From justin at redfish-group.com Tue Aug 3 02:29:15 2010 From: justin at redfish-group.com (Justin Clacherty) Date: Tue, 3 Aug 2010 12:29:15 +1000 (EST) Subject: [Ovirt-devel] Build fails due to missing ovirt-node-recipe.ks In-Reply-To: <1121038266.127.1280802313700.JavaMail.root@mail.redfish.com.au> Message-ID: <549426462.129.1280802555666.JavaMail.root@mail.redfish.com.au> > I built and installed everything from next repo in fedora 13 minimal > following the build instructions in the website. Cool, now that at least one person has managed it I'll give it another go. > - the instructions to build the server rpm are missing from the > website Did you just do what I had outlined in my last email? ie. - cd to server directory - ./autogen.sh - ./configure - make publish > - I had to make the pxe images manually and add them to cobbler Can you outline how you did this? If we can document exactly how we get it working under F13 then perhaps someone can update the build instructions on the website appropriately. Cheers, Justin. From justin at redfish-group.com Wed Aug 4 00:06:07 2010 From: justin at redfish-group.com (Justin Clacherty) Date: Wed, 4 Aug 2010 10:06:07 +1000 (EST) Subject: [Ovirt-devel] Build fails due to missing ovirt-node-recipe.ks In-Reply-To: <1215813797.138.1280880287609.JavaMail.root@mail.redfish.com.au> Message-ID: <118733368.140.1280880367000.JavaMail.root@mail.redfish.com.au> > - ace -d install ovirt has one error : package ovirt-node-image-pxe > not > found I get an error when FreeIPA is set up. Then a whole heap of notices and warning about a missing dependancy (presumably FreeIPA) debug: //freeipa::bundled/Single_exec[ipa_server_install]: Executing '/usr/sbin/ipa-server-install -r ovirt.priv -p 'supersecretpw' -P 'supersecretpw' -a 'supersecretpw' --hostname management.ovirt.priv -u dirsrv -U' debug: Executing '/usr/sbin/ipa-server-install -r ovirt.priv -p 'supersecretpw' -P 'supersecretpw' -a 'supersecretpw' --hostname management.ovirt.priv -u dirsrv -U' err: //freeipa::bundled/Single_exec[ipa_server_install]/returns: change from notrun to 0 failed: /usr/sbin/ipa-server-install -r ovirt.priv -p 'supersecretpw' -P 'supersecretpw' -a 'supersecretpw' --hostname management.ovirt.priv -u dirsrv -U returned 1 instead of 0 at /usr/share/ace/modules/ovirt/manifests/freeipa.pp:65 From justin at redfish-group.com Wed Aug 4 00:45:58 2010 From: justin at redfish-group.com (Justin Clacherty) Date: Wed, 4 Aug 2010 10:45:58 +1000 (EST) Subject: [Ovirt-devel] Build fails due to missing ovirt-node-recipe.ks In-Reply-To: <118733368.140.1280880367000.JavaMail.root@mail.redfish.com.au> Message-ID: <1488408741.142.1280882758510.JavaMail.root@mail.redfish.com.au> > I get an error when FreeIPA is set up. Then a whole heap of notices > and warning about a missing dependancy (presumably FreeIPA) > > debug: //freeipa::bundled/Single_exec[ipa_server_install]: Executing > '/usr/sbin/ipa-server-install -r ovirt.priv -p 'supersecretpw' -P > 'supersecretpw' -a 'supersecretpw' --hostname management.ovirt.priv -u > dirsrv -U' > debug: Executing '/usr/sbin/ipa-server-install -r ovirt.priv -p > 'supersecretpw' -P 'supersecretpw' -a 'supersecretpw' --hostname > management.ovirt.priv -u dirsrv -U' > err: //freeipa::bundled/Single_exec[ipa_server_install]/returns: > change from notrun to 0 failed: /usr/sbin/ipa-server-install -r > ovirt.priv -p 'supersecretpw' -P 'supersecretpw' -a 'supersecretpw' > --hostname management.ovirt.priv -u dirsrv -U returned 1 instead of 0 > at /usr/share/ace/modules/ovirt/manifests/freeipa.pp:65 I ran ipa-server-install from the command line and the error it's getting is this: Traceback (most recent call last): File "/usr/sbin/ipa-server-install", line 41, in import ipaserver.krbinstance File "/usr/lib/python2.6/site-packages/ipaserver/krbinstance.py", line 45, in from pyasn1.type import univ, namedtype ImportError: No module named type Had a look at the python packages and pyasn1 now has a v1 subdirectory so pyasn1.type seems to now be pyasn1.v1.type Is anyone else seeing this? edited krbinstance.py and replaced all pyasn1.xxx with pyasn1.v1.xxx and reran ipa-server-install, now I get: Unexpected error - see ipaserver-install.log for details: 'module' object has no attribute 'VERSION' Of course there is no ipaserver-install.log. Any help would be appreciated. From justin at redfish-group.com Wed Aug 4 06:49:31 2010 From: justin at redfish-group.com (Justin Clacherty) Date: Wed, 4 Aug 2010 16:49:31 +1000 (EST) Subject: [Ovirt-devel] Build fails due to missing ovirt-node-recipe.ks In-Reply-To: <556218855.167.1280904427174.JavaMail.root@mail.redfish.com.au> Message-ID: <1035483060.169.1280904571358.JavaMail.root@mail.redfish.com.au> ----- "Justin Clacherty" wrote: > > I ran ipa-server-install from the command line and the error it's > getting is this: > > Traceback (most recent call last): > File "/usr/sbin/ipa-server-install", line 41, in > import ipaserver.krbinstance > File "/usr/lib/python2.6/site-packages/ipaserver/krbinstance.py", > line 45, in > from pyasn1.type import univ, namedtype > ImportError: No module named type > > Had a look at the python packages and pyasn1 now has a v1 subdirectory > so pyasn1.type seems to now be pyasn1.v1.type > > Is anyone else seeing this? > > I started from scratch yet again. This time, instead of doing a minimal install of F13 in the VM I did a Software Development install. Doing it this way I do not get this error when running the ace install. I only get the missing pxe file error. I guess this means there's some dependency missing in the install somewhere, presumably python related. All the ovirt services appear to be up and running, however when I browse to https://management.ovirt.priv/ovirt the login doesn't come up, I get "This page isn't redirecting properly". Cheers, Justin. From nicolas.ochem at alcatel-lucent.com Mon Aug 9 12:59:02 2010 From: nicolas.ochem at alcatel-lucent.com (Nicolas Ochem) Date: Mon, 9 Aug 2010 14:59:02 +0200 Subject: [Ovirt-devel] Build fails due to missing ovirt-node-recipe.ks In-Reply-To: <1035483060.169.1280904571358.JavaMail.root@mail.redfish.com.au> References: <1035483060.169.1280904571358.JavaMail.root@mail.redfish.com.au> Message-ID: <4C5FFB96.4030300@alcatel-lucent.com> > All the ovirt services appear to be up and running, however when I browse to https://management.ovirt.priv/ovirt the login doesn't come up, I get "This page isn't redirecting properly". > > This is an issue with mongrel and rails 2.3.5 You should add a monkey patch in rails initializer. See my patch from july 30th. From justin at