From mburns at redhat.com Fri Sep 7 11:40:13 2012 From: mburns at redhat.com (Mike Burns) Date: Fri, 07 Sep 2012 07:40:13 -0400 Subject: [node-devel] oVirt Node 2.5.2 packages uploaded Message-ID: <1347018013.10838.27.camel@beelzebub.mburnsfire.net> I've posted packages for ovirt-node 2.5.2. Currently this is only the ovirt-node packages and not ovirt-node-iso packages. ovirt-node-iso packages will be built and uploaded when the NFS kernel issue is resolved. Changes in this build are to make systemd and dracut checks and directory structures more stable. Previously, we were checking only Fedora vs. RHEL vs Centos which breaks when a RHEL or Centos version with systemd or a newer dracut comes out. There are no changes other than some packaging and spec file changes. Thanks Mike From Charles_Rose at Dell.com Tue Sep 11 13:42:23 2012 From: Charles_Rose at Dell.com (Charles_Rose at Dell.com) Date: Tue, 11 Sep 2012 19:12:23 +0530 Subject: [node-devel] dracut: ovirt-boot-functions not found during bootup Message-ID: <504F3FBF.604@dell.com> Hello, I was building on RHEL 6.3 with latest upstream ovirt-node. Build of ovirt-node and ISO go fine. During ISO boot, I see messages that /usr/lib/dracut/modules.d/91ovirtnode/ovirt-boot-functions is not found. The install and subsequent boot succeed though. I had to fix scripts/ovirt-functions and scripts/ovirt-init-functions.sh to point to /usr/share/dracut instead of /usr/lib/dracut. Probably a remnant of 9c98bb973e7831232555f290cdd9dadb71eb9322 And recipe/common-minimizer.ks drops /usr/share/dracut, so I had to comment that out to begin with. -- Thanks, Charles Rose Linux Engineering Dell Inc. From Charles_Rose at Dell.com Tue Sep 11 15:47:38 2012 From: Charles_Rose at Dell.com (Charles_Rose at Dell.com) Date: Tue, 11 Sep 2012 21:17:38 +0530 Subject: [node-devel] dracut: ovirt-boot-functions not found during bootup In-Reply-To: <504F3FBF.604@dell.com> References: <504F3FBF.604@dell.com> Message-ID: <504F5D1A.6070403@dell.com> On Tuesday 11 September 2012 07:12 PM, Rose, Charles wrote: > Hello, > > I was building on RHEL 6.3 with latest upstream ovirt-node. Build of ovirt-node and ISO go fine. > During ISO boot, I see messages that /usr/lib/dracut/modules.d/91ovirtnode/ovirt-boot-functions is > not found. The install and subsequent boot succeed though. > > I had to fix scripts/ovirt-functions and scripts/ovirt-init-functions.sh to point to > /usr/share/dracut instead of /usr/lib/dracut. It looks like we need /usr/lib/dracut for F18 and later. How about this? --- ovirt-node.orig/scripts/ovirt-init-functions.sh 2012-09-10 20:19:43.000000000 +0530 +++ ovirt-node/scripts/ovirt-init-functions.sh 2012-09-11 19:40:53.447742348 +0530 @@ -24,7 +24,10 @@ . /etc/init.d/functions . /usr/libexec/ovirt-functions -. /usr/lib/dracut/modules.d/91ovirtnode/ovirt-boot-functions +DRACUTDIR="/usr/lib/dracut" +if [ ! -d "${DRACUTDIR}" ]; then DRACUTDIR="/usr/share/dracut"; fi + +. ${DRACUTDIR}/modules.d/91ovirtnode/ovirt-boot-functions NODE_CONFIG=/etc/sysconfig/node-config > > Probably a remnant of 9c98bb973e7831232555f290cdd9dadb71eb9322 > > And recipe/common-minimizer.ks drops /usr/share/dracut, so I had to comment that out to begin with. and we can keep /usr/share/dracut only for RHEV. --- ovirt-node.orig/recipe/rhevh6-minimizer.ks 2012-09-10 20:19:43.000000000 +0530 +++ ovirt-node/recipe/rhevh6-minimizer.ks 2012-09-11 19:25:11.786783005 +0530 @@ -5,3 +5,4 @@ droprpm procmail droprpm sendmail drop /etc/rc.d/init.d/libvirt-guests +keep /usr/share/dracut > From mburns at redhat.com Tue Sep 11 15:55:33 2012 From: mburns at redhat.com (Mike Burns) Date: Tue, 11 Sep 2012 11:55:33 -0400 Subject: [node-devel] dracut: ovirt-boot-functions not found during bootup In-Reply-To: <504F5D1A.6070403@dell.com> References: <504F3FBF.604@dell.com> <504F5D1A.6070403@dell.com> Message-ID: <1347378933.3690.3.camel@beelzebub.mburnsfire.net> On Tue, 2012-09-11 at 21:17 +0530, Charles_Rose at Dell.com wrote: > On Tuesday 11 September 2012 07:12 PM, Rose, Charles wrote: > > Hello, > > > > I was building on RHEL 6.3 with latest upstream ovirt-node. Build of ovirt-node and ISO go fine. > > During ISO boot, I see messages that /usr/lib/dracut/modules.d/91ovirtnode/ovirt-boot-functions is > > not found. The install and subsequent boot succeed though. > > > > I had to fix scripts/ovirt-functions and scripts/ovirt-init-functions.sh to point to > > /usr/share/dracut instead of /usr/lib/dracut. > > It looks like we need /usr/lib/dracut for F18 and later. How about this? > > --- ovirt-node.orig/scripts/ovirt-init-functions.sh 2012-09-10 20:19:43.000000000 +0530 > +++ ovirt-node/scripts/ovirt-init-functions.sh 2012-09-11 19:40:53.447742348 +0530 > @@ -24,7 +24,10 @@ > . /etc/init.d/functions > . /usr/libexec/ovirt-functions > > -. /usr/lib/dracut/modules.d/91ovirtnode/ovirt-boot-functions > +DRACUTDIR="/usr/lib/dracut" > +if [ ! -d "${DRACUTDIR}" ]; then DRACUTDIR="/usr/share/dracut"; fi > + > +. ${DRACUTDIR}/modules.d/91ovirtnode/ovirt-boot-functions > > NODE_CONFIG=/etc/sysconfig/node-config > > > > > Probably a remnant of 9c98bb973e7831232555f290cdd9dadb71eb9322 > > > > And recipe/common-minimizer.ks drops /usr/share/dracut, so I had to comment that out to begin with. > and we can keep /usr/share/dracut only for RHEV. > > --- ovirt-node.orig/recipe/rhevh6-minimizer.ks 2012-09-10 20:19:43.000000000 +0530 > +++ ovirt-node/recipe/rhevh6-minimizer.ks 2012-09-11 19:25:11.786783005 +0530 > @@ -5,3 +5,4 @@ > droprpm procmail > droprpm sendmail > drop /etc/rc.d/init.d/libvirt-guests > +keep /usr/share/dracut Thanks Charles. We actually were just noticing this today with our builds for el6. The patches were developed and tested on Fedora 17 which had already shifted to /usr/lib/dracut from /usr/share/dracut which is why this was caught previously. I think we may actually need to not drop /usr/*/dracut at all to support plugins, but I need to double check that first. Mike > > > > _______________________________________________ > node-devel mailing list > node-devel at ovirt.org > http://lists.ovirt.org/mailman/listinfo/node-devel From mburns at redhat.com Tue Sep 11 16:33:47 2012 From: mburns at redhat.com (Mike Burns) Date: Tue, 11 Sep 2012 12:33:47 -0400 Subject: [node-devel] dracut: ovirt-boot-functions not found during bootup In-Reply-To: <1347378933.3690.3.camel@beelzebub.mburnsfire.net> References: <504F3FBF.604@dell.com> <504F5D1A.6070403@dell.com> <1347378933.3690.3.camel@beelzebub.mburnsfire.net> Message-ID: <1347381227.3690.4.camel@beelzebub.mburnsfire.net> On Tue, 2012-09-11 at 11:55 -0400, Mike Burns wrote: > On Tue, 2012-09-11 at 21:17 +0530, Charles_Rose at Dell.com wrote: > > On Tuesday 11 September 2012 07:12 PM, Rose, Charles wrote: > > > Hello, > > > > > > I was building on RHEL 6.3 with latest upstream ovirt-node. Build of ovirt-node and ISO go fine. > > > During ISO boot, I see messages that /usr/lib/dracut/modules.d/91ovirtnode/ovirt-boot-functions is > > > not found. The install and subsequent boot succeed though. > > > > > > I had to fix scripts/ovirt-functions and scripts/ovirt-init-functions.sh to point to > > > /usr/share/dracut instead of /usr/lib/dracut. > > > > It looks like we need /usr/lib/dracut for F18 and later. How about this? > > > > --- ovirt-node.orig/scripts/ovirt-init-functions.sh 2012-09-10 20:19:43.000000000 +0530 > > +++ ovirt-node/scripts/ovirt-init-functions.sh 2012-09-11 19:40:53.447742348 +0530 > > @@ -24,7 +24,10 @@ > > . /etc/init.d/functions > > . /usr/libexec/ovirt-functions > > > > -. /usr/lib/dracut/modules.d/91ovirtnode/ovirt-boot-functions > > +DRACUTDIR="/usr/lib/dracut" > > +if [ ! -d "${DRACUTDIR}" ]; then DRACUTDIR="/usr/share/dracut"; fi > > + > > +. ${DRACUTDIR}/modules.d/91ovirtnode/ovirt-boot-functions > > > > NODE_CONFIG=/etc/sysconfig/node-config > > > > > > > > Probably a remnant of 9c98bb973e7831232555f290cdd9dadb71eb9322 > > > > > > And recipe/common-minimizer.ks drops /usr/share/dracut, so I had to comment that out to begin with. > > and we can keep /usr/share/dracut only for RHEV. > > > > --- ovirt-node.orig/recipe/rhevh6-minimizer.ks 2012-09-10 20:19:43.000000000 +0530 > > +++ ovirt-node/recipe/rhevh6-minimizer.ks 2012-09-11 19:25:11.786783005 +0530 > > @@ -5,3 +5,4 @@ > > droprpm procmail > > droprpm sendmail > > drop /etc/rc.d/init.d/libvirt-guests > > +keep /usr/share/dracut > > Thanks Charles. We actually were just noticing this today with our > builds for el6. > > The patches were developed and tested on Fedora 17 which had already > shifted to /usr/lib/dracut from /usr/share/dracut which is why this was > caught previously. > > I think we may actually need to not drop /usr/*/dracut at all to support > plugins, but I need to double check that first. > > Mike Proposed Patch: http://gerrit.ovirt.org/7932 Mike > > > > > > > > _______________________________________________ > > node-devel mailing list > > node-devel at ovirt.org > > http://lists.ovirt.org/mailman/listinfo/node-devel > > > _______________________________________________ > node-devel mailing list > node-devel at ovirt.org > http://lists.ovirt.org/mailman/listinfo/node-devel From mburns at redhat.com Tue Sep 18 11:45:48 2012 From: mburns at redhat.com (Mike Burns) Date: Tue, 18 Sep 2012 07:45:48 -0400 (EDT) Subject: [node-devel] Cancelled: oVirt Node weekly meeting Message-ID: <1658786185.1111960.1347968748972.JavaMail.root@redhat.com> A single instance of the following meeting has been cancelled: Subject: oVirt Node weekly meeting Organizer: "Mike Burns" Location: #ovirt on irc.oftc.net Time: Tuesday, September 18, 2012, 9:00:00 AM - 9:30:00 AM GMT -05:00 US/Canada Eastern Invitees: aliguori at linux.vnet.ibm.com; anthony at codemonkey.ws; node-devel at ovirt.org; whenry at redhat.com *~*~*~*~*~*~*~*~*~* Unable to make this meeting this week and I know at least one other person can't either, so cancelling. Mike -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: meeting.ics Type: text/calendar Size: 2955 bytes Desc: not available URL: From Charles_Rose at Dell.com Wed Sep 19 12:09:46 2012 From: Charles_Rose at Dell.com (Charles_Rose at Dell.com) Date: Wed, 19 Sep 2012 05:09:46 -0700 Subject: [node-devel] edit-node: mksquashfs failures on RHEL Message-ID: <5059B60A.3060006@dell.com> Hello, On RHEL 6.3, I attempted to add a package to ovirt-node-iso, but I get this