Re: [libvirt] [jenkins-ci PATCH] guests: Reduce boot delay for FreeBSD
On Fri, Oct 20, 2017 at 10:11:31PM +0200, Andrea Bolognani wrote: > Set it to 1 second instead of the default 10 seconds. This brings > it in line with Linux guests and makes boot faster. > > Signed-off-by: Andrea Bolognani > --- > guests/tasks/base.yml | 7 +-- > 1 file changed, 5 insertions(+), 2 deletions(-) Reviewed-by: Pavel Hrdina signature.asc Description: PGP signature -- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
Re: [libvirt] [PATCH] qemu_domain: Error out eariler when adding an IDE controller to a q35-based vm
s/eariler/earlier/ in the commit summary
On Sun, Oct 22, 2017 at 11:05:08PM +0800, Lin Ma wrote:
Signed-off-by: Lin Ma
---
src/qemu/qemu_domain.c | 10 +-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c
index ece8ee7..120d31c 100644
--- a/src/qemu/qemu_domain.c
+++ b/src/qemu/qemu_domain.c
@@ -3975,10 +3975,18 @@
qemuDomainControllerDefPostParse(virDomainControllerDefPtr cont,
This function is called when parsing every XML. Adding new errors here
could cause already existing machines (even though they aren't working)
disappear from libvirt.
qemuDomainDeviceDefValidate is called only for newly defined domains and
on domain startup, so the error from qemuBuildControllerDevStr can be
moved there.
Jan
}
break;
+case VIR_DOMAIN_CONTROLLER_TYPE_IDE:
+if (qemuDomainIsQ35(def)) {
+virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
+ _("IDE controllers are unsupported for q35 "
+ "machine type"));
+return -1;
+}
+break;
+
case VIR_DOMAIN_CONTROLLER_TYPE_SATA:
signature.asc
Description: Digital signature
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
Re: [libvirt] [jenkins-ci PATCH] guests: Use OpenJDK 8 on Debian 8 too
On Fri, Oct 20, 2017 at 10:12:01PM +0200, Andrea Bolognani wrote: > Signed-off-by: Andrea Bolognani > --- > guests/files/jessie-backports.preferences | 3 +++ > guests/files/jessie-backports.sources | 1 + > guests/tasks/base.yml | 20 > guests/vars/mappings.yml | 1 - > 4 files changed, 24 insertions(+), 1 deletion(-) > create mode 100644 guests/files/jessie-backports.preferences > create mode 100644 guests/files/jessie-backports.sources Reviewed-by: Pavel Hrdina signature.asc Description: PGP signature -- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
Re: [libvirt] [PATCH v2 03/11] conf: Validate user supplied aliases
On Fri, Oct 20, 2017 at 04:52:13PM +0200, Michal Privoznik wrote:
They have to be unique within the domain. As usual, backwards
compatibility takes its price. In this particular situation we
have a device that is represented twice in a domain and so is its
alias.
Signed-off-by: Michal Privoznik
---
src/conf/domain_conf.c | 148 ++-
src/conf/domain_conf.h | 5 ++
src/libvirt_private.syms | 1 +
src/qemu/qemu_driver.c | 3 +
4 files changed, 155 insertions(+), 2 deletions(-)
ACK with the 'xmlopt' and 'parseFlags' arguments removed, see below:
diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c
index 40fcbc7df..ad71e951b 100644
--- a/src/conf/domain_conf.c
+++ b/src/conf/domain_conf.c
@@ -5457,6 +5457,145 @@ virDomainDeviceDefValidateInternal(const
virDomainDeviceDef *dev,
}
+struct virDomainDefValidateAliasesData {
+virHashTablePtr aliases;
+};
+
+
+static int
+virDomainDeviceDefValidateAliasesIterator(virDomainDefPtr def,
+ virDomainDeviceDefPtr dev,
+ virDomainDeviceInfoPtr info,
+ void *opaque)
+{
+struct virDomainDefValidateAliasesData *data = opaque;
+const char *alias = info->alias;
+
+if (!alias)
+return 0;
+
+/* Some crazy backcompat for consoles. */
+if (def->nserials && def->nconsoles &&
+def->consoles[0]->deviceType == VIR_DOMAIN_CHR_DEVICE_TYPE_CONSOLE &&
+def->consoles[0]->targetType == VIR_DOMAIN_CHR_CONSOLE_TARGET_TYPE_SERIAL
&&
+dev->type == VIR_DOMAIN_DEVICE_CHR &&
+virDomainChrEquals(def->serials[0], dev->data.chr))
+return 0;
+
+if (virHashLookup(data->aliases, alias)) {
+virReportError(VIR_ERR_XML_ERROR,
+ _("non unique alias detected: %s"),
+ alias);
+return -1;
+}
+
+if (virHashAddEntry(data->aliases, alias, (void *) 1) < 0) {
+virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
+ _("Unable to construct table of device aliases"));
+return -1;
+}
+
+return 0;
+}
+
+
+/**
+ * virDomainDefValidateAliases:
+ *
+ * Check for uniqueness of device aliases. If @aliases is not
+ * NULL return hash table of all the aliases in it.