redfish.com.au Mon Aug 9 14:14:37 2010 From: justin at redfish.com.au (Justin Clacherty) Date: Tue, 10 Aug 2010 00:14:37 +1000 Subject: [Ovirt-devel] Build fails due to missing ovirt-node-recipe.ks In-Reply-To: <4C5FFB96.4030300@alcatel-lucent.com> References: <1035483060.169.1280904571358.JavaMail.root@mail.redfish.com.au> <4C5FFB96.4030300@alcatel-lucent.com> Message-ID: <4C600D4D.50505@redfish.com.au> On 9/08/2010 10:59 PM, Nicolas Ochem wrote: > >> All the ovirt services appear to be up and running, however when I >> browse to https://management.ovirt.priv/ovirt the login doesn't come >> up, I get "This page isn't redirecting properly". >> > This is an issue with mongrel and rails 2.3.5 > You should add a monkey patch in rails initializer. > See my patch from july 30th. Thanks. Thought I'd done that, I'll take another look. From nicolas.ochem at alcatel-lucent.com Fri Aug 13 07:09:18 2010 From: nicolas.ochem at alcatel-lucent.com (Nicolas Ochem) Date: Fri, 13 Aug 2010 09:09:18 +0200 Subject: [Ovirt-devel] Build fails due to missing ovirt-node-recipe.ks In-Reply-To: <691042362.266.1281682621732.JavaMail.root@mail.redfish.com.au> References: <691042362.266.1281682621732.JavaMail.root@mail.redfish.com.au> Message-ID: <4C64EF9E.1030301@alcatel-lucent.com> cc-to ovirt-devel skipping cobbler seems to be the best solution since ovirt-node-image-pxe rpm does not exist anymore. However for this to work you need to add runtime_mode=ovirt in pxelinux.cfg/default in the kernel boot options. In the node, it's useful to check /var/log/ovirt.log for correct operation. > The monkey patch was already there. The /etc/hosts entry for management.ovirt.priv on the management server was pointing to it's guest interface ip address. I changed this to be the ip address of the admin network and the login came up and authenticated me. So now I appear to have a working ovirt. > > Except. > > Booted a node, there's nothing in /var/lib/tftpboot. As Nicolas mentioned, the install no longer adds in the node pxe image, not particularly useful. I ran livecd-iso-to-pxeboot on the iso and then copied all its output to /var/lib/tftpboot. I guess it should be done via cobbler but it wasn't immediately obvious how to do that and I only ever want to boot the node anyway. > > Booted the node again. tftpboot works nicely and the node boots but doesn't appear as a host on the ovirt server. I logged in to the node and there's no network interfaces for some reason. Check dmesg and it says that it finds both Intel Pro/1000 cards. /etc/udev/rules.d/70-persistent-net.rules show both interfaces with the correct MAC addresses. There's no /etc/sysconfig/network-scripts/ifcfg-eth0 or ifcfg-eth1. Do I need to pass some boot options to the kernel? Would importing in to cobbler fix this? > > Cheers, > Justin. > From justin at redfish-group.com Fri Aug 13 07:40:05 2010 From: justin at redfish-group.com (Justin Clacherty) Date: Fri, 13 Aug 2010 17:40:05 +1000 (EST) Subject: [Ovirt-devel] Build fails due to missing ovirt-node-recipe.ks In-Reply-To: <1152708562.268.1281685147934.JavaMail.root@mail.redfish.com.au> Message-ID: <1356803917.270.1281685205253.JavaMail.root@mail.redfish.com.au> ----- "Nicolas Ochem" wrote: > > However for this to work you need to add runtime_mode=ovirt in > pxelinux.cfg/default in the kernel boot options. > Fantastic! Thanks Nicolas, that's done the trick. The node now boots and sets up it's interfaces. The host now appears in the management server user interface and appears to report all the correct hardware. Cheers, Justin. From seungjin at email.arizona.edu Mon Aug 16 18:51:52 2010 From: seungjin at email.arizona.edu (Seung-jin Kim) Date: Mon, 16 Aug 2010 11:51:52 -0700 Subject: [Ovirt-devel] where is ruby on rail Ovirt server code? Message-ID: <4C6988C8.8060506@email.arizona.edu> Hi, Ovirt, I was trying to download and see Ovirt ruby and rail server code, but I see all python code under ovirtserver folder. Where can I see Ovirt management server's web ui codes? From dpierce at redhat.com Mon Aug 16 19:02:12 2010 From: dpierce at redhat.com (Darryl L. Pierce) Date: Mon, 16 Aug 2010 15:02:12 -0400 Subject: [Ovirt-devel] where is ruby on rail Ovirt server code? In-Reply-To: <4C6988C8.8060506@email.arizona.edu> References: <4C6988C8.8060506@email.arizona.edu> Message-ID: <20100816190212.GA3899@mcpierce-laptop.rdu.redhat.com> On Mon, Aug 16, 2010 at 11:51:52AM -0700, Seung-jin Kim wrote: > I was trying to download and see Ovirt ruby and rail server code, > but I see all python code under ovirtserver folder. Where can I see > Ovirt management server's web ui codes? There's no Python code in the server, only Ruby. You can download it here: git://git.fedorahosted.org/git/ovirt/server.git -- Darryl L. Pierce, Sr. Software Engineer @ Red Hat, Inc. Delivering value year after year. Red Hat ranks #1 in value among software vendors. http://www.redhat.com/promo/vendor/ -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 198 bytes Desc: not available URL: From seungjin at email.arizona.edu Mon Aug 16 19:05:02 2010 From: seungjin at email.arizona.edu (Seung-jin Kim) Date: Mon, 16 Aug 2010 12:05:02 -0700 Subject: [Ovirt-devel] where is ruby on rail Ovirt server code? In-Reply-To: <20100816190212.GA3899@mcpierce-laptop.rdu.redhat.com> References: <4C6988C8.8060506@email.arizona.edu> <20100816190212.GA3899@mcpierce-laptop.rdu.redhat.com> Message-ID: <4C698BDE.5010803@email.arizona.edu> Thanks.. My fault.. I downloaded node source.. On 8/16/10 12:02 PM, Darryl L. Pierce wrote: > On Mon, Aug 16, 2010 at 11:51:52AM -0700, Seung-jin Kim wrote: >> I was trying to download and see Ovirt ruby and rail server code, >> but I see all python code under ovirtserver folder. Where can I see >> Ovirt management server's web ui codes? > There's no Python code in the server, only Ruby. You can download it > here: > > git://git.fedorahosted.org/git/ovirt/server.git > From justin at redfish.com.au Tue Aug 17 03:35:43 2010 From: justin at redfish.com.au (Justin Clacherty) Date: Tue, 17 Aug 2010 13:35:43 +1000 Subject: [Ovirt-devel] node becomes "unavailable" Message-ID: <4C6A038F.8080506@redfish.com.au> After running for a while the node becomes "unavailable" in the server UI. All VMs running on that node also become unavailable. The node is still running fine as are all the VMs, they're just no longer manageable. I looked on the node and everything appeared to be running fine. Looked on the server and ovirt-taskomatic was stopped (this seems to happen quite a bit). Restarted it but that didn't help. Restarting Matahari on the node sends information to