trace. It looks like mksquashfs osmin.img is being passed '-comp None': # edit-node --install vim --repo ./rhel6.repo ovirt-node-iso-2.5.999-999.1.el6.iso ... Creating Manifests Copying Manifests Editing Complete ... _ovirt-node-iso: 15171/32768 files (0.8% non-contiguous), 105637/105637 blocks ['/sbin/mksquashfs', '/var/tmp/edit-liveos-2Rpc0f/iso/LiveOS/osmin', '/var/tmp/edit-liveos-2Rpc0f/iso/LiveOS/osmin.img', '-comp', None] Traceback (most recent call last): File "/usr/sbin/edit-node", line 1498, in sys.exit(main()) File "/usr/sbin/edit-node", line 1482, in main editor.package(output) File "/usr/lib/python2.6/site-packages/imgcreate/creator.py", line 768, in package self._stage_final_image() File "/usr/lib/python2.6/site-packages/imgcreate/live.py", line 302, in _stage_final_image create_image_minimizer(self.__isodir + "/LiveOS/osmin.img", self._image, self.compress_type) File "/usr/lib/python2.6/site-packages/imgcreate/fs.py", line 618, in create_image_minimizer mksquashfs(cowloop.lofile, path, compress_type) File "/usr/lib/python2.6/site-packages/imgcreate/fs.py", line 55, in mksquashfs ret = subprocess.call(args) File "/usr/lib64/python2.6/subprocess.py", line 478, in call p = Popen(*popenargs, **kwargs) File "/usr/lib64/python2.6/subprocess.py", line 639, in __init__ errread, errwrite) File "/usr/lib64/python2.6/subprocess.py", line 1228, in _execute_child raise child_exception TypeError: execv() arg 2 must contain only strings squashfs-tools-4.0 on RHEL does not support '-comp'. So I bumped it to squashfs-tools-4.2 from upstream, still see the issue. Based on imgcreate/fs.py, setting 'self.compress_type = gzip' in edit-node seems to work. Is mksquashfs supposed to detect the compression type and use that? Any thoughts? -- Thanks, Charles Rose Linux Engineering Dell Inc. From Charles_Rose at Dell.com Wed Sep 19 12:42:36 2012 From: Charles_Rose at Dell.com (Charles_Rose at Dell.com) Date: Wed, 19 Sep 2012 05:42:36 -0700 Subject: [node-devel] edit-node: mksquashfs failures on RHEL In-Reply-To: <5059B60A.3060006@dell.com> References: <5059B60A.3060006@dell.com> Message-ID: <5059BDBC.3010503@dell.com> On Wednesday 19 September 2012 05:39 PM, Rose, Charles wrote: > Hello, > > On RHEL 6.3, I attempted to add a package to ovirt-node-iso, but I get this trace. It looks like > mksquashfs osmin.img is being passed '-comp None': > > # edit-node --install vim --repo ./rhel6.repo ovirt-node-iso-2.5.999-999.1.el6.iso > ... > Creating Manifests > Copying Manifests > Editing Complete > ... > _ovirt-node-iso: 15171/32768 files (0.8% non-contiguous), 105637/105637 blocks > ['/sbin/mksquashfs', '/var/tmp/edit-liveos-2Rpc0f/iso/LiveOS/osmin', > '/var/tmp/edit-liveos-2Rpc0f/iso/LiveOS/osmin.img', '-comp', None] > Traceback (most recent call last): > File "/usr/sbin/edit-node", line 1498, in > sys.exit(main()) > File "/usr/sbin/edit-node", line 1482, in main > editor.package(output) > File "/usr/lib/python2.6/site-packages/imgcreate/creator.py", line 768, in package > self._stage_final_image() > File "/usr/lib/python2.6/site-packages/imgcreate/live.py", line 302, in _stage_final_image > create_image_minimizer(self.__isodir + "/LiveOS/osmin.img", self._image, self.compress_type) > File "/usr/lib/python2.6/site-packages/imgcreate/fs.py", line 618, in create_image_minimizer > mksquashfs(cowloop.lofile, path, compress_type) > File "/usr/lib/python2.6/site-packages/imgcreate/fs.py", line 55, in mksquashfs > ret = subprocess.call(args) > File "/usr/lib64/python2.6/subprocess.py", line 478, in call > p = Popen(*popenargs, **kwargs) > File "/usr/lib64/python2.6/subprocess.py", line 639, in __init__ > errread, errwrite) > File "/usr/lib64/python2.6/subprocess.py", line 1228, in _execute_child > raise child_exception > TypeError: execv() arg 2 must contain only strings > > squashfs-tools-4.0 on RHEL does not support '-comp'. So I bumped it to squashfs-tools-4.2 from > upstream, still see the issue. > > Based on imgcreate/fs.py, setting 'self.compress_type = gzip' in edit-node seems to work. > > Is mksquashfs supposed to detect the compression type and use that? Any thoughts? > Looks like RHEL6 EPEL has a very old version of livecd-tools. Upstream and Fedora have "Enable reading of SquashFS compression type" - 1aabfeae37aa87607a5e083e2775f744773619eb I built livecd-tools-17.14-1 from F17 for RHEL6 and everything works now. -Charles From Charles_Rose at Dell.com Wed Sep 19 13:51:37 2012 From: Charles_Rose at Dell.com (Charles_Rose at Dell.com) Date: Wed, 19 Sep 2012 06:51:37 -0700 Subject: [node-devel] edit-node: mksquashfs failures on RHEL In-Reply-To: <5059BDBC.3010503@dell.com> References: <5059B60A.3060006@dell.com> <5059BDBC.3010503@dell.com> Message-ID: <5059CDE9.708@dell.com> On Wednesday 19 September 2012 06:12 PM, Rose, Charles wrote: > On Wednesday 19 September 2012 05:39 PM, Rose, Charles wrote: >> Hello, >> >> On RHEL 6.3, I attempted to add a package to ovirt-node-iso, but I get this trace. It looks like >> mksquashfs osmin.img is being passed '-comp None': >> >> # edit-node --install vim --repo ./rhel6.repo ovirt-node-iso-2.5.999-999.1.el6.iso >> ... >> Creating Manifests >> Copying Manifests >> Editing Complete >> ... >> _ovirt-node-iso: 15171/32768 files (0.8% non-contiguous), 105637/105637 blocks >> ['/sbin/mksquashfs', '/var/tmp/edit-liveos-2Rpc0f/iso/LiveOS/osmin', >> '/var/tmp/edit-liveos-2Rpc0f/iso/LiveOS/osmin.img', '-comp', None] >> Traceback (most recent call last): >> File "/usr/sbin/edit-node", line 1498, in >> sys.exit(main()) >> File "/usr/sbin/edit-node", line 1482, in main >> editor.package(output) >> File "/usr/lib/python2.6/site-packages/imgcreate/creator.py", line 768, in package >> self._stage_final_image() >> File "/usr/lib/python2.6/site-packages/imgcreate/live.py", line 302, in _stage_final_image >> create_image_minimizer(self.