+ *
+ * Returns 0 on success,
+ *-1 otherwise (with error reported).
+ */
+static int
+virDomainDefValidateAliases(virDomainXMLOptionPtr xmlopt ATTRIBUTE_UNUSED,
xmlopt is unused here and all the other occurences in this patch.
+const virDomainDef *def,
+virHashTablePtr *aliases,
+unsigned int parseFlags)
+{
+struct virDomainDefValidateAliasesData data;
+int ret = -1;
+
+/* validate configuration only in certain places */
+if (parseFlags & VIR_DOMAIN_DEF_PARSE_SKIP_VALIDATE)
+return 0;
This is checked from the virDomainDefValidateInternal path
before virDomainDefValidateInternal is even called.
And virDomainDeviceValidateAliasImpl always passes 0,
so this condition is redundant and 'parseFlags' can be dropped too.
+
+/* We are not storing copies of aliases. Don't free them. */
+if (!(data.aliases = virHashCreate(10, NULL)))
+goto cleanup;
+
Jan
signature.asc
Description: Digital signature
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
Re: [libvirt] [PATCH v2 00/11] Never ending story of user supplied aliases
On Fri, Oct 20, 2017 at 04:52:10PM +0200, Michal Privoznik wrote: v2 of: https://www.redhat.com/archives/libvir-list/2017-October/msg00790.html diff to v1: - new patch 1/11 to address Pavel's findings - reworked parsing so that the alias is set iff it follows the rules - added some tests - added news.xml entry Michal Privoznik (11): qemu_alias: Be more tolerant if alias don't follow our format conf: Parse user supplied aliases conf: Validate user supplied aliases qemuDomainABIStabilityCheck: Check for memory aliases too qemuxml2argvdata: Drop device aliases qemuhotplugtest: Load active XML conf: Format alias even for inactive XMLs docs: Document user aliases qemu: Parse alias from inactive XMLs tests: Test user set aliases for qemu news: Document user aliases docs/formatdomain.html.in | 23 +++ docs/news.xml | 9 ++ src/conf/domain_conf.c | 171 - src/conf/domain_conf.h | 6 + src/libvirt_private.syms | 1 + src/qemu/qemu_alias.c | 22 +-- src/qemu/qemu_domain.c | 18 ++- src/qemu/qemu_driver.c | 3 + tests/qemuhotplugtest.c| 3 +- .../qemuxml2argv-disk-cdrom-network-ftp.xml| 1 - .../qemuxml2argv-disk-cdrom-network-ftps.xml | 1 - .../qemuxml2argv-disk-cdrom-network-http.xml | 1 - .../qemuxml2argv-disk-cdrom-network-https.xml | 1 - .../qemuxml2argv-disk-cdrom-network-tftp.xml | 1 - .../qemuxml2argv-usb-redir-filter.xml | 1 - .../qemuxml2argv-user-aliases.args | 71 + .../qemuxml2argvdata/qemuxml2argv-user-aliases.xml | 140 + tests/qemuxml2argvtest.c | 5 + .../qemuxml2xmlout-user-aliases.xml| 1 + tests/qemuxml2xmltest.c| 2 + 20 files changed, 449 insertions(+), 32 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-user-aliases.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-user-aliases.xml create mode 12 tests/qemuxml2xmloutdata/qemuxml2xmlout-user-aliases.xml ACK series Jan signature.asc Description: Digital signature -- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
[libvirt] [PATCH] Add output of local time when adding the option of "--pretty" for virsh domtime
Add more human-friendly output of domain's system time.