the server but the node does not become available. The only way I've been able to get it back is to shutdown all the VMs and reboot the node and management server. Is anyone else seeing this happen? What else can I look at when it happens again? Cheers, Justin. From nicolas.ochem at alcatel-lucent.com Tue Aug 17 07:37:07 2010 From: nicolas.ochem at alcatel-lucent.com (Nicolas Ochem) Date: Tue, 17 Aug 2010 09:37:07 +0200 Subject: [Ovirt-devel] node becomes "unavailable" In-Reply-To: <4C6A038F.8080506@redfish.com.au> References: <4C6A038F.8080506@redfish.com.au> Message-ID: <4C6A3C23.4090304@alcatel-lucent.com> You can look at /var/log/ovirt-server/db-omatic.log . Probably the node times out because it does not answer to heartbeat anymore. To get more detail you can run the db-omatic script in no-daemon mode (/usr/share/ovirt-server/db-omatic/db_omatic.rb -n) I see that very often on fedora 13, a bit less on fedora 12. This is because the ruby aqmp bindings get stuck when they have to handle too many threads. There's no fix for this yet, but a workaround : whenever that happens, restart everything in the node and server with this script : http://ovirt.pastebin.com/JjNpEDak http://ovirt.pastebin.com/tPAPJBpB You can put that script in a cron job. On 08/17/2010 05:35 AM, Justin Clacherty wrote: > After running for a while the node becomes "unavailable" in the server > UI. All VMs running on that node also become unavailable. The node is > still running fine as are all the VMs, they're just no longer manageable. > > I looked on the node and everything appeared to be running fine. Looked > on the server and ovirt-taskomatic was stopped (this seems to happen > quite a bit). Restarted it but that didn't help. Restarting Matahari > on the node sends information to the server but the node does not become > available. The only way I've been able to get it back is to shutdown > all the VMs and reboot the node and management server. Is anyone else > seeing this happen? What else can I look at when it happens again? > > Cheers, > Justin. > > _______________________________________________ > Ovirt-devel mailing list > Ovirt-devel at redhat.com > https://www.redhat.com/mailman/listinfo/ovirt-devel > From aclement at linagora.com Tue Aug 17 09:43:16 2010 From: aclement at linagora.com (Arthur =?iso-8859-1?q?Cl=E9ment?=) Date: Tue, 17 Aug 2010 11:43:16 +0200 Subject: [Ovirt-devel] [PATCH] Display available space in lvm pool In-Reply-To: <1276505194-7617-1-git-send-email-mloiseleur@linagora.com> References: <1276505194-7617-1-git-send-email-mloiseleur@linagora.com> Message-ID: <201008171143.17009.aclement@linagora.com> ACK and pushed On lundi 14 juin 2010 10:46:34 Loiseleur Michel wrote: > --- > src/app/models/iscsi_storage_volume.rb | 7 ++++++- > 1 files changed, 6 insertions(+), 1 deletions(-) > > diff --git a/src/app/models/iscsi_storage_volume.rb > b/src/app/models/iscsi_storage_volume.rb index fe2cbf5..0a0797d 100644 > --- a/src/app/models/iscsi_storage_volume.rb > +++ b/src/app/models/iscsi_storage_volume.rb > @@ -19,7 +19,12 @@ > > class IscsiStorageVolume < StorageVolume > def label_components > - "#{storage_pool[:target]}:#{lun}" > + available = self.size_in_gb - > self.lvm_storage_pool.storage_volumes.total_size_in_gb if > self.lvm_storage_pool + if self.lvm_storage_pool > + "#{storage_pool[:target]}:#{lun} #{available}GB available" > + else > + "#{storage_pool[:target]}:#{lun}" > + end > end > > def volume_name -- Arthur CLEMENT Linagora Paris From aclement at linagora.com Wed Aug 18 09:51:14 2010 From: aclement at linagora.com (Arthur Clement) Date: Wed, 18 Aug 2010 11:51:14 +0200 Subject: [Ovirt-devel] [PATCH] Don't show vlan interfaces in NIC host management Message-ID: <1282125074-26338-1-git-send-email-aclement@linagora.com> Signed-off-by: Arthur Clement --- src/host-browser/host-register.rb | 16 ++++++++++++---- 1 files changed, 12 insertions(+), 4 deletions(-) diff --git a/src/host-browser/host-register.rb b/src/host-browser/host-register.rb index 693a6f2..5ff6a90 100755 --- a/src/host-browser/host-register.rb +++ b/src/host-browser/host-register.rb @@ -393,13 +393,18 @@ class HostRegister < Qmf::ConsoleHandler # if we have a match, then update the database and remove # the received data to avoid creating a dupe later @logger.info "Searching for existing record for: #{detail.macaddr.upcase} in host #{host_qmf.hostname}" - if detail.macaddr.upcase == nic.mac + # if detail.interface ~ /eth.*\..*/ || detail.interface ~ /vnet.*/ + if detail.interface =~ /eth\d+\.\d+/ || detail.interface =~ /vnet\d+/ + @logger.info "Don't manage #{detail.interface.inspect}" + else + if detail.macaddr.upcase == nic.mac @logger.info "Updating details for: #{detail.interface} [#{nic.mac}]}" nic.bandwidth = detail.bandwidth nic.interface_name = detail.interface nic.save! found = true nic_info.delete(detail) + end end end @@ -425,9 +430,12 @@ class HostRegister < Qmf::ConsoleHandler 'interface_name' => nic.interface, 'usage_type' => 1) - host_db.nics << detail - - @logger.info "Added NIC #{nic.interface} with MAC #{nic.macaddr} to host #{host_qmf.hostname}" + if detail.interface =~ /eth\d+\.\d+/ || detail.interface =~ /vnet\d+/ + @logger.info "Don't Add #{detail.interface.inspect}" + else + host_db.nics << detail + @logger.info "Added NIC #{nic.interface} with MAC #{nic.macaddr} to host #{host_qmf.hostname}" + end end end -- 1.7.2.1 From aclement at linagora.com Wed Aug 18 09:53:23 2010 From: aclement at linagora.com (Arthur =?iso-8859-15?q?Cl=E9ment?=) Date: Wed, 18 Aug 2010 11:53:23 +0200 Subject: [Ovirt-devel] [PATCH] Don't show vlan interfaces in NIC host management In-Reply-To: <1282125074-26338-1-git-send-email-aclement@linagora.com> References: <1282125074-26338-1-git-send-email-aclement@linagora.com> Message-ID: <201008181153.23831.aclement@linagora.com> This is the third patch for vlan support To add vlan : [PATCH] Add support for tagged VLAN 21/07/2010 [PATCH] Add vlan support in taskomatic Patch from Patrick Guiran 21/07/2010 and this one On mercredi 18 ao?t 2010 11:51:14 you wrote: > Signed-off-by: Arthur Clement > --- > src/host-browser/host-register.rb | 16 ++++++++++++---- > 1 files changed, 12 insertions(+), 4 deletions(-) > > diff --git a/src/host-browser/host-register.rb > b/src/host-browser/host-register.rb index 693a6f2..5ff6a90 100755 > --- a/src/host-browser/host-register.rb > +++ b/src/host-browser/host-register.rb > @@ -393,13 +393,18 @@ class HostRegister < Qmf::ConsoleHandler > # if we have a match, then update the database and remove > # the received data to avoid creating a dupe later > @logger.info "Searching for existing record for: > #{detail.macaddr.upcase} in host #{host_qmf.hostname}" - if > detail.macaddr.upcase == nic.mac > + # if detail.interface ~ /eth.