__isodir + "/LiveOS/osmin.img", self._image, self.compress_type) >> File "/usr/lib/python2.6/site-packages/imgcreate/fs.py", line 618, in create_image_minimizer >> mksquashfs(cowloop.lofile, path, compress_type) >> File "/usr/lib/python2.6/site-packages/imgcreate/fs.py", line 55, in mksquashfs >> ret = subprocess.call(args) >> File "/usr/lib64/python2.6/subprocess.py", line 478, in call >> p = Popen(*popenargs, **kwargs) >> File "/usr/lib64/python2.6/subprocess.py", line 639, in __init__ >> errread, errwrite) >> File "/usr/lib64/python2.6/subprocess.py", line 1228, in _execute_child >> raise child_exception >> TypeError: execv() arg 2 must contain only strings >> >> squashfs-tools-4.0 on RHEL does not support '-comp'. So I bumped it to squashfs-tools-4.2 from >> upstream, still see the issue. >> >> Based on imgcreate/fs.py, setting 'self.compress_type = gzip' in edit-node seems to work. >> >> Is mksquashfs supposed to detect the compression type and use that? Any thoughts? >> > > Looks like RHEL6 EPEL has a very old version of livecd-tools. Upstream and Fedora have "Enable > reading of SquashFS compression type" - 1aabfeae37aa87607a5e083e2775f744773619eb > > I built livecd-tools-17.14-1 from F17 for RHEL6 and everything works now. The livecd-tools version in the spec file has a '020-2'. Setting this to '16.0' will ensure we have support to detect compression type for squashfs. diff --git a/ovirt-node.spec.in b/ovirt-node.spec.in index aa01679..d8f1ef9 100644 --- a/ovirt-node.spec.in +++ b/ovirt-node.spec.in @@ -84,7 +84,7 @@ Provides a series of daemons and support utilities for hypervisor distribution. Summary: Recipes for building and running %{product_family} image Group: Applications/System Requires: pykickstart >= 1.54 -Requires: livecd-tools >= 020-2 +Requires: livecd-tools >= 16.0 Obsoletes: ovirt-node-tools <= 2.3.0-999 %define recipe_root %{_datadir}/ovirt-node-recipe @@ -98,7 +98,7 @@ however on a development machine to help to build the image. %package tools Summary: Recipes for building and running %{product_family} image Group: Applications/System -Requires: livecd-tools >= 020-2 +Requires: livecd-tools >= 16.0 %define tools_root %{_datadir}/ovirt-node-tools > > -Charles > From mburns at redhat.com Wed Sep 19 18:33:42 2012 From: mburns at redhat.com (Mike Burns) Date: Wed, 19 Sep 2012 14:33:42 -0400 Subject: [node-devel] edit-node: mksquashfs failures on RHEL In-Reply-To: <5059CDE9.708@dell.com> References: <5059B60A.3060006@dell.com> <5059BDBC.3010503@dell.com> <5059CDE9.708@dell.com> Message-ID: <1348079622.4066.11.camel@beelzebub.mburnsfire.net> On Wed, 2012-09-19 at 06:51 -0700, Charles_Rose at Dell.com wrote: > On Wednesday 19 September 2012 06:12 PM, Rose, Charles wrote: > > On Wednesday 19 September 2012 05:39 PM, Rose, Charles wrote: > >> Hello, > >> > >> On RHEL 6.3, I attempted to add a package to ovirt-node-iso, but I get this trace. It looks like > >> mksquashfs osmin.img is being passed '-comp None': > >> > >> # edit-node --install vim --repo ./rhel6.repo ovirt-node-iso-2.5.999-999.1.el6.iso > >> ... > >> Creating Manifests > >> Copying Manifests > >> Editing Complete > >> ... > >> _ovirt-node-iso: 15171/32768 files (0.8% non-contiguous), 105637/105637 blocks > >> ['/sbin/mksquashfs', '/var/tmp/edit-liveos-2Rpc0f/iso/LiveOS/osmin', > >> '/var/tmp/edit-liveos-2Rpc0f/iso/LiveOS/osmin.img', '-comp', None] > >> Traceback (most recent call last): > >> File "/usr/sbin/edit-node", line 1498, in > >> sys.exit(main()) > >> File "/usr/sbin/edit-node", line 1482, in main > >> editor.package(output) > >> File "/usr/lib/python2.6/site-packages/imgcreate/creator.py", line 768, in package > >> self._stage_final_image() > >> File "/usr/lib/python2.6/site-packages/imgcreate/live.py", line 302, in _stage_final_image > >> create_image_minimizer(self.