I don't think Charles' patch would help this situation. So is there something that libvirt can wait on here to ensure proper start? Or is there a problem in qemu? (I'm still running 0.11. I'll also try upgrading to 0.12 and see if there are changes in behavior.)

1 0

[libvirt] PATCH [0/3] Filesystem pool formatting
by David Allan 22 Feb '10

22 Feb '10

I finished these patches this afternoon, and I need to do a bit more testing before I'm comfortable having these patches committed, but I'm sending them along so I can get everybody's feedback overnight. In particular, I'm wondering if my move of the pool unref call is correct. Dave

5 11

[libvirt] Failures in existing qemu domain save/restore.
by Laine Stump 21 Feb '10

21 Feb '10

After running a bunch more tests, I am now noticing failures in domain save/restore *without* my patch (ie sources built straight from git), not 100% in either case (with or without patches), and the same saved domain sometimes fails to restore and sometimes is successful. (Most of the time (but not all), when I try to restore a domain immediately after a save, it fails, but if I then force shutdown of that failed restore and restore again from the same file, it succeeds. (Note that this is all to/from a local disk - it's not related to NFS) Is there a known problem in domain save/restore? Is anyone else seeing any failures? On 02/19/2010 02:30 AM, Laine Stump wrote: > (This patch requires that you first apply the virFileOperation patches > that I sent to the list yesterday.) > > I'm hoping someone else's keen eye can see what I'm failing to see > here. > > The patch following this message makes domain save "work" on > root-squash NFS (both the libvirt header and the QEMU state are saved > to the file), but the files that it saves don't restore properly, even > when simply saving to a local directory w/o forking - the guest screen > comes up black, or reboots immediately, or some other such thing. I've > looked at the data files and the code, and can't see what difference > is causing this - at least the header written by libvirt seems to be > the same, and the part written by qemu starts in the same place and > looks similar. > > If anyone's got an idea, please let me know! I'd like to get this > figured out and a working patch submitted before the day is done. > > -- > libvir-list mailing list > libvir-list(a)redhat.com > https://www.redhat.com/mailman/listinfo/libvir-list > >

1 0

[libvirt] [PATCH] docs: add 3 missing spaces
by Dan Kenigsberg 21 Feb '10

21 Feb '10

From: danken <danken(a)redhat.com> --- docs/architecture.html.in | 2 +- docs/auth.html.in | 2 +- docs/logging.html.in | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/docs/architecture.html.in b/docs/architecture.html.in index 8eb6458..f5cc520 100644 --- a/docs/architecture.html.in +++ b/docs/architecture.html.in @@ -83,7 +83,7 @@ drivers present in driver.h: <li>xen_internal: provide the implementation of the functions possible via direct hypervisor access</li> <li>proxy_internal: provide read-only Xen access via a proxy, the proxy code - is in the <code>proxy/</code>directory.</li> + is in the <code>proxy/</code> directory.</li> <li>xm_internal: provide support for Xen defined but not running domains.</li> <li>qemu_internal: implement the driver functions for QEmu and diff --git a/docs/auth.html.in b/docs/auth.html.in index 3ca40c3..ab6c3e9 100644 --- a/docs/auth.html.in +++ b/docs/auth.html.in @@ -72,7 +72,7 @@ available. The two libvirt daemon actions available are named <code>org.libvirt. for the RO socket, and <code>org.libvirt.unix.manage</code> for the RW socket. -As an example, to allow a user <code>fred</code>full access to the RW socket, +As an example, to allow a user <code>fred</code> full access to the RW socket, while requiring <code>joe</code> to authenticate with the admin password, would require adding the following snippet to <code>PolicyKit.conf</code>. diff --git a/docs/logging.html.in b/docs/logging.html.in index 63a0b05..f4a5878 100644 --- a/docs/logging.html.in +++ b/docs/logging.html.in @@ -96,7 +96,7 @@ <li><code>x:stderr</code> output goes to stderr</li> <li><code>x:syslog:name</code> use syslog for the output and use the given <code>name</code> as the ident</li> - <li><code>x:file:file_path</code>output to a file, with the given + <li><code>x:file:file_path</code> output to a file, with the given filepath</li> </ul> In all cases the x prefix is the minimal level, acting as a filter: -- 1.6.6

2 1