*\..*/ || detail.interface ~ /vnet.*/ > + if detail.interface =~ /eth\d+\.\d+/ || detail.interface =~ /vnet\d+/ > + @logger.info "Don't manage #{detail.interface.inspect}" > + else > + if detail.macaddr.upcase == nic.mac > @logger.info "Updating details for: > #{detail.interface} [#{nic.mac}]}" nic.bandwidth = detail.bandwidth > nic.interface_name = detail.interface > nic.save! > found = true > nic_info.delete(detail) > + end > end > end > > @@ -425,9 +430,12 @@ class HostRegister < Qmf::ConsoleHandler > 'interface_name' => nic.interface, > 'usage_type' => 1) > > - host_db.nics << detail > - > - @logger.info "Added NIC #{nic.interface} with MAC > #{nic.macaddr} to host #{host_qmf.hostname}" + if detail.interface =~ > /eth\d+\.\d+/ || detail.interface =~ /vnet\d+/ + > @logger.info "Don't Add #{detail.interface.inspect}" + else > + host_db.nics << detail > + @logger.info "Added NIC #{nic.interface} with MAC > #{nic.macaddr} to host #{host_qmf.hostname}" + end > end > end -- Arthur CLEMENT Linagora Paris From aclement at linagora.com Wed Aug 18 12:19:11 2010 From: aclement at linagora.com (Arthur =?iso-8859-15?q?Cl=E9ment?=) Date: Wed, 18 Aug 2010 14:19:11 +0200 Subject: [Ovirt-devel] [PATCH] Remove obsolete packages (nash and rhpl) In-Reply-To: <1279637839-29339-1-git-send-email-aclement@linagora.com> References: <1279637839-29339-1-git-send-email-aclement@linagora.com> Message-ID: <201008181419.16719.aclement@linagora.com> pushed On mardi 20 juillet 2010 16:57:19 Arthur Clement wrote: > Signed-off-by: Arthur Clement > --- > recipe/common-pkgs.ks | 2 -- > 1 files changed, 0 insertions(+), 2 deletions(-) > > diff --git a/recipe/common-pkgs.ks b/recipe/common-pkgs.ks > index 7578335..f5785ce 100644 > --- a/recipe/common-pkgs.ks > +++ b/recipe/common-pkgs.ks > @@ -49,7 +49,6 @@ newt-python > -newt > -kudzu > -libselinux-python > --rhpl > -kbd > -usermode > -fedora-logos > @@ -63,7 +62,6 @@ generic-logos > -less > -which > -parted > --nash > -tar > -libuser > -mtools -- Arthur CLEMENT Linagora Paris From scourtois at linagora.com Wed Aug 18 13:21:27 2010 From: scourtois at linagora.com (Simon COURTOIS) Date: Wed, 18 Aug 2010 15:21:27 +0200 Subject: [Ovirt-devel] [PATCH] Setting the migration timeout to 3 minutes Message-ID: <1282137687-92857-1-git-send-email-scourtois@linagora.com> --- src/task-omatic/taskomatic.rb | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/task-omatic/taskomatic.rb b/src/task-omatic/taskomatic.rb index 7de725d..0314e6a 100755 --- a/src/task-omatic/taskomatic.rb +++ b/src/task-omatic/taskomatic.rb @@ -548,7 +548,7 @@ class TaskOmatic dest_uri = "qemu+tcp://" + dest_node.hostname + "/system" @logger.debug("Migrating from #{src_uri} to #{dest_uri}") - result = vm.migrate(dest_uri, Libvirt::Domain::MIGRATE_LIVE, '', '', 0, :timeout => 60 * 50) + result = vm.migrate(dest_uri, Libvirt::Domain::MIGRATE_LIVE, '', '', 0, :timeout => 60 * 300) @logger.error "Error migrating VM: #{result.text}" unless result.status == 0 # undefine can fail, for instance, if we live migrated from A -> B, and -- 1.7.2.1 From aclement at linagora.com Wed Aug 18 13:31:38 2010 From: aclement at linagora.com (Arthur =?iso-8859-1?q?Cl=E9ment?=) Date: Wed, 18 Aug 2010 15:31:38 +0200 Subject: [Ovirt-devel] [PATCH] Setting the migration timeout to 3 minutes In-Reply-To: <1282137687-92857-1-git-send-email-scourtois@linagora.com> References: <1282137687-92857-1-git-send-email-scourtois@linagora.com> Message-ID: <201008181531.38737.aclement@linagora.com> ACK and Pushed On mercredi 18 ao?t 2010 15:21:27 Simon COURTOIS wrote: > --- > src/task-omatic/taskomatic.rb | 2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) > > diff --git a/src/task-omatic/taskomatic.rb b/src/task-omatic/taskomatic.rb > index 7de725d..0314e6a 100755 > --- a/src/task-omatic/taskomatic.rb > +++ b/src/task-omatic/taskomatic.rb > @@ -548,7 +548,7 @@ class TaskOmatic > dest_uri = "qemu+tcp://" + dest_node.hostname + "/system" > @logger.debug("Migrating from #{src_uri} to #{dest_uri}") > > - result = vm.migrate(dest_uri, Libvirt::Domain::MIGRATE_LIVE, '', '', > 0, :timeout => 60 * 50) + result = vm.migrate(dest_uri, > Libvirt::Domain::MIGRATE_LIVE, '', '', 0, :timeout => 60 * 300) > @logger.error "Error migrating VM: #{result.text}" unless result.status == > 0 > > # undefine can fail, for instance, if we live migrated from A -> B, > and -- Arthur CLEMENT Linagora Paris From aclement at linagora.com Wed Aug 18 13:37:22 2010 From: aclement at linagora.com (Arthur =?iso-8859-15?q?Cl=E9ment?=) Date: Wed, 18 Aug 2010 15:37:22 +0200 Subject: [Ovirt-devel] [PATCH] Get timestamp at check_heartbeat to avoid time lag In-Reply-To: <1279704434-17237-1-git-send-email-aclement@linagora.com> References: <1279704434-17237-1-git-send-email-aclement@linagora.com> Message-ID: <201008181537.22387.aclement@linagora.com> pushed (Wed Jul 21) On mercredi 21 juillet 2010 11:27:14 Arthur Clement wrote: > Avoid time lag and disconnection after several checks > > Signed-off-by: Arthur Clement > --- > src/host-browser/host-register.rb | 7 ++++--- > 1 files changed, 4 insertions(+), 3 deletions(-) > > diff --git a/src/host-browser/host-register.rb > b/src/host-browser/host-register.rb index e49f08e..693a6f2 100755 > --- a/src/host-browser/host-register.rb > +++ b/src/host-browser/host-register.rb > @@ -178,6 +178,7 @@ class HostRegister < Qmf::ConsoleHandler > def agent_disconnected(agent) > synchronize do > debugputs "Marking objects for agent #{agent.key} inactive" > + @logger.info "Marking objects for agent #{agent.key} inactive" > @cached_hosts.keys.each do |objkey| > if @cached_hosts[objkey][:agent_key] == agent.key > cached_host = @cached_hosts[objkey] > @@ -435,10 +436,10 @@ class HostRegister < Qmf::ConsoleHandler > while true > sleep(5) > synchronize do > - # Get seconds from the epoch > - t = Time.new.to_i > > @heartbeats.keys.each do | key | > + # Get seconds from the epoch > + t = Time.new.to_i > agent, timestamp = @heartbeats[key] > > # Heartbeats from qpid are in microseconds, we > just need seconds.. @@ -447,7 +448,7 @@ class HostRegister < > Qmf::ConsoleHandler > > if delta > 30 > # No heartbeat for 30 seconds.. deal with > dead/disconnected agent. - debugputs "Agent > #{key} timed out!" > + @logger.info "Agent #{key} timed out! Check > NTP || with delta #{delta} = t #{t} - s #{s}" @heartbeats.delete(key) > > agent_disconnected(agent) -- Arthur CLEMENT Linagora Paris From aclement at linagora.com Wed Aug 18 13:37:44 2010 From: aclement at linagora.com (Arthur =?iso-8859-15?q?Cl=E9ment?=) Date: Wed, 18 Aug 2010 15:37:44 +0200 Subject: [Ovirt-devel] [PATCH] Add support for tagged VLAN In-Reply-To: <1279706091-20184-1-git-send-email-aclement@linagora.com> References: <1279706091-20184-1-git-send-email-aclement@linagora.com> Message-ID: <201008181537.44821.aclement@linagora.com> pushed (Wed Jul 21) On mercredi 21 juillet 2010 11:54:51 Arthur Clement wrote: > VLAN are supported with this patch from Patrick Guiran > Setup instructions coming soon > (https://fedorahosted.org/ovirt/wiki/VLANSetup) > > Signed-off-by: Arthur Clement > --- > src/lib/managed_node_configuration.rb | 20 ++++++++++++++++++++ > 1 files changed, 20 insertions(+), 0 deletions(-) > > diff --git a/src/lib/managed_node_configuration.rb > b/src/lib/managed_node_configuration.rb index fac653b..47360c0 100644 > --- a/src/lib/managed_node_configuration.rb > +++ b/src/lib/managed_node_configuration.rb > @@ -95,6 +95,21 @@ class ManagedNodeConfiguration > nic.netmask, nic.broadcast, > nic.gateway) > add_nic(result, nic.mac, iface_name) > + > + # process the vlan tagging > + > + nic.network.usages.map do |usage| > + usage.networks.map do |net| > + if net.type == "Vlan" > + eth_vlan_name = "#{nic.interface_name}.#{net.number}" > + add_bridge(result, 'none', eth_vlan_name, > + nic.boot_protocol, nic.ip_address, > + nic.netmask, nic.broadcast, > + nic.gateway) > + add_vlan(result, eth_vlan_name) > + end > + end # end of : usage.networks.map do |net| > + end # end of : nic.network.usages.map do |usage| > end > end > end > @@ -121,4 +136,9 @@ class ManagedNodeConfiguration > def self.add_slave(result, mac, iface_name, master) > result.puts > "ifcfg=#{mac}|#{iface_name}|MASTER=#{master}|SLAVE=yes|ONBOOT=yes" end > + > + def self.add_vlan(result, iface_name) > + result.puts > "ifcfg=none|#{iface_name}|BRIDGE=br#{iface_name}|ONBOOT=yes|VLAN=yes" + > end > + > end -- Arthur CLEMENT Linagora Paris From aclement at linagora.com Wed Aug 18 13:38:14 2010 From: aclement at linagora.com (Arthur =?iso-8859-15?q?Cl=E9ment?=) Date: Wed, 18 Aug 2010 15:38:14 +0200 Subject: [Ovirt-devel] [PATCH] Add vlan support in taskomatic Patch from Patrick Guiran In-Reply-To: <1279726045-21361-1-git-send-email-aclement@linagora.com> References: <1279726045-21361-1-git-send-email-aclement@linagora.com> Message-ID: <201008181538.14651.aclement@linagora.com> pushed (Wed Jul 21) On mercredi 21 juillet 2010 17:27:25 Arthur Clement wrote: > Signed-off-by: Arthur Clement > --- > src/task-omatic/taskomatic.rb | 20 ++++++++++++++++++-- > 1 files changed, 18 insertions(+), 2 deletions(-) > > diff --git a/src/task-omatic/taskomatic.rb b/src/task-omatic/taskomatic.rb > index 7de725d..1fb8132 100755 > --- a/src/task-omatic/taskomatic.rb > +++ b/src/task-omatic/taskomatic.rb > @@ -376,20 +376,36 @@ class TaskOmatic > # loop through each nic/network assigned to vm, > # finding necessary host devices to bridge > net_interfaces = [] > + device_interface_name = nil > db_vm.nics.each { |nic| > device = net_device = nil > if nic.network.class == PhysicalNetwork > device = Nic.find(:first, > > :conditions => ["host_id = ? AND network_id = ?", > > db_host.id, nic.network_id ]) > + device_interface_name = device.interface_name > else > device = Bonding.find(:first, > > :conditions => ["host_id = ? AND vlan_id = ?", > > db_host.id, nic.network_id ]) > + # here we should check if in usage group associate to this lan > + # we can find a lan that match ... > + if device > + device_interface_name = device.interface_name > + else > + # tag du vlan = nic.netwoerk.number > + usage = nic.network.usages.first > + db_host.nics.map do |n| > + if n.network and n.network.usages.first == usage > + device_interface_name = > "#{n.interface_name}.#{nic.network.number}" + break > + end > + end > + end > end > > - unless device.nil? > - net_device = "br" + device.interface_name > + if device_interface_name > + net_device = "br" + device_interface_name > else > net_device = "breth0" # FIXME remove this default at some point > end -- Arthur CLEMENT Linagora Paris From aclement at linagora.com Wed Aug 18 13:39:03 2010 From: aclement at linagora.com (Arthur =?iso-8859-15?q?Cl=E9ment?=) Date: Wed, 18 Aug 2010 15:39:03 +0200 Subject: [Ovirt-devel] [PATCH] Don't show vlan interfaces in NIC host management In-Reply-To: <1282125074-26338-1-git-send-email-aclement@linagora.com> References: <1282125074-26338-1-git-send-email-aclement@linagora.com> Message-ID: <201008181539.03457.aclement@linagora.com> pushed On mercredi 18 ao?t 2010 11:51:14 Arthur Clement wrote: > Signed-off-by: Arthur Clement > --- > src/host-browser/host-register.rb | 16 ++++++++++++---- > 1 files changed, 12 insertions(+), 4 deletions(-) > > diff --git a/src/host-browser/host-register.rb > b/src/host-browser/host-register.rb index 693a6f2..5ff6a90 100755 > --- a/src/host-browser/host-register.rb > +++ b/src/host-browser/host-register.rb > @@ -393,13 +393,18 @@ class HostRegister < Qmf::ConsoleHandler > # if we have a match, then update the database and remove > # the received data to avoid creating a dupe later > @logger.info "Searching for existing record for: > #{detail.macaddr.upcase} in host #{host_qmf.hostname}" - if > detail.macaddr.upcase == nic.mac > + # if detail.interface ~ /eth.*\..*/ || detail.interface ~ /vnet.*/ > + if detail.interface =~ /eth\d+\.\d+/ || detail.interface =~ /vnet\d+/ > + @logger.info "Don't manage #{detail.interface.inspect}" > + else > + if detail.macaddr.upcase == nic.mac > @logger.info "Updating details for: > #{detail.interface} [#{nic.mac}]}" nic.bandwidth = detail.bandwidth > nic.interface_name = detail.interface > nic.save! > found = true > nic_info.delete(detail) > + end > end > end > > @@ -425,9 +430,12 @@ class HostRegister < Qmf::ConsoleHandler > 'interface_name' => nic.interface, > 'usage_type' => 1) > > - host_db.nics << detail > - > - @logger.info "Added NIC #{nic.interface} with MAC > #{nic.macaddr} to host #{host_qmf.hostname}" + if detail.interface =~ > /eth\d+\.\d+/ || detail.interface =~ /vnet\d+/ + > @logger.info "Don't Add #{detail.interface.inspect}" + else > + host_db.nics << detail > + @logger.info "Added NIC #{nic.interface} with MAC > #{nic.macaddr} to host #{host_qmf.hostname}" + end > end > end -- Arthur CLEMENT Linagora Paris From aclement at linagora.com Wed Aug 18 13:49:23 2010 From: aclement at linagora.com (Arthur =?iso-8859-15?q?Cl=E9ment?=) Date: Wed, 18 Aug 2010 15:49:23 +0200 Subject: [Ovirt-devel] [PATCH] Solve update problems Thanks to Thibault Deloffre In-Reply-To: <1279641189-2977-1-git-send-email-aclement@linagora.com> References: <1279641189-2977-1-git-send-email-aclement@linagora.com> Message-ID: <201008181549.23697.aclement@linagora.com> Pushed (Tue Jul 20) On mardi 20 juillet 2010 17:53:09 Arthur Clement wrote: > Patch in production for six months > > Signed-off-by: Arthur Clement > --- > src/host-collect/host-collect.rb | 79 > ++++++++++++++++++++++---------------- 1 files changed, 46 insertions(+), > 33 deletions(-) > > diff --git a/src/host-collect/host-collect.rb > b/src/host-collect/host-collect.rb index 1694191..b3abca8 100755 > --- a/src/host-collect/host-collect.rb > +++ b/src/host-collect/host-collect.rb > @@ -1,7 +1,7 @@ > #!/usr/bin/ruby > # > -# Copyright (C) 2008 Red Hat, Inc. > -# Written by Ian Main > +#1;2000;0c Copyright (C) 2008 Red Hat, Inc. > +#1;2000;0c Written by Ian Main > # > # This program is free software; you can redistribute it and/or modify > # it under the terms of the GNU General Public License as published by > @@ -55,14 +55,17 @@ if do_daemon > daemonize > end > > -f = UNIXSocket.new("/var/lib/collectd/unixsock") > > database_connect > > loop do > + f = UNIXSocket.new("/var/lib/collectd/unixsock") > + > f.write("LISTVAL\n") > + puts("****** SOCKET asked ******") > > count = f.gets > + puts("****** SOCKET return #{count} ******") > count = count.to_i > > vals = [] > @@ -71,39 +74,49 @@ loop do > vals.push(value) > count = count - 1 > end > - > - for val in vals do > - timestamp, keystring = val.split(" ") > - > - hostname,plugin,type = keystring.split("/") > - > - if plugin == "load" and type == "load" > - f.write("GETVAL #{keystring}\n") > - valuestring = f.gets > - > - values = valuestring.split("=") > - if values.length != 4 > - puts("GACK! Should have 4 values for load") > - next > - end > - short = values[1].to_f > - med = values[2].to_f > - long = values[3].to_f > - > - # You only see this in non-daemon mode.. > - puts("hostname: #{hostname} --> short: #{short}, med: #{med}, long: > #{long}") - > - # We have our values now, just need to update the db. > - host = Host.find(:first, :conditions => [ "hostname = ?", hostname]) > - if host == nil > - puts("GACK! No such host in database: #{hostname}") > - else > - host.load_average = med > - host.save > + > + if (vals.length > 0) > + for val in vals do > + timestamp, keystring = val.split(" ") > + > + hostname,plugin,type = keystring.split("/") > + > + if plugin == "load" and type == "load" > + puts("KEYSTRING -> #{keystring}") > + f.write("GETVAL #{keystring}\n") > + valuestring = f.gets > + 3.times { > + valuestring += f.gets > + } > + > + puts(" ------------------- VALUESTRING:: #{valuestring}") > + values = valuestring.split("=") > + if values.length != 4 > + puts("GACK! Should have 4 values for load") > + next > + end > + short = values[1].to_f > + med = values[2].to_f > + long = values[3].to_f > + > + # You only see this in non-daemon mode.. > + puts("hostname: #{hostname} --> short: #{short}, med: #{med}, > long: #{long}") + > + # We have our values now, just need to update the db. > + host = Host.find(:first, :conditions => [ "hostname = ?", > hostname]) + if host == nil > + puts("GACK! No such host in database: #{hostname}") > + else > + host.load_average = short > + host.save > + > + end > end > end > end > - > + > + puts("****** SLEEP ******") > + f.close > sleep sleeptime > > end -- Arthur CLEMENT Linagora Paris From nicolas.ochem at alcatel-lucent.com Wed Aug 18 14:52:26 2010 From: nicolas.ochem at alcatel-lucent.com (Nicolas Ochem) Date: Wed, 18 Aug 2010 16:52:26 +0200 Subject: [Ovirt-devel] node becomes "unavailable" In-Reply-To: <4C6A3C23.4090304@alcatel-lucent.com> References: <4C6A038F.8080506@redfish.com.au> <4C6A3C23.4090304@alcatel-lucent.com> Message-ID: <4C6BF3AA.6080300@alcatel-lucent.com> I found out that this problem is probably caused by the change of qmf daemon in march. I solved the problem that you describe, and another problem (taskomatic hanging) by reverting the following commits : ea01a6cbcd53105a39b14e7a216234ca8c2f7ab3 fix storage problem. 3a6c7737f7b1d0249aa70678f0a0f85126d786f9 Replace the occurence of the type @qmfc.object( 60a37d26cd488b85a409b02313995c75fecc918e Missed this for QMF update. 1eee8e47f4a1ddb3bd4a69616fef2102db7aa24 Update daemons to use new QMF. Let us know if it works, or if you have difficulties reverting the commits. On 08/17/2010 09:37 AM, Nicolas Ochem wrote: > You can look at /var/log/ovirt-server/db-omatic.log . Probably the node > times out because it does not answer to heartbeat anymore. > > To get more detail you can run the db-omatic script in no-daemon mode > (/usr/share/ovirt-server/db-omatic/db_omatic.rb -n) > > I see that very often on fedora 13, a bit less on fedora 12. > > This is because the ruby aqmp bindings get stuck when they have to > handle too many threads. > > There's no fix for this yet, but a workaround : whenever that happens, > restart everything in the node and server with this script : > > http://ovirt.pastebin.com/JjNpEDak > http://ovirt.pastebin.com/tPAPJBpB > > You can put that script in a cron job. > > On 08/17/2010 05:35 AM, Justin Clacherty wrote: > >> After running for a while the node becomes "unavailable" in the server >> UI. All VMs running on that node also become unavailable. The node is >> still running fine as are all the VMs, they're just no longer manageable. >> >> I looked on the node and everything appeared to be running fine. Looked >> on the server and ovirt-taskomatic was stopped (this seems to happen >> quite a bit). Restarted it but that didn't help. Restarting Matahari >> on the node sends information to the server but the node does not become >> available. The only way I've been able to get it back is to shutdown >> all the VMs and reboot the node and management server. Is anyone else >> seeing this happen? What else can I look at when it happens again? >> >> Cheers, >> Justin. >> >> _______________________________________________ >> Ovirt-devel mailing list >> Ovirt-devel at redhat.com >> https://www.redhat.com/mailman/listinfo/ovirt-devel >> >> > _______________________________________________ > Ovirt-devel mailing list > Ovirt-devel at redhat.com > https://www.redhat.com/mailman/listinfo/ovirt-devel > From nicolas.ochem at alcatel-lucent.com Wed Aug 18 14:57:41 2010 From: nicolas.ochem at alcatel-lucent.com (Nicolas Ochem) Date: Wed, 18 Aug 2010 16:57:41 +0200 Subject: [Ovirt-devel] Build fails due to missing ovirt-node-recipe.ks In-Reply-To: <1356803917.270.1281685205253.JavaMail.root@mail.redfish.com.au> References: <1356803917.270.1281685205253.JavaMail.root@mail.redfish.com.au> Message-ID: <4C6BF4E5.9010601@alcatel-lucent.com> > Fantastic! Thanks Nicolas, that's done the trick. The node now boots and sets up it's interfaces. The host now appears in the management server user interface and appears to report all the correct hardware. > Probably there's something missing still, but you'll only notice the difference if you have several networks. In the same line in pxelinux.cfg/default, you should also add kssendmac ksdevice=bootif and you should add an extra line immediately below with ipappend 2 If you don't do that, ovirt-early does not download the augeas interface configuration at startup and all network connections are bridged to eth0 = default behaviour. You can check if it works by checking /var/log/ovirt.log. There should be the output of a wget command there. From justin at redfish.com.au Wed Aug 18 15:09:56 2010 From: justin at redfish.com.au (Justin Clacherty) Date: Thu, 19 Aug 2010 01:09:56 +1000 Subject: [Ovirt-devel] node becomes "unavailable" In-Reply-To: <4C6BF3AA.6080300@alcatel-lucent.com> References: <4C6A038F.8080506@redfish.com.au> <4C6A3C23.4090304@alcatel-lucent.com> <4C6BF3AA.6080300@alcatel-lucent.com> Message-ID: <4C6BF7C4.602@redfish.com.au> On 19/08/2010 12:52 AM, Nicolas Ochem wrote: > I found out that this problem is probably caused by the change of qmf > daemon in march. > > I solved the problem that you describe, and another problem > (taskomatic hanging) by reverting the following commits : > > ea01a6cbcd53105a39b14e7a216234ca8c2f7ab3 fix storage problem. > 3a6c7737f7b1d0249aa70678f0a0f85126d786f9 Replace the occurence of the > type @qmfc.object( > 60a37d26cd488b85a409b02313995c75fecc918e Missed this for QMF update. > 1eee8e47f4a1ddb3bd4a69616fef2102db7aa24 Update daemons to use new QMF. > > Let us know if it works, or if you have difficulties reverting the > commits. Thanks Nicolas, I'll take a look at that. Justin. From justin at redfish.com.au Wed Aug 18 15:14:39 2010 From: justin at redfish.com.au (Justin Clacherty) Date: Thu, 19 Aug 2010 01:14:39 +1000 Subject: [Ovirt-devel] Build fails due to missing ovirt-node-recipe.ks In-Reply-To: <4C6BF4E5.9010601@alcatel-lucent.com> References: <1356803917.270.1281685205253.JavaMail.root@mail.redfish.com.au> <4C6BF4E5.9010601@alcatel-lucent.com> Message-ID: <4C6BF8DF.1080302@redfish.com.au> On 19/08/2010 12:57 AM, Nicolas Ochem wrote: > > Probably there's something missing still, but you'll only notice the > difference if you have several networks. > In the same line in pxelinux.cfg/default, you should also add > kssendmac > ksdevice=bootif > and you should add an extra line immediately below with > ipappend 2 > > > If you don't do that, ovirt-early does not download the augeas > interface configuration at startup and all network connections are > bridged to eth0 = default behaviour. > You can check if it works by checking /var/log/ovirt.log. There should > be the output of a wget command there. I did notice some oddities with the network. If I tried to bring up a VM on another node where the nic to network order was swapped it didn't seem to like it. ie. node1 - eth0 guest, eth1 admin and node2 - eth0 admin, eth1 guest. Also, nodes don't always appear. Start up one node and it all works, shut it down, start it up with the cables swapped over everything appears to work but nothing appears in the UI. Is there a way to remove nodes? I have one node that's in there twice because I was playing around with which NIC would be plugged in to which network. For the moment I've just created another group and moved any defunct nodes to there. Justin. From nicolas.ochem at alcatel-lucent.com Thu Aug 19 09:21:20 2010 From: nicolas.ochem at alcatel-lucent.com (Nicolas Ochem) Date: Thu, 19 Aug 2010 11:21:20 +0200 Subject: [Ovirt-devel] Build fails due to missing ovirt-node-recipe.ks In-Reply-To: <4C6BF8DF.1080302@redfish.com.au> References: <1356803917.270.1281685205253.JavaMail.root@mail.redfish.com.au> <4C6BF4E5.9010601@alcatel-lucent.com> <4C6BF8DF.1080302@redfish.com.au> Message-ID: <4C6CF790.2090504@alcatel-lucent.com> > I did notice some oddities with the network. If I tried to bring up a > VM on another node where the nic to network order was swapped it didn't > seem to like it. ie. node1 - eth0 guest, eth1 admin and node2 - eth0 > admin, eth1 guest. > > Also, nodes don't always appear. Start up one node and it all works, > shut it down, start it up with the cables swapped over everything > appears to work but nothing appears in the UI. > It's not obvious, but when you first boot a node, you need to "edit network" in the Hosts tab and attach the Nics to the Networks manually. > Is there a way to remove nodes? I have one node that's in there twice > because I was playing around with which NIC would be plugged in to which > network. For the moment I've just created another group and moved any > defunct nodes to there. > > > To remove a node you need to : -remove it in the kerberos database using the shell : kadmin.local -p ovirtadmin (type ? for help) -remove it in the freeipa database in /usr/share/ipa/html/*-libvirt.tab - remove it in ovirt database using /usr/share/ovirt-server/script/console production or directly in postgres CLI But I'd rather recommend running the management server in a VM and making a "clean snapshot" with no nodes. From justin at redfish.com.au Thu Aug 19 17:51:41 2010 From: justin at redfish.com.au (Justin Clacherty) Date: Fri, 20 Aug 2010 03:51:41 +1000 Subject: [Ovirt-devel] Build fails due to missing ovirt-node-recipe.ks In-Reply-To: <4C6CF790.2090504@alcatel-lucent.com> References: <1356803917.270.1281685205253.JavaMail.root@mail.redfish.com.au> <4C6BF4E5.9010601@alcatel-lucent.com> <4C6BF8DF.1080302@redfish.com.au> <4C6CF790.2090504@alcatel-lucent.com> Message-ID: <4C6D6F2D.9030409@redfish.com.au> >> Also, nodes don't always appear. Start up one node and it all works, >> shut it down, start it up with the cables swapped over everything >> appears to work but nothing appears in the UI. > > It's not obvious, but when you first boot a node, you need to "edit > network" in the Hosts tab and attach the Nics to the Networks manually. Yes, I've been doing that. The problem I'm talking about is that the node doesn't come up in the hosts tab. Justin. From aclement at linagora.com Thu Aug 19 19:55:36 2010 From: aclement at linagora.com (Arthur =?iso-8859-1?q?Cl=E9ment?=) Date: Thu, 19 Aug 2010 21:55:36 +0200 Subject: [Ovirt-devel] Build fails due to missing ovirt-node-recipe.ks In-Reply-To: <4C6D6F2D.9030409@redfish.com.au> References: <1356803917.270.1281685205253.JavaMail.root@mail.redfish.com.au> <4C6CF790.2090504@alcatel-lucent.com> <4C6D6F2D.9030409@redfish.com.au> Message-ID: <201008192155.36215.aclement@linagora.com> Hi Justin, Does /etc/init.d/ovirt-host-register and /etc/init.d/ovirt-db-omatic are running (/etc/init.d/ovirt-host-register status..) ? I'm experiencing this problem right now. The node boot correctly but it can not register because these scripts keep crashing. If they are running, check /var/log/ovirt-server/host-register.log host- collect.log and db-omatic.log. I you see an error like "wrong schemas" in host-register.log, there is a solution. On jeudi 19 ao?t 2010 19:51:41 Justin Clacherty wrote: > >> Also, nodes don't always appear. Start up one node and it all works, > >> shut it down, start it up with the cables swapped over everything > >> appears to work but nothing appears in the UI. > > > > It's not obvious, but when you first boot a node, you need to "edit > > network" in the Hosts tab and attach the Nics to the Networks manually. > > Yes, I've been doing that. The problem I'm talking about is that the > node doesn't come up in the hosts tab. > > Justin. > > _______________________________________________ > Ovirt-devel mailing list > Ovirt-devel at redhat.com > https://www.redhat.com/mailman/listinfo/ovirt-devel From aclement at linagora.com Tue Aug 24 12:12:13 2010 From: aclement at linagora.com (Arthur Clement) Date: Tue, 24 Aug 2010 14:12:13 +0200 Subject: [Ovirt-devel] [PATCH] Clean dig return (excess dot) Message-ID: <1282651933-1400-1-git-send-email-aclement@linagora.com> Signed-off-by: Arthur Clement --- scripts/ovirt-functions | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/scripts/ovirt-functions b/scripts/ovirt-functions index 7b95ea8..9e96f7b 100644 --- a/scripts/ovirt-functions +++ b/scripts/ovirt-functions @@ -186,7 +186,7 @@ find_srv() local search=$(grep search /etc/resolv.conf) search=${search#search } for d in $domain $search; do - dnsreply=$(dig +short -t srv _$1._$2.$d) + dnsreply=$(dig +short -t srv _$1._$2.$d | sed s/.$// ) rc=$? if [ $rc -eq 0 ]; then set _ $dnsreply; shift -- 1.7.2.1 From apevec at gmail.com Tue Aug 24 16:18:30 2010 From: apevec at gmail.com (Alan Pevec) Date: Tue, 24 Aug 2010 18:18:30 +0200 Subject: [Ovirt-devel] [PATCH] Clean dig return (excess dot) In-Reply-To: <1282651933-1400-1-git-send-email-aclement@linagora.com> References: <1282651933-1400-1-git-send-email-aclement@linagora.com> Message-ID: > - ? ? ? ?dnsreply=$(dig +short -t srv _$1._$2.$d) > + ? ? ? ?dnsreply=$(dig +short -t srv _$1._$2.$d | sed s/.$// ) if you want to remove dot only then + dnsreply=$(dig +short -t srv _$1._$2.$d | sed 's/\.$//' ) But why is that trailing dot a problem? It's legal in DNS. From aclement at linagora.com Tue Aug 24 18:11:25 2010 From: aclement at linagora.com (Arthur =?iso-8859-1?q?Cl=E9ment?=) Date: Tue, 24 Aug 2010 20:11:25 +0200 Subject: [Ovirt-devel] [PATCH] Clean dig return (excess dot) In-Reply-To: References: <1282651933-1400-1-git-send-email-aclement@linagora.com> Message-ID: <201008242011.25192.aclement@linagora.com> We removed this dot because we found it in the collectd.conf (network plugin did not like him). We applied this patch one month ago and it looks fine. Maybe it's better to remove the dot when the collectd.conf is generated. On mardi 24 ao?t 2010 18:18:30 Alan Pevec wrote: > > - dnsreply=$(dig +short -t srv _$1._$2.$d) > > + dnsreply=$(dig +short -t srv _$1._$2.$d | sed s/.$// ) > > if you want to remove dot only then > + dnsreply=$(dig +short -t srv _$1._$2.$d | sed 's/\.$//' ) > > But why is that trailing dot a problem? It's legal in DNS. From aclement at linagora.com Wed Aug 25 09:05:02 2010 From: aclement at linagora.com (Arthur Clement) Date: Wed, 25 Aug 2010 11:05:02 +0200 Subject: [Ovirt-devel] [PATCH] Virtio support Message-ID: <1282727102-9054-1-git-send-email-aclement@linagora.com> From: Simon Courtois Signed-off-by: Simon Courtois --- src/app/controllers/vm_controller.rb | 19 ++++++++++++++++- src/app/views/vm/_form.rhtml | 25 ++++++++++++++++++++--- src/db/migrate/044_add_virtio_to_vm_and_nic.rb | 11 ++++++++++ src/public/stylesheets/components.css | 5 ++++ src/task-omatic/task_vm.rb | 10 ++++++++- src/task-omatic/taskomatic.rb | 4 +- 6 files changed, 65 insertions(+), 9 deletions(-) create mode 100644 src/db/migrate/044_add_virtio_to_vm_and_nic.rb diff --git a/src/app/controllers/vm_controller.rb b/src/app/controllers/vm_controller.rb index 9860843..bc01a8c 100644 --- a/src/app/controllers/vm_controller.rb +++ b/src/app/controllers/vm_controller.rb @@ -146,6 +146,18 @@ class VmController < ApplicationController render :layout => false end + def edit_vmpool + svc_modify(params[:id]) + @vm = Vm.find(params[:id]) + render :layout => 'popup' + end + + def update_vmpool + svc_modify(params[:id]) + @vm = Vm.find(params[:id]) + # TODO + end + protected def _setup_provisioning_options @provisioning_options = [[Vm::PXE_OPTION_LABEL, Vm::PXE_OPTION_VALUE], @@ -176,7 +188,8 @@ class VmController < ApplicationController @vm.nics.each { |nic| nnic = Nic.new(:mac => nic.mac, :vm_id => @vm.id, - :network => nic.network) + :network => nic.network, + :virtio => nic.virtio) if(nic.network.boot_type.proto == 'static') nnic.ip_addresses << IpAddress.new(:address => nic.ip_address) @@ -216,7 +229,9 @@ class VmController < ApplicationController network_id = params[:networks][i] unless network_id.nil? || network_id == "" nic = { :mac => params[:macs][i], - :network_id => network_id, :bandwidth => 0 } + :network_id => network_id, + :bandwidth => 0, + :virtio => params[:virtio][i.to_s] } if(Network.find(network_id).boot_type.proto == 'static') # FIXME make this able to be v4 or v6 address diff --git a/src/app/views/vm/_form.rhtml b/src/app/views/vm/_form.rhtml index adb75d2..272889c 100644 --- a/src/app/views/vm/_form.rhtml +++ b/src/app/views/vm/_form.rhtml @@ -44,6 +44,11 @@
Total:
+ +
+ <%= check_box "vm", "virtio" %> + <%= label "vm", "virtio", "Use virtio" %> +
@@ -87,6 +92,12 @@   +
+ <%= hidden_field_tag "virtio[#{i}]", "0" %> + <%= check_box_tag "virtio[#{i}]", "1", @nics[i].virtio, :id => "vm_network_config_virtio_#{i}" %> + <%= label_tag "vm_network_config_virtio_#{i}", "Use virtio" %> +
+ <% if i != 0 %>
Remove @@ -113,10 +124,11 @@
-
- <%= check_box_tag_with_label "Start VM Now? (pending current resource availability)", "start_now", nil if create or @vm.state == Vm::STATE_STOPPED %> - <%= check_box_tag_with_label "Restart VM Now? (pending current resource availability)", "restart_now", nil if @vm.state == Vm::STATE_RUNNING %> - +
Restart
+