__isodir + "/LiveOS/osmin.img", self._image, self.compress_type) > >> File "/usr/lib/python2.6/site-packages/imgcreate/fs.py", line 618, in create_image_minimizer > >> mksquashfs(cowloop.lofile, path, compress_type) > >> File "/usr/lib/python2.6/site-packages/imgcreate/fs.py", line 55, in mksquashfs > >> ret = subprocess.call(args) > >> File "/usr/lib64/python2.6/subprocess.py", line 478, in call > >> p = Popen(*popenargs, **kwargs) > >> File "/usr/lib64/python2.6/subprocess.py", line 639, in __init__ > >> errread, errwrite) > >> File "/usr/lib64/python2.6/subprocess.py", line 1228, in _execute_child > >> raise child_exception > >> TypeError: execv() arg 2 must contain only strings > >> > >> squashfs-tools-4.0 on RHEL does not support '-comp'. So I bumped it to squashfs-tools-4.2 from > >> upstream, still see the issue. > >> > >> Based on imgcreate/fs.py, setting 'self.compress_type = gzip' in edit-node seems to work. > >> > >> Is mksquashfs supposed to detect the compression type and use that? Any thoughts? > >> > > > > Looks like RHEL6 EPEL has a very old version of livecd-tools. Upstream and Fedora have "Enable > > reading of SquashFS compression type" - 1aabfeae37aa87607a5e083e2775f744773619eb > > > > I built livecd-tools-17.14-1 from F17 for RHEL6 and everything works now. > > The livecd-tools version in the spec file has a '020-2'. Setting this to '16.0' will ensure we have > support to detect compression type for squashfs. > > diff --git a/ovirt-node.spec.in b/ovirt-node.spec.in > index aa01679..d8f1ef9 100644 > --- a/ovirt-node.spec.in > +++ b/ovirt-node.spec.in > @@ -84,7 +84,7 @@ Provides a series of daemons and support utilities for hypervisor distribution. > Summary: Recipes for building and running %{product_family} image > Group: Applications/System > Requires: pykickstart >= 1.54 > -Requires: livecd-tools >= 020-2 > +Requires: livecd-tools >= 16.0 > Obsoletes: ovirt-node-tools <= 2.3.0-999 > > %define recipe_root %{_datadir}/ovirt-node-recipe > @@ -98,7 +98,7 @@ however on a development machine to help to build the image. > %package tools > Summary: Recipes for building and running %{product_family} image > Group: Applications/System > -Requires: livecd-tools >= 020-2 > +Requires: livecd-tools >= 16.0 > > %define tools_root %{_datadir}/ovirt-node-tools > > > > > -Charles > > Hi Charles, Thanks for the debugging. It looks like we need to pull in the livecd-tools maintainer to see if we can update the EPEL version. Brian, is this something we can get updated (livecd-tools in EPEL6)? Thanks Mike > > _______________________________________________ > node-devel mailing list > node-devel at ovirt.org > http://lists.ovirt.org/mailman/listinfo/node-devel From apevec at gmail.com Thu Sep 20 07:40:19 2012 From: apevec at gmail.com (Alan Pevec) Date: Thu, 20 Sep 2012 09:40:19 +0200 Subject: [node-devel] edit-node: mksquashfs failures on RHEL In-Reply-To: <5059CDE9.708@dell.com> References: <5059B60A.3060006@dell.com> <5059BDBC.3010503@dell.com> <5059CDE9.708@dell.com> Message-ID: On Wed, Sep 19, 2012 at 3:51 PM, wrote: > -Requires: livecd-tools >= 020-2 > +Requires: livecd-tools >= 16.0 This should be >= 1:16.0 livecd-tools bumped Epoch when it changed versioning to match Fedora releases Re. updating livecd-tools in EPEL: not sure what versioning would makes sense for EPEL and what backport compat patches might be needed e.g. we can't just update squashfs-tools in RHEL, we'd need to make livecd-tools detect version and use appropriate options. Cheers, Alan From dfediuck at redhat.com Thu Sep 20 14:02:00 2012 From: dfediuck at redhat.com (Doron Fediuck) Date: Thu, 20 Sep 2012 10:02:00 -0400 (EDT) Subject: [node-devel] [Users] non-operational state as host does not meet clusters' minimu CPU level. In-Reply-To: <007101cd9710$fe89ff60$fb9dfe20$@com> Message-ID: <95404892.2673663.1348149720531.JavaMail.root@redhat.com> wujieke, Can you please run in the host- vdsClient 0 getVdsCaps and post the results? ----- Original Message ----- > From: "wujieke" > To: node-devel at ovirt.org, users at ovirt.org > Cc: "wujieke" > Sent: Thursday, September 20, 2012 12:19:10 PM > Subject: [Users] non-operational state as host does not meet > clusters' minimu CPU level. > Hi, everyone, if it?s not the right mail list, pls point out.. > thanks.. > I am trying to install the ovirt on my Xeon E5-2650 process on Dell > server, which is installed with Fedora 17. While I create a new host > , which actually is the same server as overt-engine is running. > The host is created ,and starting to ?installing?. But it ends with > ?Non operational state?. > Error: > Host CPU type is not compatible with cluster properties, missing CPU > feature: model_sandybridge. > But in my cluster, I select ?sandybridge? CPU, and my Xeon C5 is also > in Sandy bridge family. And also this error lead my server reboot. > Any help is appreciated. > Btw: I have enable INTEL-VT in BIOS. And modprobe KVM and kvm-intel > modules. . attached is screen shot for error. > _______________________________________________ > Users mailing list > Users at ovirt.org > http://lists.ovirt.org/mailman/listinfo/users -------------- next part -------------- An HTML attachment was scrubbed... URL: From iheim at redhat.com Thu Sep 20 14:04:11 2012 From: iheim at redhat.com (Itamar Heim) Date: Thu, 20 Sep 2012 17:04:11 +0300 Subject: [node-devel] [Users] non-operational state as host does not meet clusters' minimu CPU level. In-Reply-To: <007101cd9710$fe89ff60$fb9dfe20$@com> References: <007101cd9710$fe89ff60$fb9dfe20$@com> Message-ID: <505B225B.2040500@redhat.com> On 09/20/2012 12:19 PM, wujieke wrote: > Hi, everyone, if it?s not the right mail list, pls point out.. thanks.. > > I am trying to install the ovirt on my Xeon E5-2650 process on Dell > server, which is installed with Fedora 17. While I create a new host , > which actually is the same server