Signed-off-by: ZhiPeng Lu
Reviewed-by: Jiyun Fan
---
tools/virsh-domain-monitor.c | 11 ---
1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/tools/virsh-domain-monitor.c b/tools/virsh-domain-monitor.c
index 35d04b2..d9520ff 100644
--- a/tools/virsh-domain-monitor.c
+++ b/tools/virsh-domain-monitor.c
@@ -1384,17 +1384,22 @@ cmdDomTime(vshControl *ctl, const vshCmd *cmd)
goto cleanup;
if (pretty) {
-char timestr[100];
+char timestr[100], localtimestr[100];
time_t cur_time = seconds;
-struct tm time_info;
+struct tm time_info, local_time_info;
if (!gmtime_r(&cur_time, &time_info)) {
vshError(ctl, _("Unable to format time"));
goto cleanup;
}
+if (!localtime_r(&cur_time, &local_time_info)) {
+vshError(ctl, _("Unable to format local time"));
+goto cleanup;
+}
strftime(timestr, sizeof(timestr), "%Y-%m-%d %H:%M:%S",
&time_info);
+strftime(localtimestr, sizeof(localtimestr), "%Y-%m-%d %H:%M:%S",
&local_time_info);
-vshPrint(ctl, _("Time: %s"), timestr);
+vshPrint(ctl, _("Time: %s\n Local time: %s"), timestr,
localtimestr);
} else {
vshPrint(ctl, _("Time: %lld"), seconds);
}
--
1.8.3.1
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
[libvirt] [PATCH] qemu_domain: Error out eariler when adding an IDE controller to a q35-based vm
Signed-off-by: Lin Ma
---
src/qemu/qemu_domain.c | 10 +-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c
index ece8ee7..120d31c 100644
--- a/src/qemu/qemu_domain.c
+++ b/src/qemu/qemu_domain.c
@@ -3975,10 +3975,18 @@
qemuDomainControllerDefPostParse(virDomainControllerDefPtr cont,
}
break;
+case VIR_DOMAIN_CONTROLLER_TYPE_IDE:
+if (qemuDomainIsQ35(def)) {
+virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
+ _("IDE controllers are unsupported for q35 "
+ "machine type"));
+return -1;
+}
+break;
+
case VIR_DOMAIN_CONTROLLER_TYPE_SATA:
case VIR_DOMAIN_CONTROLLER_TYPE_VIRTIO_SERIAL:
case VIR_DOMAIN_CONTROLLER_TYPE_CCID:
-case VIR_DOMAIN_CONTROLLER_TYPE_IDE:
case VIR_DOMAIN_CONTROLLER_TYPE_FDC:
case VIR_DOMAIN_CONTROLLER_TYPE_LAST:
break;
--
2.9.2
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
Re: [libvirt] libvirt not wanting to read back its own interface XML
Hi Guido, I didn't mean to accuse Debian of doing a bad job with netcf. On Sun, Oct 22, 2017 at 09:26:31AM +0200, Guido Günther wrote: > On Thu, Oct 19, 2017 at 02:41:31PM +0200, Marc Haber wrote: > > On Thu, Oct 19, 2017 at 01:37:45PM +0200, Michal Privoznik wrote: > > > Aha! the thing is, you're using netcf backend while I'm using the udev > > > one. This error message comes from netcf. It's a netcf's bug. CCing > > > Laine who should know more. > > > > Where can I read up about the different backends available? Is it a > > compile-time setting or can I configure that on my system? > > > > Actually, I'm doing the configuration outside of libvirt anyway (with > > systemd-networkd), I just want the interface to be in virt-manager's > > selection dropdown, so I don't care too much about netcf. > > https://bugs.debian.org/cgi-bin/pkgreport.cgi?archive=0;dist=unstable;ordering=normal;repeatmerged=0;src=netcf > > doesn't look that bad. Not following includes is troublesome (#875762) but > shouldn't be too hard to fix. Can you file bugs for other issues you're > seeing? This would make it simpler to decide whether we disable netcf > support or fix the issues that are there. It is just that I have configuring my networks in special ways for nearly two decades, using the "manual" method of ifupdown way before it was officially introduced into the Debian packages (#88948, sic!), and am therefore used to other packages doing network configuration not playing around pretty well on my systems. And I have stopped using ifupdown entirely with jessie, and have migrated over to systemd-networkd. My expertise with current ifupdown is to limited to be of any use. I would love to see systemd-networkd support for netcf, though. Greetings Marc -- - Marc Haber | "I don't trust Computers. They | Mailadresse im Header Leimen, Germany| lose things."Winona Ryder | Fon: *49 6224 1600402 Nordisch by Nature | How to make an American Quilt | Fax: *49 6224 1600421 -- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
Re: [libvirt] libvirt not wanting to read back its own interface XML
Hi Marc, On Thu, Oct 19, 2017 at 02:41:31PM +0200, Marc Haber wrote: > On Thu, Oct 19, 2017 at 01:37:45PM +0200, Michal Privoznik wrote: > > Aha! the thing is, you're using netcf backend while I'm using the udev > > one. This error message comes from netcf. It's a netcf's bug. CCing > > Laine who should know more. > > Where can I read up about the different backends available? Is it a > compile-time setting or can I configure that on my system? > > Actually, I'm doing the configuration outside of libvirt anyway (with > systemd-networkd), I just want the interface to be in virt-manager's > selection dropdown, so I don't care too much about netcf. https://bugs.debian.org/cgi-bin/pkgreport.cgi?archive=0;dist=unstable;ordering=normal;repeatmerged=0;src=netcf doesn't look that bad. Not following includes is troublesome (#875762) but shouldn't be too hard to fix. Can you file bugs for other issues you're seeing? This would make it simpler to decide whether we disable netcf support or fix the issues that are there. Cheers, -- Guido > > Greetings > Marc > > -- > - > Marc Haber | "I don't trust Computers. They | Mailadresse im Header > Leimen, Germany| lose things."Winona Ryder | Fon: *49 6224 1600402 > Nordisch by Nature | How to make an American Quilt | Fax: *49 6224 1600421 > > -- > libvir-list mailing list > libvir-list@redhat.com > https://www.redhat.com/mailman/listinfo/libvir-list > -- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