as overt-engine is running. > > The host is created ,and starting to ?installing?. But it ends with ?Non > operational state?. > > Error: > > Host CPU type is not compatible with cluster properties, missing CPU > feature: model_sandybridge. > > But in my cluster, I select ?sandybridge? CPU, and my Xeon C5 is also in > Sandy bridge family. And also this error lead my server reboot. > > Any help is appreciated. > > Btw: I have enable INTEL-VT in BIOS. And modprobe KVM and kvm-intel > modules. . attached is screen shot for error. > > > > _______________________________________________ > Users mailing list > Users at ovirt.org > http://lists.ovirt.org/mailman/listinfo/users > please send output of this command from the host (not engine) vdsClient -s 0 getVdsCaps | grep -i flags From wujieke at qiyi.com Thu Sep 20 09:19:10 2012 From: wujieke at qiyi.com (wujieke) Date: Thu, 20 Sep 2012 17:19:10 +0800 Subject: [node-devel] non-operational state as host does not meet clusters' minimu CPU level. Message-ID: <007101cd9710$fe89ff60$fb9dfe20$@com> Hi, everyone, if it's not the right mail list, pls point out.. thanks.. I am trying to install the ovirt on my Xeon E5-2650 process on Dell server, which is installed with Fedora 17. While I create a new host , which actually is the same server as overt-engine is running. The host is created ,and starting to "installing". But it ends with "Non operational state". Error: Host CPU type is not compatible with cluster properties, missing CPU feature: model_sandybridge. But in my cluster, I select "sandybridge" CPU, and my Xeon C5 is also in Sandy bridge family. And also this error lead my server reboot. Any help is appreciated. Btw: I have enable INTEL-VT in BIOS. And modprobe KVM and kvm-intel modules. . attached is screen shot for error. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ovirt-e.jpg Type: image/jpeg Size: 306589 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ovirt-e1.jpg Type: image/jpeg Size: 206494 bytes Desc: not available URL: From wujieke at qiyi.com Fri Sep 21 00:54:17 2012 From: wujieke at qiyi.com (wujieke) Date: Fri, 21 Sep 2012 08:54:17 +0800 Subject: [node-devel] [Users] non-operational state as host does not meet clusters' minimu CPU level. In-Reply-To: <505B225B.2040500@redhat.com> References: <007101cd9710$fe89ff60$fb9dfe20$@com> <505B225B.2040500@redhat.com> Message-ID: <000601cd9793$a0fd6480$e2f82d80$@com> [root at localhost ~]# vdsClient -s 0 getVdsCaps | grep -i flags cpuFlags = fpu,vme,de,pse,tsc,msr,pae,mce,cx8,apic,sep,mtrr,pge,mca,cmov,pat,pse36,clfl ush,dts,acpi,mmx,fxsr,sse,sse2,ss,ht,tm,pbe,syscall,nx,pdpe1gb,rdtscp,lm,con stant_tsc,arch_perfmon,pebs,bts,rep_good,nopl,xtopology,nonstop_tsc,aperfmpe rf,pni,pclmulqdq,dtes64,monitor,ds_cpl,vmx,smx,est,tm2,ssse3,cx16,xtpr,pdcm, pcid,dca,sse4_1,sse4_2,x2apic,popcnt,tsc_deadline_timer,aes,xsave,avx,lahf_l m,ida,arat,epb,xsaveopt,pln,pts,dts,tpr_shadow,vnmi,flexpriority,ept,vpid,mo del_coreduo,model_Conroe seems only support model_Conroe? -----Original Message----- From: Itamar Heim [mailto:iheim at redhat.com] Sent: Thursday, September 20, 2012 10:04 PM To: wujieke Cc: node-devel at ovirt.org; users at ovirt.org Subject: Re: [Users] non-operational state as host does not meet clusters' minimu CPU level. On 09/20/2012 12:19 PM, wujieke wrote: > Hi, everyone, if it's not the right mail list, pls point out.. thanks.. > > I am trying to install the ovirt on my Xeon E5-2650 process on Dell > server, which is installed with Fedora 17. While I create a new host , > which actually is the same server as overt-engine is running. > > The host is created ,and starting to "installing". But it ends with > "Non operational state". > > Error: > > Host CPU type is not compatible with cluster properties, missing CPU > feature: model_sandybridge. > > But in my cluster, I select "sandybridge" CPU, and my Xeon C5 is also > in Sandy bridge family. And also this error lead my server reboot. > > Any help is appreciated. > > Btw: I have enable INTEL-VT in BIOS. And modprobe KVM and kvm-intel > modules. . attached is screen shot for error. > > > > _______________________________________________ > Users mailing list > Users at ovirt.org > http://lists.ovirt.org/mailman/listinfo/users > please send output of this command from the host (not engine) vdsClient -s 0 getVdsCaps | grep -i flags From iheim at redhat.com Fri Sep 21 04:44:04 2012 From: iheim at redhat.com (Itamar Heim) Date: Fri, 21 Sep 2012 07:44:04 +0300 Subject: [node-devel] [Users] non-operational state as host does not meet clusters' minimu CPU level. In-Reply-To: <000601cd9793$a0fd6480$e2f82d80$@com> References: <007101cd9710$fe89ff60$fb9dfe20$@com> <505B225B.2040500@redhat.com> <000601cd9793$a0fd6480$e2f82d80$@com> Message-ID: <505BF094.6020501@redhat.com> On 09/21/2012 03:54 AM, wujieke wrote: > [root at localhost ~]# vdsClient -s 0 getVdsCaps | grep -i flags > cpuFlags = > fpu,vme,de,pse,tsc,msr,pae,mce,cx8,apic,sep,mtrr,pge,mca,cmov,pat,pse36,clfl > ush,dts,acpi,mmx,fxsr,sse,sse2,ss,ht,tm,pbe,syscall,nx,pdpe1gb,rdtscp,lm,con > stant_tsc,arch_perfmon,pebs,bts,rep_good,nopl,xtopology,nonstop_tsc,aperfmpe > rf,pni,pclmulqdq,dtes64,monitor,ds_cpl,vmx,smx,est,tm2,ssse3,cx16,xtpr,pdcm, > pcid,dca,sse4_1,sse4_2,x2apic,popcnt,tsc_deadline_timer,aes,xsave,avx,lahf_l > m,ida,arat,epb,xsaveopt,pln,pts,dts,tpr_shadow,vnmi,flexpriority,ept,vpid,mo > del_coreduo,model_Conroe > > seems only support model_Conroe? and output of: virsh capabilities? > > -----Original Message----- > From: Itamar Heim [mailto:iheim at redhat.com] > Sent: Thursday, September 20, 2012 10:04 PM > To: wujieke > Cc: node-devel at ovirt.org; users at ovirt.org > Subject: Re: [Users] non-operational state as host does not meet clusters' > minimu CPU level. > > On 09/20/2012 12:19 PM, wujieke wrote: >> Hi, everyone, if it's not the right mail list, pls point out.. thanks.. >> >> I am trying to install the ovirt on my Xeon E5-2650 process on Dell >> server, which is installed with Fedora 17. While I create a new host , >> which actually is the same server as overt-engine is running. >> >> The host is created ,and starting to "installing". But it ends with >> "Non operational state". >> >> Error: >> >> Host CPU type is not compatible with cluster properties, missing CPU >> feature: model_sandybridge. >> >> But in my cluster, I select "sandybridge" CPU, and my Xeon C5 is also >> in Sandy bridge family. And also this error lead my server reboot. >> >> Any help is appreciated. >> >> Btw: I have enable INTEL-VT in BIOS. And modprobe KVM and kvm-intel >> modules. . attached is screen shot for error. >> >> >> >> _______________________________________________ >> Users mailing list >> Users at ovirt.org >> http://lists.ovirt.org/mailman/listinfo/users >> > > please send output of this command from the host (not engine) vdsClient -s 0 > getVdsCaps | grep -i flags > From wujieke at qiyi.com Fri Sep 21 05:01:48 2012 From: wujieke at qiyi.com (wujieke) Date: Fri, 21 Sep 2012 13:01:48 +0800 Subject: [node-devel] [Users] non-operational state as host does not meet clusters' minimu CPU level. In-Reply-To: <505BF094.6020501@redhat.com> References: <007101cd9710$fe89ff60$fb9dfe20$@com> <505B225B.2040500@redhat.com> <000601cd9793$a0fd6480$e2f82d80$@com> <505BF094.6020501@redhat.com> Message-ID: <001601cd97b6$34bd5dc0$9e381940$@com> I follow the wiki page to re-install ovirt with all-in-one version . my local host in ovirt is working now. Thanks a lot. Btw: the cmd " virsh capabilities" complains out : [root at localhost ~]# virsh capabilities Please enter your authentication name: Please enter your password: error: Failed to reconnect to the hypervisor error: no valid connection error: authentication failed: Failed to step SASL negotiation: -1 (SASL(-1): generic failure: All-whitespace username.) any idea? -----Original Message----- From: Itamar Heim [mailto:iheim at redhat.com] Sent: Friday, September 21, 2012 12:44 PM To: wujieke Cc: node-devel at ovirt.org; users at ovirt.org Subject: Re: [Users] non-operational state as host does not meet clusters' minimu CPU level. On 09/21/2012 03:54 AM, wujieke wrote: > [root at localhost ~]# vdsClient -s 0 getVdsCaps | grep -i flags > cpuFlags = > fpu,vme,de,pse,tsc,msr,pae,mce,cx8,apic,sep,mtrr,pge,mca,cmov,pat,pse3 > 6,clfl > ush,dts,acpi,mmx,fxsr,sse,sse2,ss,ht,tm,pbe,syscall,nx,pdpe1gb,rdtscp, > lm,con > stant_tsc,arch_perfmon,pebs,bts,rep_good,nopl,xtopology,nonstop_tsc,ap > erfmpe > rf,pni,pclmulqdq,dtes64,monitor,ds_cpl,vmx,smx,est,tm2,ssse3,cx16,xtpr > ,pdcm, > pcid,dca,sse4_1,sse4_2,x2apic,popcnt,tsc_deadline_timer,aes,xsave,avx, > lahf_l > m,ida,arat,epb,xsaveopt,pln,pts,dts,tpr_shadow,vnmi,flexpriority,ept,v > pid,mo > del_coreduo,model_Conroe > > seems only support model_Conroe? and output of: virsh capabilities? > > -----Original Message----- > From: Itamar Heim [mailto:iheim at redhat.com] > Sent: Thursday, September 20, 2012 10:04 PM > To: wujieke > Cc: node-devel at ovirt.org; users at ovirt.org > Subject: Re: [Users] non-operational state as host does not meet clusters' > minimu CPU level. > > On 09/20/2012 12:19 PM, wujieke wrote: >> Hi, everyone, if it's not the right mail list, pls point out.. thanks.. >> >> I am trying to install the ovirt on my Xeon E5-2650 process on Dell >> server, which is installed with Fedora 17. While I create a new host >> , which actually is the same server as overt-engine is running. >> >> The host is created ,and starting to "installing". But it ends with >> "Non operational state". >> >> Error: >> >> Host CPU type is not compatible with cluster properties, missing CPU >> feature: model_sandybridge. >> >> But in my cluster, I select "sandybridge" CPU, and my Xeon C5 is also >> in Sandy bridge family. And also this error lead my server reboot. >> >> Any help is appreciated. >> >> Btw: I have enable INTEL-VT in BIOS. And modprobe KVM and kvm-intel >> modules. . attached is screen shot for error. >> >> >> >> _______________________________________________ >> Users mailing list >> Users at ovirt.org >> http://lists.ovirt.org/mailman/listinfo/users >> > > please send output of this command from the host (not engine) > vdsClient -s 0 getVdsCaps | grep -i flags > From apevec at gmail.com Fri Sep 21 10:33:16 2012 From: apevec at gmail.com (Alan Pevec) Date: Fri, 21 Sep 2012 12:33:16 +0200 Subject: [node-devel] [Users] non-operational state as host does not meet clusters' minimu CPU level. In-Reply-To: <001601cd97b6$34bd5dc0$9e381940$@com> References: <007101cd9710$fe89ff60$fb9dfe20$@com> <505B225B.2040500@redhat.com> <000601cd9793$a0fd6480$e2f82d80$@com> <505BF094.6020501@redhat.com> <001601cd97b6$34bd5dc0$9e381940$@com> Message-ID: On Fri, Sep 21, 2012 at 7:01 AM, wujieke wrote: > [root at localhost ~]# virsh capabilities > Please enter your authentication name: vdsm protects r/w libvirt socket, to avoid administrators accidentaly mess up w/ VMs under its control. You can use r/o connection virsh -r capabilities Cheers, Alan From mburns at redhat.com Tue Sep 25 13:00:06 2012 From: mburns at redhat.com (Mike Burns) Date: Tue, 25 Sep 2012 09:00:06 -0400 (EDT) Subject: [node-devel] No node meeting today Message-ID: An HTML attachment was scrubbed... URL: From fabiand at redhat.com Wed Sep 26 10:45:22 2012 From: fabiand at redhat.com (Fabian Deutsch) Date: Wed, 26 Sep 2012 12:45:22 +0200 Subject: [node-devel] Thoughts and code for an updated TUI Message-ID: <1348656322.2363.35.camel@fdeutsch-laptop.local> Hey, the current Node TUI (ovirt-config-setup) has a couple of limitations which can not be solved using newt (as it is right now), e.g. resizing. Besides that there are implementation issues (with constraints on newt's functionality) e.g. recursion when drawing the TUI, which are hard or not to solve with newt. Further more some problems repeat itself over different parts of the TUI, e.g. validation of the user input. And looking from a testing perspective it's currently also hard to write unit tests for the "pages" of the TUI. On the other hand the TUI has to provide just enough configurations options to the user to modify a couple of files to be manageable by some entity like oVirt Engine. Seeing these shortcomings and the requirements I started to think about a more modular approach compared to what ovirt-config-setup currently is, to address the technical problems and still provide the TUI the user needs. The idea is to separate the model - which has to be modified (the config) - from the user interface by pushing all the functionality into plugins. A plugin is just exposing the model and a set of widgets (which are bound to the model) which are used to modify the model through the TUI. Events tell the plugin when the model has changed or shall be saved. This separation addresses the need for testability and the general need for plug-able TUI extensions. When starting to write some code for this ideas and trying newt for the TUI, it quickly became clear that the technical limitation can not be solved quickly. Therefor I used python-urwid [1] as the toolkit for the UI. urwid is event based, supports mouse input, utf8 and resizable consoles, features which newt currently doesn't support. urwid is also actively maintained, used in a couple of projects (also in the debian world afaik) and packaged for Fedora. This is not (yet) a proposal for new TUI for Node, but at least I wanted to share my thoughts and point you to my current playground [2]. The intention at last is however to come up with a solution solving our TUI problems. Check out the code and have a look at the example.py plugin (uses most of the functionality). plugins/__init__.py has also lot's of documentation for the interface between plugins and tui. Greetings fabian -- [1] http://excess.org/urwid/ [2] http://fedorapeople.org/cgit/fabiand/public_git/molch.git/ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 836 bytes Desc: This is a digitally signed message part URL: From apevec at gmail.com Wed Sep 26 13:56:21 2012 From: apevec at gmail.com (Alan Pevec) Date: Wed, 26 Sep 2012 15:56:21 +0200 Subject: [node-devel] Thoughts and code for an updated TUI In-Reply-To: <1348656322.2363.35.camel@fdeutsch-laptop.local> References: <1348656322.2363.35.camel@fdeutsch-laptop.local> Message-ID: Hi Fabian, this looks great, I'll have a closer look, but for now just a quick note: > [2] http://fedorapeople.org/cgit/fabiand/public_git/molch.git/ Looks like git urls at bottom of that page are not correct, this one worked: git://fedorapeople.org/~fabiand/public_git/molch.git Cheers, Alan From fabiand at redhat.com Wed Sep 26 14:49:59 2012 From: fabiand at redhat.com (Fabian Deutsch) Date: Wed, 26 Sep 2012 16:49:59 +0200 Subject: [node-devel] Thoughts and code for an updated TUI In-Reply-To: References: <1348656322.2363.35.camel@fdeutsch-laptop.local> Message-ID: <1348670999.2363.36.camel@fdeutsch-laptop.local> Am Mittwoch, den 26.09.2012, 15:56 +0200 schrieb Alan Pevec: > Hi Fabian, > > this looks great, I'll have a closer look, but for now just a quick note: IT's rough around the edges and a wip to evaluate some ways to do stuff. Patches are welcome :) > > [2] http://fedorapeople.org/cgit/fabiand/public_git/molch.git/ > > Looks like git urls at bottom of that page are not correct, this one > worked: git://fedorapeople.org/~fabiand/public_git/molch.git Thanks. I've informed the fedora-admin ppl about this, Greetings fabian -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 836 bytes Desc: This is a digitally signed message part URL: