[Users] Can't create ISO domain
My ovirt-node host was created and approved in ovirt-engine. I couldn't activate an new ISO domain in the ovirt-engine if it's not existing. It could be attached to the data center though. On the other hand, I have no problem adding a data domain. How should I debug that? - David ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Host problems
I think the real issue might be that I couldn't activate an new ISO domain in the ovirt-engine if it's not existing. It could be attached to the data center though. n the other hand, I have no problem adding a data domain. The host registration problem doesn't seem to affect ovirt-node approval and data domain addition. How should I debug that? - David From: Itamar Heim [ih...@redhat.com] Sent: Friday, April 13, 2012 8:50 AM To: Li, David Cc: users@ovirt.org Subject: Re: [Users] Host problems On 04/13/2012 04:36 PM, Li, David wrote: Itamar, No, the node (or host) has been up for a week. I set up the engine using its' IP addr instead of FQDN. Could that cause any issues? you can't remove the host until you move it in engine to maintenance mode. (using ip address should be fine) - David From: Itamar Heim [ih...@redhat.com] Sent: Thursday, April 12, 2012 10:08 PM To: Li, David Cc: users@ovirt.org Subject: Re: [Users] Host problems On 04/13/2012 12:52 AM, Li, David wrote: Hi, I have two problems with hosts in the ovirt-engine: 1.I couldn’t remove the host. The remove is grayed out in the menu. I rebooted and did everything I could. How do I get around this problem without having to reinstall the engine? did you move the host to maintenance? 2.The ovirt-engine complained about “Host registration failed … (localhost.localdomain)” when I tried to add an ISO storage domain. Before that I already had a Data domain up and running. The fact is I never added the host using this name. I only used IP address. Because of this I had tried to remove the host and hence the problem #1. How can I prevent the engine from using “localhost” in the registration? i assume this is an ovirt-node? fix it's dns/hostname? Thanks in advance. David ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Host problems
Itamar, No, the node (or host) has been up for a week. I set up the engine using its' IP addr instead of FQDN. Could that cause any issues? - David From: Itamar Heim [ih...@redhat.com] Sent: Thursday, April 12, 2012 10:08 PM To: Li, David Cc: users@ovirt.org Subject: Re: [Users] Host problems On 04/13/2012 12:52 AM, Li, David wrote: Hi, I have two problems with hosts in the ovirt-engine: 1.I couldn’t remove the host. The remove is grayed out in the menu. I rebooted and did everything I could. How do I get around this problem without having to reinstall the engine? did you move the host to maintenance? 2.The ovirt-engine complained about “Host registration failed … (localhost.localdomain)” when I tried to add an ISO storage domain. Before that I already had a Data domain up and running. The fact is I never added the host using this name. I only used IP address. Because of this I had tried to remove the host and hence the problem #1. How can I prevent the engine from using “localhost” in the registration? i assume this is an ovirt-node? fix it's dns/hostname? Thanks in advance. David ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
[Users] Host problems
Hi, I have two problems with hosts in the ovirt-engine: 1. I couldn't remove the host. The remove is grayed out in the menu. I rebooted and did everything I could. How do I get around this problem without having to reinstall the engine? 2. The ovirt-engine complained about Host registration failed ... (localhost.localdomain) when I tried to add an ISO storage domain. Before that I already had a Data domain up and running. The fact is I never added the host using this name. I only used IP address. Because of this I had tried to remove the host and hence the problem #1. How can I prevent the engine from using localhost in the registration? Thanks in advance. David ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
[Users] Frozen login screen - how to debug?
Hi, I am not able to use the ovirt GUI admin login portal. The screen is completely frozen (no response at all) after I enter the admin login name and password. The only indication is a warning message Firefox 3.6 is not currently supported. I am not sure if this makes any sense. The same thing happens on Chrome and IE too. I have no such problem using CLI though. How can I debug this problem? - David ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] CLI: can't attach storagedomain
Omer,
My storage domain was in unattached stage. So I ran the following CLI to
attach it to the data center:
[oVirt shell (connected)]# create storagedomain --name
fc16-lid-engine_iso-domain --datacenter-identifier
ac2690e4-72b2-11e1-882d-000c29bf47ca
unknown error:
status: 400
reason: Bad Request
detail: [Cannot attach Storage Domain.]
The vdsm.log shows nothing specific to this but periodically spits out the
following. I am not sure if this is relevant.
Thread-1025::DEBUG::2012-03-27
00:20:16,952::task::588::TaskManager.Task::(_updateState)
Task=`2ede9ab5-f184-4f98-af9b-ac292285cc14`::moving from state init - state
preparing
Thread-1025::INFO::2012-03-27 00:20:16,953::logUtils::37::dispatcher::(wrapper)
Run and protect: repoStats(options=None)
Thread-1025::INFO::2012-03-27 00:20:16,953::logUtils::39::dispatcher::(wrapper)
Run and protect: repoStats, Return response: {}
Thread-1025::DEBUG::2012-03-27
00:20:16,953::task::1174::TaskManager.Task::(prepare)
Task=`2ede9ab5-f184-4f98-af9b-ac292285cc14`::finished: {}
Thread-1025::DEBUG::2012-03-27
00:20:16,954::task::588::TaskManager.Task::(_updateState)
Task=`2ede9ab5-f184-4f98-af9b-ac292285cc14`::moving from state preparing -
state finished
Thread-1025::DEBUG::2012-03-27
00:20:16,954::resourceManager::806::ResourceManager.Owner::(releaseAll)
Owner.releaseAll requests {} resources {}
Thread-1025::DEBUG::2012-03-27
00:20:16,954::resourceManager::841::ResourceManager.Owner::(cancelAll)
Owner.cancelAll requests {}
Thread-1025::DEBUG::2012-03-27
00:20:16,955::task::980::TaskManager.Task::(_decref)
Task=`2ede9ab5-f184-4f98-af9b-ac292285cc14`::ref 0 aborting False
- David
Date: Sun, 25 Mar 2012 06:03:59 -0400 (EDT)
From: Omer Frenkel ofren...@redhat.com
To: David Li david...@sbcglobal.net
Cc: users@ovirt.org
Subject: Re: [Users] CLI: can't attach storagedomain
Message-ID: 323fb20e-c89d-45f7-945c-c70bfb2ec...@ofrenkel.csb
Content-Type: text/plain; charset=utf-8
- Original Message -
From: David Li david...@sbcglobal.net
To: Omer Frenkel ofren...@redhat.com
Cc: users@ovirt.org
Sent: Thursday, March 22, 2012 7:18:45 PM
Subject: Re: [Users] CLI: can't attach storagedomain
Omer,
Here the engine.log after the CLI cmd execution.
2012-03-22 09:48:27,629 INFO
[org.ovirt.engine.core.bll.LoginAdminUserCommand]
(http--0.0.0.0-8080-2) Running command: LoginAdminUserCommand
internal: false.
2012-03-22 09:48:27,655 INFO
[org.ovirt.engine.core.vdsbroker.vdsbroker.ValidateStorageServerConnectionVDSCommand]
(http--0.0.0.0-8080-2) START,
ValidateStorageServerConnectionVDSCommand(vdsId =
bcb946e8-72b9-11e1-8c2b-000c29bf47ca, storagePoolId =
----, storageType = NFS,
connectionList = [{ id:
null, connection:
10.10.2.187:/home/lid2/workspace/ovirt/storagedomain/data-domain
};]), log id:
323927f6
2012-03-22 09:48:27,761 INFO
[org.ovirt.engine.core.vdsbroker.vdsbroker.ValidateStorageServerConnectionVDSCommand]
(http--0.0.0.0-8080-2) FINISH,
ValidateStorageServerConnectionVDSCommand,
return: {----=453}, log id: 323927f6
2012-03-22 09:48:27,764 ERROR
[org.ovirt.engine.core.bll.storage.NFSStorageHelper]
(http--0.0.0.0-8080-2) The
connection with details
10.10.2.187:/home/lid2/workspace/ovirt/storagedomain/data-domain
failed because
of error code 453 and error message is: the specified path does not
exist or
cannot be reached.
verify the path is correct, and for remote storage,
check the connection to your storage
2012-03-22 09:48:27,764 WARN
[org.ovirt.engine.core.bll.storage.AddStorageServerConnectionCommand]
(http--0.0.0.0-8080-2) CanDoAction of action
AddStorageServerConnection failed.
Reasons:ACTION_TYPE_FAILED_STORAGE_CONNECTION
2012-03-22 09:48:27,765 ERROR
[org.ovirt.engine.api.restapi.resource.AbstractBackendResource]
(http--0.0.0.0-8080-2) Operation Failed: [Cannot connect server to
Storage. ]
2012-03-22 09:48:27,768 INFO
[org.ovirt.engine.core.bll.LogoutUserCommand]
(http--0.0.0.0-8080-2) Running command: LogoutUserCommand internal:
false.
The path /home/lid2/workspace/ovirt/storagedomain/data-domain is
definitely real
and accessible:
drwxrwxr-x 2 vdsm kvm 4096 Mar 21 09:34 data-domain
The CLI was run on the same machine as the engine. Could that be a
problem?
i dont think so,
looks like the problem is with your host, which fails to mount this dir.
maybe attach the vdsm.log for this host and for that action (look for the
validateStorageServerConnection that fails.)
David
- Original Message
From: Omer Frenkel ofren...@redhat.com
To: David Li david...@sbcglobal.net
Cc: users@ovirt.org
Sent: Thu, March 22, 2012 4:42:56 AM
Subject: Re: [Users] CLI: can't attach storagedomain
- Original Message -
From: David Li david...@sbcglobal.net
To: Omer Frenkel ofren...@redhat.com
Cc: users@ovirt.org
Sent: Wednesday, March 21
[Users] Frozen Engine GUI
Hi, My Engine GUI freezes up at the Admin Portal once admin password is entered. This happens to all three browser types: IE, Firefox and Chrome. Apart from reinstalling the engine, are there any log files I can take a look to understand why? David ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
[Users] Can I bypass VDSM?
Hi, Is it possible to bypass VDSM but use libvirt directly to manage ovirt node? David ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] adding host with ovirt-engine-cli
Hi Michael, I have a storagedomain that was created along with the engine. It's in unattached state. When I tried to use cli to activate, I got the error: [oVirt shell (connected)]# action storagedomain iso --datacenter-identifier c479b8b5-1c43-4421-a68d-01f9ceebfb06 activate error: syntax error What's the correct syntax? - David From: Michael Pasternak [mpast...@redhat.com] Sent: Wednesday, March 07, 2012 7:04 AM To: Li, David Cc: users@ovirt.org Subject: Re: [Users] adding host with ovirt-engine-cli On 03/07/2012 04:49 PM, Li, David wrote: Hi Michael, Do I need to reinstall the SDK as well? yes. In general if I start from scrach, is the right order of installation like: engine-sdk-cli? yep, but please hold on, I'm about to do few important changes on server side, I'll let you know when you can pull the code. - David -- Michael Pasternak RedHat, ENG-Virtualization RD ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] adding host with ovirt-engine-cli
Hi Michael, Do I need to reinstall the SDK as well? In general if I start from scrach, is the right order of installation like: engine-sdk-cli? - David From: Michael Pasternak [mpast...@redhat.com] Sent: Tuesday, March 06, 2012 10:56 PM To: Li, David Cc: users@ovirt.org Subject: Re: [Users] adding host with ovirt-engine-cli Hi David, On 03/07/2012 12:08 AM, Li, David wrote: Hi Michael, Now I could create a host but couldn't delete it. This is because you using not official release of the cli, (you build cli rpm from the latest git right?!, you should use rpm i gave you) so this is due to partly developed/committed feature, if you'll build new rpm now (from the git) and reinstall it (don't forget removing the old one first cause both rpms will have same version) it will work, in future please try to avoid installing non officially released tarballs. [oVirt shell (connected)]# delete host 72b4107c-679d-11e1-9302-000c29cfd59e error: failed locating type async what's the correct syntax for deleting a host? - David -- Michael Pasternak RedHat, ENG-Virtualization RD ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] new wiki for ovirt-engine-cli
Hi Michael, I installed the cli rpms and connected to the engine. But the create host keeps giving errors and I have no clue why: [oVirt shell (connected)]# create host --address '10.10.62.106' --cluster-id 'default' --name 'xxx' --root_password 'yyy' error: incomplete command From the help menu, the rest are optional. So why is this error? You have examples creating vm and datacenter. Can you add some how to create hosts? - David From: Michael Pasternak [mpast...@redhat.com] Sent: Saturday, March 03, 2012 11:14 PM To: Li, David Cc: users@ovirt.org Subject: Re: [Users] new wiki for ovirt-engine-cli On 03/02/2012 05:02 PM, Li, David wrote: I see the Feb 19 SDK dn CLI tar balls. those aren't latest, you should wait for the next drop or build new ones by yourself (see how in wiki) I already have an engine up and running. Do I need to install both SDK and CLI to use the cli? yes Any wiki pages about installation instructions? http://www.ovirt.org/wiki/SDK#Deployment http://www.ovirt.org/wiki/CLI#Deployment - David From: Michael Pasternak [mpast...@redhat.com] Sent: Thursday, March 01, 2012 7:42 AM To: Itamar Heim Cc: Li, David; users@ovirt.org Subject: Re: [Users] new wiki for ovirt-engine-cli On 03/01/2012 05:09 PM, Itamar Heim wrote: On 03/01/2012 05:06 PM, Li, David wrote: Hi Michael, Thanks for the info. Now on to more questions:) 1. Do you have a link where to get the ovirt-engine-cli rpm or tar ball? http://www.ovirt.org/releases/ i see that ovirt-engine-cli-2.0-3 is yet not available in nightly repo, will ask to upload it, also don't forget that you need latest ovirt-engine to use it 2. I guess my real question is if the cli package can replace the engine GUI. This is necessary if I don't want to use the GUI but integrate the engine management functionalities with my own code. I don't know in this case if I should directly use the SDK or script the CLI cmds. Any suggestions? for coding (in python), SDK sounds the right integration level. if you want to code in a different language, worth discussing. sdk is the right tool for integration, see http://www.ovirt.org/wiki/SDK -- Michael Pasternak RedHat, ENG-Virtualization RD -- Michael Pasternak RedHat, ENG-Virtualization RD ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] new wiki for ovirt-engine-cli
I see the Feb 19 SDK dn CLI tar balls. I already have an engine up and running. Do I need to install both SDK and CLI to use the cli? Any wiki pages about installation instructions? - David From: Michael Pasternak [mpast...@redhat.com] Sent: Thursday, March 01, 2012 7:42 AM To: Itamar Heim Cc: Li, David; users@ovirt.org Subject: Re: [Users] new wiki for ovirt-engine-cli On 03/01/2012 05:09 PM, Itamar Heim wrote: On 03/01/2012 05:06 PM, Li, David wrote: Hi Michael, Thanks for the info. Now on to more questions:) 1. Do you have a link where to get the ovirt-engine-cli rpm or tar ball? http://www.ovirt.org/releases/ i see that ovirt-engine-cli-2.0-3 is yet not available in nightly repo, will ask to upload it, also don't forget that you need latest ovirt-engine to use it 2. I guess my real question is if the cli package can replace the engine GUI. This is necessary if I don't want to use the GUI but integrate the engine management functionalities with my own code. I don't know in this case if I should directly use the SDK or script the CLI cmds. Any suggestions? for coding (in python), SDK sounds the right integration level. if you want to code in a different language, worth discussing. sdk is the right tool for integration, see http://www.ovirt.org/wiki/SDK -- Michael Pasternak RedHat, ENG-Virtualization RD ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] new wiki for ovirt-engine-cli
Michael, I am interested in the ovirt-shell and I might have missed some earlier discussions. The wiki doesn't say much about how to install it. My questions are: 1. Is this part of ovirt-engine install? 2. Can it entirely replace the engine? Do you have a 'how to get started' page? David Date: Wed, 29 Feb 2012 17:08:13 +0200 From: Michael Pasternak mpast...@redhat.com To: users@ovirt.org Subject: [Users] new wiki for ovirt-engine-cli Message-ID: 4f4e3f5d.7010...@redhat.com Content-Type: text/plain; charset=ISO-8859-1 http://www.ovirt.org/wiki/CLI -- Michael Pasternak RedHat, ENG-Virtualization RD ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Question: ovirt-engine jboss startup failure
Oved, yeah, you are right. I put an entry into /etc/hosts for the host itself. The problem seems gone now. - David From: Oved Ourfalli [ov...@redhat.com] Sent: Tuesday, February 21, 2012 3:44 AM To: Li, David Cc: users@ovirt.org Subject: Re: [Users] Question: ovirt-engine jboss startup failure It's the first time I'm seeing such an issue, so I guess a good start in understanding the problem is checking if you have a proper localhost entry in /etc/hosts. Also, can you attach the boot.log (or any other log files created in the jboss log directory)? It might contain more useful details. Thank you, Oved - Original Message - From: David Li l...@cloudshield.com To: users@ovirt.org Sent: Tuesday, February 21, 2012 3:01:52 AM Subject: [Users] Question: ovirt-engine jboss startup failure Hi, This is my first time installing ovirt-engine. I already have a node up and running. Engine-setup script went ok except for FQDN. It complained about missing PTR record for the FQDN. If I went ahead to use the name, the JBOSS startup would fail. Upon examining the log file in /usr/share/jboss-as/standalone/log/boot.log, there was an error in java.net.InetAddress.getLocalHost. I think this is related to the PTR record complain. Haven’t done extensive search on the subject yet. How should get around this problem? David ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
[Users] Question: ovirt-engine jboss startup failure
Hi, This is my first time installing ovirt-engine. I already have a node up and running. Engine-setup script went ok except for FQDN. It complained about missing PTR record for the FQDN. If I went ahead to use the name, the JBOSS startup would fail. Upon examining the log file in /usr/share/jboss-as/standalone/log/boot.log, there was an error in java.net.InetAddress.getLocalHost. I think this is related to the PTR record complain. Haven't done extensive search on the subject yet. How should get around this problem? David ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Pxeboot - pwd/login issue
Mike, Ran into a problem at make ovirt-node-image.iso: ) ./node-creator ovirt-node-image.ks Traceback (most recent call last): File /usr/bin/livecd-creator, line 210, in module sys.exit(main()) File /usr/bin/livecd-creator, line 167, in main ks = imgcreate.read_kickstart(options.kscfg) File /usr/lib/python2.7/site-packages/imgcreate/kickstart.py, line 61, in read_kickstart e.args[0])) imgcreate.errors.KickstartError: Failed to read kickstart file '/mnt/storage/home/lid/ovirt-cache/node-creator-x86_64-tmp/ovirt-node-image.ks' : Local file does not exist: /mnt/storage/home/lid/ovirt-cache/node-creator-x86_64-tmp/ovirt-node-image.ks mv ovirt-node-image.iso ovirt-node-image-2.2.3-0.1.fc16.gitd7dbc01.iso mv: cannot stat `ovirt-node-image.iso': No such file or directory make: *** [ovirt-node-image.iso] Error 1 I checked but the file is there: [lid@fc16-lid recipe]$ ls -al /mnt/storage/home/lid/ovirt-cache/node-creator-x86_64-tmp/ovirt-node-image.ks -rw-r--r--. 1 lid Engineering 29910 Feb 13 09:13 /mnt/storage/home/lid/ovirt-cache/node-creator-x86_64-tmp/ovirt-node-image.ks Any clue? - David From: Mike Burns [mbu...@redhat.com] Sent: Monday, February 13, 2012 7:28 AM To: Li, David Cc: users@ovirt.org Subject: RE: [Users] Pxeboot - pwd/login issue On Mon, 2012-02-13 at 07:20 -0800, Li, David wrote: Hi Mike, I will also double check my side starting from a new clone. To get all the patches you 've put in for stateless, are these the steps I need to take? git clone http://gerrit.ovirt.org/p/ovirt-node.git cd ovirt-node git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/38/1738/3 don't need this one ^^. It's already in master. git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/07/1807/1 git cherry-pick FETCH_HEAD git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/08/1808/1 git cherry-pick FETCH_HEAD David -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Monday, February 13, 2012 5:56 AM To: Li, David Cc: users@ovirt.org Subject: RE: [Users] Pxeboot - pwd/login issue On Sat, 2012-02-11 at 07:39 -0800, Li, David wrote: Mike, Tried the new ssh patch and the problem still there. Both options were used BOOTIF and ssh_pwauth When I logged in first time and changed pwd, any subsequent logins were rejected. Again I couldn't capture the error message. Tried to ssh into the node, but was rejected too: [xxx oVirtNode]# ssh admin@192.168.36.114 The authenticity of host '192.168.36.114 (192.168.36.114)' can't be established. RSA key fingerprint is 23:6f:98:80:80:ec:d7:bb:e5:2a:02:59:93:85:38:39. Are you sure you want to continue connecting (yes/no)? yes Warning: Permanently added '192.168.36.114' (RSA) to the list of known hosts. Permission denied (publickey,gssapi-keyex,gssapi-with-mic). Hmm, looks like something isn't right with ssh_pwauth then. I'll look into that, but for now, try using rootpw instead of adminpw. Then login as root. Once you're in as root, you can su - admin to go the the TUI (or in this case, get the stack trace for why the TUI is failing). Mike - David From: Mike Burns [mbu...@redhat.com] Sent: Friday, February 10, 2012 11:34 AM To: Li, David Cc: users@ovirt.org Subject: RE: [Users] Pxeboot - pwd/login issue On Fri, 2012-02-10 at 11:22 -0800, Li, David wrote: ok, I have already patched like this. Hope this is ok: [lid@FC16-1 ovirt-node]$ git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/07/1807/1 git checkout FETCH_HEAD remote: Counting objects: 57, done remote: Finding sources: 100% (39/39) remote: Total 39 (delta 28), reused 39 (delta 28) Unpacking objects: 100% (39/39), done. From http://gerrit.ovirt.org/p/ovirt-node * branchrefs/changes/07/1807/1 - FETCH_HEAD Warning: you are leaving 1 commit behind, not connected to any of your branches: 07d8a0f Add archipel option with stateless boot If you want to keep them by creating a new branch, this may be a good time to do so with: git branch new_branch_name 07d8a0fc5007c5f74f9421b7a565cd2db2d9e772 HEAD is now at a94c33c... fix ssh_pwauth handling in autoinstall [lid@FC16-1 ovirt-node]$ git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/08/1808/1 git checkout FETCH_HEAD remote: Counting objects: 62, done remote: Finding sources: 100% (44/44) remote: Total 44 (delta 32), reused 44 (delta 32) Unpacking objects: 100% (44/44), done. From http://gerrit.ovirt.org/p/ovirt-node * branchrefs/changes/08/1808/1 - FETCH_HEAD Warning: you are leaving 5 commits behind, not connected to any of your branches: a94c33c fix ssh_pwauth handling in autoinstall 893a095 Add archipel option with stateless boot a1053cb regenerate repos.ks when
Re: [Users] Pxeboot - pwd/login issue
Mike, Tried the new ssh patch and the problem still there. Both options were used BOOTIF and ssh_pwauth When I logged in first time and changed pwd, any subsequent logins were rejected. Again I couldn't capture the error message. Tried to ssh into the node, but was rejected too: [xxx oVirtNode]# ssh admin@192.168.36.114 The authenticity of host '192.168.36.114 (192.168.36.114)' can't be established. RSA key fingerprint is 23:6f:98:80:80:ec:d7:bb:e5:2a:02:59:93:85:38:39. Are you sure you want to continue connecting (yes/no)? yes Warning: Permanently added '192.168.36.114' (RSA) to the list of known hosts. Permission denied (publickey,gssapi-keyex,gssapi-with-mic). - David From: Mike Burns [mbu...@redhat.com] Sent: Friday, February 10, 2012 11:34 AM To: Li, David Cc: users@ovirt.org Subject: RE: [Users] Pxeboot - pwd/login issue On Fri, 2012-02-10 at 11:22 -0800, Li, David wrote: ok, I have already patched like this. Hope this is ok: [lid@FC16-1 ovirt-node]$ git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/07/1807/1 git checkout FETCH_HEAD remote: Counting objects: 57, done remote: Finding sources: 100% (39/39) remote: Total 39 (delta 28), reused 39 (delta 28) Unpacking objects: 100% (39/39), done. From http://gerrit.ovirt.org/p/ovirt-node * branchrefs/changes/07/1807/1 - FETCH_HEAD Warning: you are leaving 1 commit behind, not connected to any of your branches: 07d8a0f Add archipel option with stateless boot If you want to keep them by creating a new branch, this may be a good time to do so with: git branch new_branch_name 07d8a0fc5007c5f74f9421b7a565cd2db2d9e772 HEAD is now at a94c33c... fix ssh_pwauth handling in autoinstall [lid@FC16-1 ovirt-node]$ git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/08/1808/1 git checkout FETCH_HEAD remote: Counting objects: 62, done remote: Finding sources: 100% (44/44) remote: Total 44 (delta 32), reused 44 (delta 32) Unpacking objects: 100% (44/44), done. From http://gerrit.ovirt.org/p/ovirt-node * branchrefs/changes/08/1808/1 - FETCH_HEAD Warning: you are leaving 5 commits behind, not connected to any of your branches: a94c33c fix ssh_pwauth handling in autoinstall 893a095 Add archipel option with stateless boot a1053cb regenerate repos.ks when building iso 237870c Add network configure before dropping into emergency shell 5bb3613 add error handling to recent rhn.py commit If you want to keep them by creating a new branch, this may be a good time to do so with: git branch new_branch_name a94c33c6d0957391a48ab0cd2e78e4a15e5ad80e HEAD is now at 3daa6d3... when stateless, don't execute store_config and remove_config [lid@FC16-1 ovirt-node]$ I *think* you're ok, because I pushed the commits such that this last command should get both. The way to check is to run $ git log and look at the commits in the log. You should see both the commits fix ssh_pwauth handling... and when stateless... commits at the top of the log. ONE MORE QUESTION: do I need to run 'make publish' before make iso? Yes, always rerun make publish. In general, I'd say rerun the autogen.sh too, but in this case you're fine. Mike - David From: Mike Burns [mbu...@redhat.com] Sent: Friday, February 10, 2012 11:19 AM To: Li, David Cc: users@ovirt.org Subject: Re: [Users] Pxeboot - pwd/login issue On Fri, 2012-02-10 at 11:09 -0800, Li, David wrote: Ok, what's the right cmd to get both? Do I have to start from scratch? No need to start from scratch. First, figure out the name of the remote. $ git remote show It will most likely return just one entry, and it's probably named origin. If it's something other than origin, then just substitute in the following commands: $ git fetch origin $ git reset --hard origin/master $ git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/07/1807/1 git cherry-pick FETCH_HEAD $ git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/08/1808/1 git cherry-pick FETCH_HEAD $ git status # should show something like: # On branch master # Your branch is ahead of 'origin/master' by 2 commits. Now you should be good to build. Mike -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Friday, February 10, 2012 11:01 AM To: Li, David Cc: users@ovirt.org Subject: Re: [Users] Pxeboot - pwd/login issue On Fri, 2012-02-10 at 10:49 -0800, Li, David wrote: Mike, I think I got a glimpse of the error message - something like persistent storage not available after I tried to login with the new passwd. Does this make sense? Try this patch with the other one I just sent: http://gerrit.ovirt.org/1808 Mike David -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Friday, February 10, 2012 7:37 AM To: Li, David Cc: users
Re: [Users] Pxeboot - pwd/login issue
Mike, ssh_pwauth=1 didn't work. The boot dropped into the dracut shell. I attached a snapshot so you can see the error. BOOTIF worked. I could ping the node but ssh was rejected. - David From: Mike Burns [mbu...@redhat.com] Sent: Friday, February 10, 2012 7:36 AM To: Li, David Cc: users@ovirt.org Subject: RE: [Users] Pxeboot - pwd/login issue On Fri, 2012-02-10 at 07:17 -0800, Li, David wrote: Hi Mike, I ran into the login/pwd problem again with this new build. Kernel option adminpw worked but only for a short time. I was asked to change pwd immediately after login but then rejected again after I did that. There was an short-lived error message on the screen but I wasn't able to capture it. You can try booting with BOOTIF=eth0 ssh_pwauth=1 in the command line. This will configure networking and enable SSH password authentication. Then you can ssh admin@your host and it will show you the error. I tried the latest from archipel source and it worked without password issue. I pulled it from: git clone https://github.com/primalmotion/Archipel.git I'm using git://github.com/primalmotion/Node.git The only differences I see between the 2 are some minor formatting and conditional changes in recipe/Makefile.am and a couple patches that were pushed since the github repo was updated. These patches are all on the gerrit.ovirt.org side. Note: there is an incompatible change that was merged into master this morning. I'm working on a rebase of the archipel stuff now to include it. Mike There might be a few updates right after you pulled it last time. Would you mind merge the new updates into ovirt node again? Thanks. David -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Thursday, February 09, 2012 5:10 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 17:05 -0800, Li, David wrote: Mike, Built it finally! Thanks a lot! Woot! -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Thursday, February 09, 2012 4:18 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 15:50 -0800, Li, David wrote: MIke, Got something new this time in make publish: /bin/sh: -c: line 36: syntax error near unexpected token `)' /bin/sh: -c: line 36: `)' make[1]: *** [repos.ks] Error 1 make[1]: Leaving directory `/home/lid/workspace/ovirt/node-archipel- 2.2.2/ovirt-node/recipe' make: *** [distdir] Error 1 [lid@FC16-1 ovirt-node]$ echo $OVIRT_CACHE_DIR $OVIRT_LOCAL_REPO /home/lid/ovirt-cache file:///home/lid/ovirt-cache/ovirt [lid@FC16-1 ovirt-node]$ Oops, bad merge on my part. $ git reset --hard origin/master #sets you back to head of the master branch $ git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/38/1738/3 git cherry-pick FETCH_HEAD # checks out the new version of the patch set. Mike - David From: Mike Burns [mbu...@redhat.com] Sent: Thursday, February 09, 2012 2:12 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 13:34 -0800, Li, David wrote: Mike, It seems it needs couple vdsm pkgs: ./node-creator ovirt-node-image.ks Error creating Live CD : Failed to find package 'vdsm-cli' : No package(s) available to install mv ovirt-node-image.iso ovirt-node-image-2.2.3-0.1fc16.git16688ea.iso mv: cannot stat `ovirt-node-image.iso': No such file or directory make: *** [ovirt-node-image.iso] Error 1 can I remove them from ks files? Not if you want to use ovirt-node with ovirt-engine. There was a bug in that patch set that is fixed in patch set 2 (along with the bad = package). I hadn't pulled the patch set together and uploaded to gerrit yet, but it's there now. git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/38/1738/2 git checkout FETCH_HEAD This should fix the problem. Mike - David From: Mike Burns [mbu...@redhat.com] Sent: Thursday, February 09, 2012 12:07 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 11:52 -0800, Li, David wrote: It's still the same error: [lid@FC16-1 ovirt-node]$ cd recipe/ [lid@FC16-1 recipe]$ make ovirt-node-image.iso ( \ echo PRODUCT='oVirt Node Hypervisor' ;\ echo PRODUCT_SHORT='oVirt Node Hypervisor' ;\ echo PACKAGE=ovirt-node-image ;\ echo VERSION=2.2.3 ;\ echo RELEASE=0.1fc16.git16688ea ;\ ) version.ks ./node-creator ovirt-node-image.ks Error creating Live CD : Failed to find package '===' : No package(s) available to install mv
Re: [Users] Pxeboot - pwd/login issue
Mike, I think I got a glimpse of the error message - something like persistent storage not available after I tried to login with the new passwd. Does this make sense? David -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Friday, February 10, 2012 7:37 AM To: Li, David Cc: users@ovirt.org Subject: RE: [Users] Pxeboot - pwd/login issue On Fri, 2012-02-10 at 07:17 -0800, Li, David wrote: Hi Mike, I ran into the login/pwd problem again with this new build. Kernel option adminpw worked but only for a short time. I was asked to change pwd immediately after login but then rejected again after I did that. There was an short-lived error message on the screen but I wasn't able to capture it. You can try booting with BOOTIF=eth0 ssh_pwauth=1 in the command line. This will configure networking and enable SSH password authentication. Then you can ssh admin@your host and it will show you the error. I tried the latest from archipel source and it worked without password issue. I pulled it from: git clone https://github.com/primalmotion/Archipel.git I'm using git://github.com/primalmotion/Node.git The only differences I see between the 2 are some minor formatting and conditional changes in recipe/Makefile.am and a couple patches that were pushed since the github repo was updated. These patches are all on the gerrit.ovirt.org side. Note: there is an incompatible change that was merged into master this morning. I'm working on a rebase of the archipel stuff now to include it. Mike There might be a few updates right after you pulled it last time. Would you mind merge the new updates into ovirt node again? Thanks. David -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Thursday, February 09, 2012 5:10 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 17:05 -0800, Li, David wrote: Mike, Built it finally! Thanks a lot! Woot! -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Thursday, February 09, 2012 4:18 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 15:50 -0800, Li, David wrote: MIke, Got something new this time in make publish: /bin/sh: -c: line 36: syntax error near unexpected token `)' /bin/sh: -c: line 36: `)' make[1]: *** [repos.ks] Error 1 make[1]: Leaving directory `/home/lid/workspace/ovirt/node-archipel- 2.2.2/ovirt-node/recipe' make: *** [distdir] Error 1 [lid@FC16-1 ovirt-node]$ echo $OVIRT_CACHE_DIR $OVIRT_LOCAL_REPO /home/lid/ovirt-cache file:///home/lid/ovirt-cache/ovirt [lid@FC16-1 ovirt-node]$ Oops, bad merge on my part. $ git reset --hard origin/master #sets you back to head of the master branch $ git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/38/1738/3 git cherry-pick FETCH_HEAD # checks out the new version of the patch set. Mike - David From: Mike Burns [mbu...@redhat.com] Sent: Thursday, February 09, 2012 2:12 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 13:34 -0800, Li, David wrote: Mike, It seems it needs couple vdsm pkgs: ./node-creator ovirt-node-image.ks Error creating Live CD : Failed to find package 'vdsm-cli' : No package(s) available to install mv ovirt-node-image.iso ovirt-node-image-2.2.3-0.1fc16.git16688ea.iso mv: cannot stat `ovirt-node-image.iso': No such file or directory make: *** [ovirt-node-image.iso] Error 1 can I remove them from ks files? Not if you want to use ovirt-node with ovirt-engine. There was a bug in that patch set that is fixed in patch set 2 (along with the bad = package). I hadn't pulled the patch set together and uploaded to gerrit yet, but it's there now. git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/38/1738/2 git checkout FETCH_HEAD This should fix the problem. Mike - David From: Mike Burns [mbu...@redhat.com] Sent: Thursday, February 09, 2012 12:07 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 11:52 -0800, Li, David wrote: It's still the same error: [lid@FC16-1 ovirt-node]$ cd recipe/ [lid@FC16-1 recipe]$ make ovirt-node-image.iso ( \ echo PRODUCT='oVirt Node Hypervisor' ;\ echo PRODUCT_SHORT='oVirt Node Hypervisor' ;\ echo PACKAGE=ovirt-node-image ;\ echo VERSION=2.2.3 ;\ echo RELEASE=0.1fc16.git16688ea ;\ ) version.ks ./node-creator ovirt-node-image.ks Error creating Live CD : Failed to find package '===' : No package(s) available to install mv ovirt-node
Re: [Users] Pxeboot - pwd/login issue
Ok, what's the right cmd to get both? Do I have to start from scratch? -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Friday, February 10, 2012 11:01 AM To: Li, David Cc: users@ovirt.org Subject: Re: [Users] Pxeboot - pwd/login issue On Fri, 2012-02-10 at 10:49 -0800, Li, David wrote: Mike, I think I got a glimpse of the error message - something like persistent storage not available after I tried to login with the new passwd. Does this make sense? Try this patch with the other one I just sent: http://gerrit.ovirt.org/1808 Mike David -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Friday, February 10, 2012 7:37 AM To: Li, David Cc: users@ovirt.org Subject: RE: [Users] Pxeboot - pwd/login issue On Fri, 2012-02-10 at 07:17 -0800, Li, David wrote: Hi Mike, I ran into the login/pwd problem again with this new build. Kernel option adminpw worked but only for a short time. I was asked to change pwd immediately after login but then rejected again after I did that. There was an short-lived error message on the screen but I wasn't able to capture it. You can try booting with BOOTIF=eth0 ssh_pwauth=1 in the command line. This will configure networking and enable SSH password authentication. Then you can ssh admin@your host and it will show you the error. I tried the latest from archipel source and it worked without password issue. I pulled it from: git clone https://github.com/primalmotion/Archipel.git I'm using git://github.com/primalmotion/Node.git The only differences I see between the 2 are some minor formatting and conditional changes in recipe/Makefile.am and a couple patches that were pushed since the github repo was updated. These patches are all on the gerrit.ovirt.org side. Note: there is an incompatible change that was merged into master this morning. I'm working on a rebase of the archipel stuff now to include it. Mike There might be a few updates right after you pulled it last time. Would you mind merge the new updates into ovirt node again? Thanks. David -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Thursday, February 09, 2012 5:10 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 17:05 -0800, Li, David wrote: Mike, Built it finally! Thanks a lot! Woot! -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Thursday, February 09, 2012 4:18 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 15:50 -0800, Li, David wrote: MIke, Got something new this time in make publish: /bin/sh: -c: line 36: syntax error near unexpected token `)' /bin/sh: -c: line 36: `)' make[1]: *** [repos.ks] Error 1 make[1]: Leaving directory `/home/lid/workspace/ovirt/node-archipel- 2.2.2/ovirt-node/recipe' make: *** [distdir] Error 1 [lid@FC16-1 ovirt-node]$ echo $OVIRT_CACHE_DIR $OVIRT_LOCAL_REPO /home/lid/ovirt-cache file:///home/lid/ovirt-cache/ovirt [lid@FC16-1 ovirt-node]$ Oops, bad merge on my part. $ git reset --hard origin/master #sets you back to head of the master branch $ git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/38/1738/3 git cherry-pick FETCH_HEAD # checks out the new version of the patch set. Mike - David From: Mike Burns [mbu...@redhat.com] Sent: Thursday, February 09, 2012 2:12 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 13:34 -0800, Li, David wrote: Mike, It seems it needs couple vdsm pkgs: ./node-creator ovirt-node-image.ks Error creating Live CD : Failed to find package 'vdsm-cli' : No package(s) available to install mv ovirt-node-image.iso ovirt-node-image-2.2.3-0.1fc16.git16688ea.iso mv: cannot stat `ovirt-node-image.iso': No such file or directory make: *** [ovirt-node-image.iso] Error 1 can I remove them from ks files? Not if you want to use ovirt-node with ovirt-engine. There was a bug in that patch set that is fixed in patch set 2 (along with the bad = package). I hadn't pulled the patch set together and uploaded to gerrit yet, but it's there now. git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/38/1738/2 git checkout FETCH_HEAD This should fix the problem. Mike - David From: Mike Burns [mbu...@redhat.com] Sent: Thursday, February 09, 2012 12:07 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 11:52 -0800, Li
Re: [Users] Pxeboot - pwd/login issue
ok, I have already patched like this. Hope this is ok: [lid@FC16-1 ovirt-node]$ git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/07/1807/1 git checkout FETCH_HEAD remote: Counting objects: 57, done remote: Finding sources: 100% (39/39) remote: Total 39 (delta 28), reused 39 (delta 28) Unpacking objects: 100% (39/39), done. From http://gerrit.ovirt.org/p/ovirt-node * branchrefs/changes/07/1807/1 - FETCH_HEAD Warning: you are leaving 1 commit behind, not connected to any of your branches: 07d8a0f Add archipel option with stateless boot If you want to keep them by creating a new branch, this may be a good time to do so with: git branch new_branch_name 07d8a0fc5007c5f74f9421b7a565cd2db2d9e772 HEAD is now at a94c33c... fix ssh_pwauth handling in autoinstall [lid@FC16-1 ovirt-node]$ git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/08/1808/1 git checkout FETCH_HEAD remote: Counting objects: 62, done remote: Finding sources: 100% (44/44) remote: Total 44 (delta 32), reused 44 (delta 32) Unpacking objects: 100% (44/44), done. From http://gerrit.ovirt.org/p/ovirt-node * branchrefs/changes/08/1808/1 - FETCH_HEAD Warning: you are leaving 5 commits behind, not connected to any of your branches: a94c33c fix ssh_pwauth handling in autoinstall 893a095 Add archipel option with stateless boot a1053cb regenerate repos.ks when building iso 237870c Add network configure before dropping into emergency shell 5bb3613 add error handling to recent rhn.py commit If you want to keep them by creating a new branch, this may be a good time to do so with: git branch new_branch_name a94c33c6d0957391a48ab0cd2e78e4a15e5ad80e HEAD is now at 3daa6d3... when stateless, don't execute store_config and remove_config [lid@FC16-1 ovirt-node]$ ONE MORE QUESTION: do I need to run 'make publish' before make iso? - David From: Mike Burns [mbu...@redhat.com] Sent: Friday, February 10, 2012 11:19 AM To: Li, David Cc: users@ovirt.org Subject: Re: [Users] Pxeboot - pwd/login issue On Fri, 2012-02-10 at 11:09 -0800, Li, David wrote: Ok, what's the right cmd to get both? Do I have to start from scratch? No need to start from scratch. First, figure out the name of the remote. $ git remote show It will most likely return just one entry, and it's probably named origin. If it's something other than origin, then just substitute in the following commands: $ git fetch origin $ git reset --hard origin/master $ git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/07/1807/1 git cherry-pick FETCH_HEAD $ git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/08/1808/1 git cherry-pick FETCH_HEAD $ git status # should show something like: # On branch master # Your branch is ahead of 'origin/master' by 2 commits. Now you should be good to build. Mike -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Friday, February 10, 2012 11:01 AM To: Li, David Cc: users@ovirt.org Subject: Re: [Users] Pxeboot - pwd/login issue On Fri, 2012-02-10 at 10:49 -0800, Li, David wrote: Mike, I think I got a glimpse of the error message - something like persistent storage not available after I tried to login with the new passwd. Does this make sense? Try this patch with the other one I just sent: http://gerrit.ovirt.org/1808 Mike David -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Friday, February 10, 2012 7:37 AM To: Li, David Cc: users@ovirt.org Subject: RE: [Users] Pxeboot - pwd/login issue On Fri, 2012-02-10 at 07:17 -0800, Li, David wrote: Hi Mike, I ran into the login/pwd problem again with this new build. Kernel option adminpw worked but only for a short time. I was asked to change pwd immediately after login but then rejected again after I did that. There was an short-lived error message on the screen but I wasn't able to capture it. You can try booting with BOOTIF=eth0 ssh_pwauth=1 in the command line. This will configure networking and enable SSH password authentication. Then you can ssh admin@your host and it will show you the error. I tried the latest from archipel source and it worked without password issue. I pulled it from: git clone https://github.com/primalmotion/Archipel.git I'm using git://github.com/primalmotion/Node.git The only differences I see between the 2 are some minor formatting and conditional changes in recipe/Makefile.am and a couple patches that were pushed since the github repo was updated. These patches are all on the gerrit.ovirt.org side. Note: there is an incompatible change that was merged into master this morning. I'm working on a rebase of the archipel stuff now to include it. Mike There might be a few updates right after you pulled it last time. Would you mind merge the new updates into ovirt node again
Re: [Users] Pxeboot
Mike, Two quick questions before I kick off a build: 1. Git errors. Do they matter? [lid@FC16-1 node-archipel-2.2.2]$ git clone http://gerrit.ovirt.org/p/ovirt-node.git refs/changes/38/1738/1git checkout FETCH_HEAD Cloning into refs/changes/38/1738/1... remote: Counting objects: 14791, done remote: Finding sources: 100% (14791/14791) remote: Total 14791 (delta 10855), reused 14778 (delta 10855) Receiving objects: 100% (14791/14791), 16.92 MiB | 154 KiB/s, done. Resolving deltas: 100% (10855/10855), done. fatal: Not a git repository (or any of the parent directories): .git 2. Build machine. I am using a x86_64 FC16 VM as the build machine. My target is x86_64 Intel board. Is this ok? - David From: Mike Burns [mbu...@redhat.com] Sent: Thursday, February 09, 2012 8:16 AM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: RE: [Users] Pxeboot On Thu, 2012-02-09 at 08:10 -0800, Li, David wrote: Not sure where the boot log is since I can't login. I got the source from https://github.com/primalmotion/archipel-node and followed the build instructions in http://ovirt.org/wiki/Node_Building maybe I should rebuild the iso from the ovirt node source tree as you have pulled in archipel changes yesterday. How do I do that? git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/38/1738/1 git checkout FETCH_HEAD That command will checkout the current version of the archipel-node changes. Then you can just follow the Node_Building process. Note that the changes aren't merged into master yet. I'm still in the process of reviewing them. Mike - David -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Thursday, February 09, 2012 7:22 AM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 07:06 -0800, Li, David wrote: Mike, I added adminpw as a kernel option. After boot, I tried admin/abc123 and root/abc123 but not able to login. Any idea? Are there any errors in your build or boot logs? Mike David -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Wednesday, February 08, 2012 6:53 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Wed, 2012-02-08 at 17:11 -0800, Li, David wrote: Ok, I built an iso out of archipel and pxebooted it with stateless option. Now I got a localhost login prompt. Any idea what should I put in as user/pwd? David You will need to pass a password in on the command line as well adminpw=hashed_password You can generate the hashed password using: $ openssl passwd Assuming the password was abc123 $ openssl passwd abc123 tKrQfufCbosr6 adminpw=tKrQfufCbosr6 Mike -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Wednesday, February 08, 2012 4:31 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: RE: [Users] Pxeboot On Wed, 2012-02-08 at 15:43 -0800, Li, David wrote: Ok, I think we are on the same page now regarding stateless node. It looks like the achipel diff you just pulled in might be able to do this. But I can't quite make it out what the kernel option should be. Is it stateless=1? stateless=1 or just stateless, both will work. David -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Wednesday, February 08, 2012 3:27 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Wed, 2012-02-08 at 15:19 -0800, Li, David wrote: Perry, I am glad you brought this up. I 'd be perfectly happy to run the root fs entirely in RAM. In fact that would be ideal for me. It's just I am not quite sure how to use pxeboot to achieve this. I am doing a test now using the tftpboot files created from the iso. As far as I see, the kernel boot options (pxelinux.cfg/default) has: root=live:/ovirt-node-image-2.2.2-1.1.fc16.iso With this I can only pxeboot to the intall screen. What should I use to let the kernel mount the root fs in memory? Something like root=/dev/ram0? David It's not currently possible with ovirt-node. That's the whole stateless feature that we outlined. Once we actually have it implemented, it will be something along the lines of adding stateless to the kernel commandline. Mike -Original Message- From: Perry Myers [mailto:pmy...@redhat.com] Sent: Wednesday, February 08, 2012 2:54 PM To: Mike Burns Cc: Li, David; users@ovirt.org Subject: Re: [Users] Pxeboot On 02/08/2012 05:03 PM, Mike Burns wrote: On Wed, 2012-02-08 at 13:46 -0800, Li, David wrote: Mike, If I understand this correctly
Re: [Users] Pxeboot
ok, that's my bad:) Ran into error at the last iso build step: Start building ISO image ( \ echo PRODUCT='oVirt Node Hypervisor' ;\ echo PRODUCT_SHORT='oVirt Node Hypervisor' ;\ echo PACKAGE=ovirt-node-image ;\ echo VERSION=2.2.3 ;\ echo RELEASE=0.1.fc16.git16688ea ;\ ) version.ks ./node-creator ovirt-node-image.ks Error creating Live CD : Unable to download from repo : Cannot retrieve repository metadata (repomd.xml) for repository: local. Please verify its path and try again mv ovirt-node-image.iso ovirt-node-image-2.2.3-0.1.fc16.git16688ea.iso mv: cannot stat `ovirt-node-image.iso': No such file or directory make: *** [ovirt-node-image.iso] Error 1 I wonder which repo my machine is lacking? - David From: Mike Burns [mbu...@redhat.com] Sent: Thursday, February 09, 2012 10:02 AM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 09:47 -0800, Li, David wrote: Got an error: [lid@FC16-1 ovirt-node]$ git fetch http://gerrit.ovirt.org/ refs/changes/38/1738/1git cherry-pick FETCH_HEAD fatal: http://gerrit.ovirt.org/info/refs not found: did you run git update-server-info on the server? Bad copy/paste: git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/38/1738/1 git cherry-pick FETCH_HEAD any ideas? - David From: Mike Burns [mbu...@redhat.com] Sent: Thursday, February 09, 2012 8:37 AM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 08:30 -0800, Li, David wrote: Mike, Two quick questions before I kick off a build: 1. Git errors. Do they matter? [lid@FC16-1 node-archipel-2.2.2]$ git clone http://gerrit.ovirt.org/p/ovirt-node.git refs/changes/38/1738/1git checkout FETCH_HEAD Cloning into refs/changes/38/1738/1... remote: Counting objects: 14791, done remote: Finding sources: 100% (14791/14791) remote: Total 14791 (delta 10855), reused 14778 (delta 10855) Receiving objects: 100% (14791/14791), 16.92 MiB | 154 KiB/s, done. Resolving deltas: 100% (10855/10855), done. fatal: Not a git repository (or any of the parent directories): .git try this instead: git clone http://gerrit.ovirt.org/p/ovirt-node.git cd ovirt-node git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/38/1738/1 git cherry-pick FETCH_HEAD 2. Build machine. I am using a x86_64 FC16 VM as the build machine. My target is x86_64 Intel board. Is this ok? Should be fine. Mike - David From: Mike Burns [mbu...@redhat.com] Sent: Thursday, February 09, 2012 8:16 AM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: RE: [Users] Pxeboot On Thu, 2012-02-09 at 08:10 -0800, Li, David wrote: Not sure where the boot log is since I can't login. I got the source from https://github.com/primalmotion/archipel-node and followed the build instructions in http://ovirt.org/wiki/Node_Building maybe I should rebuild the iso from the ovirt node source tree as you have pulled in archipel changes yesterday. How do I do that? git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/38/1738/1 git checkout FETCH_HEAD That command will checkout the current version of the archipel-node changes. Then you can just follow the Node_Building process. Note that the changes aren't merged into master yet. I'm still in the process of reviewing them. Mike - David -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Thursday, February 09, 2012 7:22 AM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 07:06 -0800, Li, David wrote: Mike, I added adminpw as a kernel option. After boot, I tried admin/abc123 and root/abc123 but not able to login. Any idea? Are there any errors in your build or boot logs? Mike David -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Wednesday, February 08, 2012 6:53 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Wed, 2012-02-08 at 17:11 -0800, Li, David wrote: Ok, I built an iso out of archipel and pxebooted it with stateless option. Now I got a localhost login prompt. Any idea what should I put in as user/pwd? David You will need to pass a password in on the command line as well adminpw=hashed_password You can generate the hashed password using: $ openssl passwd Assuming the password was abc123 $ openssl passwd abc123 tKrQfufCbosr6 adminpw=tKrQfufCbosr6 Mike -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Wednesday, February 08, 2012
Re: [Users] Pxeboot
Mike,
It seems it needs couple vdsm pkgs:
./node-creator ovirt-node-image.ks
Error creating Live CD : Failed to find package 'vdsm-cli' : No package(s)
available to install
mv ovirt-node-image.iso ovirt-node-image-2.2.3-0.1fc16.git16688ea.iso
mv: cannot stat `ovirt-node-image.iso': No such file or directory
make: *** [ovirt-node-image.iso] Error 1
can I remove them from ks files?
- David
From: Mike Burns [mbu...@redhat.com]
Sent: Thursday, February 09, 2012 12:07 PM
To: Li, David
Cc: Perry Myers; users@ovirt.org
Subject: Re: [Users] Pxeboot
On Thu, 2012-02-09 at 11:52 -0800, Li, David wrote:
It's still the same error:
[lid@FC16-1 ovirt-node]$ cd recipe/
[lid@FC16-1 recipe]$ make ovirt-node-image.iso
( \
echo PRODUCT='oVirt Node Hypervisor' ;\
echo PRODUCT_SHORT='oVirt Node Hypervisor' ;\
echo PACKAGE=ovirt-node-image ;\
echo VERSION=2.2.3 ;\
echo RELEASE=0.1fc16.git16688ea ;\
) version.ks
./node-creator ovirt-node-image.ks
Error creating Live CD : Failed to find package '===' : No package(s)
available to install
mv ovirt-node-image.iso ovirt-node-image-2.2.3-0.1fc16.git16688ea.iso
mv: cannot stat `ovirt-node-image.iso': No such file or directory
make: *** [ovirt-node-image.iso] Error 1
[lid@FC16-1 recipe]$ echo $OVIRT_LOCAL_REPO
file:///home/lid/ovirt-cache/ovirt
[lid@FC16-1 recipe]$ ll /home/lid/ovirt-cache/ovirt/
total 12
drwxr-xr-x. 2 lid lid 4096 Feb 9 11:10 noarch
drwxrwxr-x. 2 lid lid 4096 Feb 9 11:10 repodata
drwxrwxr-x. 2 lid lid 4096 Feb 9 11:10 src
That's a bug in the patch. edit recipe/common-pkgs.ks and remove the
last line which is ===
Mike
Did make publish store the rpms in $OVIRT_LOCAL_REPO?
- David
From: Mike Burns [mbu...@redhat.com]
Sent: Thursday, February 09, 2012 11:44 AM
To: Li, David
Cc: Perry Myers; users@ovirt.org
Subject: RE: [Users] Pxeboot
On Thu, 2012-02-09 at 11:40 -0800, Li, David wrote:
The error seems a little stubborn.
[lid@FC16-1 recipe]$ echo $OVIRT_LOCAL_REPO
file://home/lid/ovirt-cache/ovirt
You need 3 slashes. file:///home/lid/ovirt-cache/ovirt
[lid@FC16-1 recipe]$ make ovirt-node-image.iso
( \
echo PRODUCT='oVirt Node Hypervisor' ;\
echo PRODUCT_SHORT='oVirt Node Hypervisor' ;\
echo PACKAGE=ovirt-node-image ;\
echo VERSION=2.2.3 ;\
echo RELEASE=0.1fc16.git16688ea ;\
) version.ks
./node-creator ovirt-node-image.ks
Error creating Live CD : Unable to download from repo : Cannot retrieve
repository metadata (repomd.xml) for repository: local. Please verify its
path and try again
mv ovirt-node-image.iso ovirt-node-image-2.2.3-0.1fc16.git16688ea.iso
mv: cannot stat `ovirt-node-image.iso': No such file or directory
make: *** [ovirt-node-image.iso] Error 1
There is also a pop-up window: Unable to open a folder for _ovirt-node-imag
maybe it's looking at the wrong place?
- David
From: Mike Burns [mbu...@redhat.com]
Sent: Thursday, February 09, 2012 11:15 AM
To: Li, David
Cc: Perry Myers; users@ovirt.org
Subject: RE: [Users] Pxeboot
On Thu, 2012-02-09 at 10:57 -0800, Li, David wrote:
ok, that's my bad:)
Ran into error at the last iso build step:
Start building ISO image
( \
echo PRODUCT='oVirt Node Hypervisor' ;\
echo PRODUCT_SHORT='oVirt Node Hypervisor' ;\
echo PACKAGE=ovirt-node-image ;\
echo VERSION=2.2.3 ;\
echo RELEASE=0.1.fc16.git16688ea ;\
) version.ks
./node-creator ovirt-node-image.ks
Error creating Live CD : Unable to download from repo : Cannot retrieve
repository metadata (repomd.xml) for repository: local. Please verify its
path and try again
mv ovirt-node-image.iso ovirt-node-image-2.2.3-0.1.fc16.git16688ea.iso
mv: cannot stat `ovirt-node-image.iso': No such file or directory
make: *** [ovirt-node-image.iso] Error 1
IIRC, this is usually because you have OVIRT_LOCAL_REPO defined, but not
correctly. It needs to be like this: file:///home/mburns/rpmbuild/RPMS
Ok, I just looked at the Node Building page, and it needs some updates.
Try this from the base directory of the git repo
$ export OVIRT_CACHE_DIR=${HOME}/ovirt-cache
$ export OVIRT_LOCAL_REPO=file://${OVIRT_CACHE_DIR}/ovirt
$ ./autogen.sh --with-image-minimizer
$ make publish #this builds the rpms and puts them in
${OVIRT_CACHE_DIR}
$ cd recipe
$ make ovirt-node-image.iso
Mike
I wonder which repo my machine is lacking?
- David
From: Mike Burns [mbu...@redhat.com]
Sent: Thursday, February 09, 2012 10:02 AM
To: Li, David
Cc: Perry Myers; users@ovirt.org
Subject: Re: [Users] Pxeboot
On Thu, 2012-02-09 at 09:47 -0800, Li, David wrote:
Got an error:
[lid@FC16-1 ovirt-node]$ git fetch http://gerrit.ovirt.org
Re: [Users] Pxeboot
Mike, I 'd like to start again clean. Will the following give me the latest stateless source and iso build: git clone http://gerrit.ovirt.org/p/ovirt-node.git cd ovirt-node git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/38/1738/1 git cherry-pick FETCH_HEAD git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/38/1738/2 git checkout FETCH_HEAD -David -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Thursday, February 09, 2012 2:12 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 13:34 -0800, Li, David wrote: Mike, It seems it needs couple vdsm pkgs: ./node-creator ovirt-node-image.ks Error creating Live CD : Failed to find package 'vdsm-cli' : No package(s) available to install mv ovirt-node-image.iso ovirt-node-image-2.2.3-0.1fc16.git16688ea.iso mv: cannot stat `ovirt-node-image.iso': No such file or directory make: *** [ovirt-node-image.iso] Error 1 can I remove them from ks files? Not if you want to use ovirt-node with ovirt-engine. There was a bug in that patch set that is fixed in patch set 2 (along with the bad = package). I hadn't pulled the patch set together and uploaded to gerrit yet, but it's there now. git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/38/1738/2 git checkout FETCH_HEAD This should fix the problem. Mike - David From: Mike Burns [mbu...@redhat.com] Sent: Thursday, February 09, 2012 12:07 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 11:52 -0800, Li, David wrote: It's still the same error: [lid@FC16-1 ovirt-node]$ cd recipe/ [lid@FC16-1 recipe]$ make ovirt-node-image.iso ( \ echo PRODUCT='oVirt Node Hypervisor' ;\ echo PRODUCT_SHORT='oVirt Node Hypervisor' ;\ echo PACKAGE=ovirt-node-image ;\ echo VERSION=2.2.3 ;\ echo RELEASE=0.1fc16.git16688ea ;\ ) version.ks ./node-creator ovirt-node-image.ks Error creating Live CD : Failed to find package '===' : No package(s) available to install mv ovirt-node-image.iso ovirt-node-image-2.2.3-0.1fc16.git16688ea.iso mv: cannot stat `ovirt-node-image.iso': No such file or directory make: *** [ovirt-node-image.iso] Error 1 [lid@FC16-1 recipe]$ echo $OVIRT_LOCAL_REPO file:///home/lid/ovirt-cache/ovirt [lid@FC16-1 recipe]$ ll /home/lid/ovirt-cache/ovirt/ total 12 drwxr-xr-x. 2 lid lid 4096 Feb 9 11:10 noarch drwxrwxr-x. 2 lid lid 4096 Feb 9 11:10 repodata drwxrwxr-x. 2 lid lid 4096 Feb 9 11:10 src That's a bug in the patch. edit recipe/common-pkgs.ks and remove the last line which is === Mike Did make publish store the rpms in $OVIRT_LOCAL_REPO? - David From: Mike Burns [mbu...@redhat.com] Sent: Thursday, February 09, 2012 11:44 AM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: RE: [Users] Pxeboot On Thu, 2012-02-09 at 11:40 -0800, Li, David wrote: The error seems a little stubborn. [lid@FC16-1 recipe]$ echo $OVIRT_LOCAL_REPO file://home/lid/ovirt-cache/ovirt You need 3 slashes. file:///home/lid/ovirt-cache/ovirt [lid@FC16-1 recipe]$ make ovirt-node-image.iso ( \ echo PRODUCT='oVirt Node Hypervisor' ;\ echo PRODUCT_SHORT='oVirt Node Hypervisor' ;\ echo PACKAGE=ovirt-node-image ;\ echo VERSION=2.2.3 ;\ echo RELEASE=0.1fc16.git16688ea ;\ ) version.ks ./node-creator ovirt-node-image.ks Error creating Live CD : Unable to download from repo : Cannot retrieve repository metadata (repomd.xml) for repository: local. Please verify its path and try again mv ovirt-node-image.iso ovirt-node-image-2.2.3-0.1fc16.git16688ea.iso mv: cannot stat `ovirt-node-image.iso': No such file or directory make: *** [ovirt-node-image.iso] Error 1 There is also a pop-up window: Unable to open a folder for _ovirt-node-imag maybe it's looking at the wrong place? - David From: Mike Burns [mbu...@redhat.com] Sent: Thursday, February 09, 2012 11:15 AM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: RE: [Users] Pxeboot On Thu, 2012-02-09 at 10:57 -0800, Li, David wrote: ok, that's my bad:) Ran into error at the last iso build step: Start building ISO image ( \ echo PRODUCT='oVirt Node Hypervisor' ;\ echo PRODUCT_SHORT='oVirt Node Hypervisor' ;\ echo PACKAGE=ovirt-node-image ;\ echo VERSION=2.2.3 ;\ echo RELEASE=0.1.fc16.git16688ea ;\ ) version.ks ./node-creator ovirt-node-image.ks Error creating Live CD : Unable to download from repo : Cannot retrieve repository metadata (repomd.xml) for repository: local. Please verify its path and try again mv ovirt-node-image.iso ovirt-node-image-2.2.3-0.1.fc16.git16688ea.iso
Re: [Users] Pxeboot
Mike, Built it finally! Thanks a lot! -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Thursday, February 09, 2012 4:18 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 15:50 -0800, Li, David wrote: MIke, Got something new this time in make publish: /bin/sh: -c: line 36: syntax error near unexpected token `)' /bin/sh: -c: line 36: `)' make[1]: *** [repos.ks] Error 1 make[1]: Leaving directory `/home/lid/workspace/ovirt/node-archipel- 2.2.2/ovirt-node/recipe' make: *** [distdir] Error 1 [lid@FC16-1 ovirt-node]$ echo $OVIRT_CACHE_DIR $OVIRT_LOCAL_REPO /home/lid/ovirt-cache file:///home/lid/ovirt-cache/ovirt [lid@FC16-1 ovirt-node]$ Oops, bad merge on my part. $ git reset --hard origin/master #sets you back to head of the master branch $ git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/38/1738/3 git cherry-pick FETCH_HEAD # checks out the new version of the patch set. Mike - David From: Mike Burns [mbu...@redhat.com] Sent: Thursday, February 09, 2012 2:12 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 13:34 -0800, Li, David wrote: Mike, It seems it needs couple vdsm pkgs: ./node-creator ovirt-node-image.ks Error creating Live CD : Failed to find package 'vdsm-cli' : No package(s) available to install mv ovirt-node-image.iso ovirt-node-image-2.2.3-0.1fc16.git16688ea.iso mv: cannot stat `ovirt-node-image.iso': No such file or directory make: *** [ovirt-node-image.iso] Error 1 can I remove them from ks files? Not if you want to use ovirt-node with ovirt-engine. There was a bug in that patch set that is fixed in patch set 2 (along with the bad = package). I hadn't pulled the patch set together and uploaded to gerrit yet, but it's there now. git fetch http://gerrit.ovirt.org/p/ovirt-node refs/changes/38/1738/2 git checkout FETCH_HEAD This should fix the problem. Mike - David From: Mike Burns [mbu...@redhat.com] Sent: Thursday, February 09, 2012 12:07 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Thu, 2012-02-09 at 11:52 -0800, Li, David wrote: It's still the same error: [lid@FC16-1 ovirt-node]$ cd recipe/ [lid@FC16-1 recipe]$ make ovirt-node-image.iso ( \ echo PRODUCT='oVirt Node Hypervisor' ;\ echo PRODUCT_SHORT='oVirt Node Hypervisor' ;\ echo PACKAGE=ovirt-node-image ;\ echo VERSION=2.2.3 ;\ echo RELEASE=0.1fc16.git16688ea ;\ ) version.ks ./node-creator ovirt-node-image.ks Error creating Live CD : Failed to find package '===' : No package(s) available to install mv ovirt-node-image.iso ovirt-node-image-2.2.3-0.1fc16.git16688ea.iso mv: cannot stat `ovirt-node-image.iso': No such file or directory make: *** [ovirt-node-image.iso] Error 1 [lid@FC16-1 recipe]$ echo $OVIRT_LOCAL_REPO file:///home/lid/ovirt-cache/ovirt [lid@FC16-1 recipe]$ ll /home/lid/ovirt-cache/ovirt/ total 12 drwxr-xr-x. 2 lid lid 4096 Feb 9 11:10 noarch drwxrwxr-x. 2 lid lid 4096 Feb 9 11:10 repodata drwxrwxr-x. 2 lid lid 4096 Feb 9 11:10 src That's a bug in the patch. edit recipe/common-pkgs.ks and remove the last line which is === Mike Did make publish store the rpms in $OVIRT_LOCAL_REPO? - David From: Mike Burns [mbu...@redhat.com] Sent: Thursday, February 09, 2012 11:44 AM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: RE: [Users] Pxeboot On Thu, 2012-02-09 at 11:40 -0800, Li, David wrote: The error seems a little stubborn. [lid@FC16-1 recipe]$ echo $OVIRT_LOCAL_REPO file://home/lid/ovirt-cache/ovirt You need 3 slashes. file:///home/lid/ovirt-cache/ovirt [lid@FC16-1 recipe]$ make ovirt-node-image.iso ( \ echo PRODUCT='oVirt Node Hypervisor' ;\ echo PRODUCT_SHORT='oVirt Node Hypervisor' ;\ echo PACKAGE=ovirt-node-image ;\ echo VERSION=2.2.3 ;\ echo RELEASE=0.1fc16.git16688ea ;\ ) version.ks ./node-creator ovirt-node-image.ks Error creating Live CD : Unable to download from repo : Cannot retrieve repository metadata (repomd.xml) for repository: local. Please verify its path and try again mv ovirt-node-image.iso ovirt-node-image-2.2.3-0.1fc16.git16688ea.iso mv: cannot stat `ovirt-node-image.iso': No such file or directory make: *** [ovirt-node-image.iso] Error 1 There is also a pop-up window: Unable to open a folder for _ovirt-node-imag maybe it's looking at the wrong place? - David From: Mike Burns [mbu...@redhat.com] Sent: Thursday, February 09, 2012 11:15 AM To: Li, David Cc: Perry Myers; users
Re: [Users] Pxeboot
Hi Douglas, Thanks for the link. However I can't quite tell if this is pxeboot installation or a truly stateless pxeboot that you not only suck in the configurations from the server but also NFS mount the final RHEV-H root fs from the same server. Mike has mentioned oVirt Node 2.3 can only do the install but not the stateless boot. I wonder if this is the same case with RHEV-H. David -Original Message- From: Douglas Landgraf [mailto:dougsl...@redhat.com] Sent: Wednesday, February 08, 2012 1:23 PM To: Mike Burns Cc: Li, David; users@ovirt.org Subject: Re: [Users] Pxeboot Hi Li, David, On 02/08/2012 01:11 PM, Mike Burns wrote: On Wed, 2012-02-08 at 09:53 -0800, Li, David wrote: Since oVirt node doesn’t support pxeboot in the current release, does anyone know if I can use RHEV-H to pxeboot and nfs mout the rootfs? Where does it say that ovirt-node doesn't support pxe boot? I use PXE in my testing. I haven't heard of any issues with PXE either. The livecd-iso-to-pxeboot script in livecd-tools will create the initrd and vmlinuz files for you to import into your PXE server. http://docs.redhat.com/docs/en- US/Red_Hat_Enterprise_Virtualization_for_Servers/2.1/html/5.4- 2.1_Hypervisor_Deployment_Guide/sect-Deployment_Guide- Preparing_Red_Hat_Enterprise_Virtualization_Hypervisor_installation_media -Deploying_RHEV_Hypervisors_with_PXE_and_tftp.html and http://dougsland.livejournal.com/122359.html might help you. Fell free to share any doubt with the mailing list. -- Cheers Douglas ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Pxeboot
Perry, I am glad you brought this up. I 'd be perfectly happy to run the root fs entirely in RAM. In fact that would be ideal for me. It's just I am not quite sure how to use pxeboot to achieve this. I am doing a test now using the tftpboot files created from the iso. As far as I see, the kernel boot options (pxelinux.cfg/default) has: root=live:/ovirt-node-image-2.2.2-1.1.fc16.iso With this I can only pxeboot to the intall screen. What should I use to let the kernel mount the root fs in memory? Something like root=/dev/ram0? David -Original Message- From: Perry Myers [mailto:pmy...@redhat.com] Sent: Wednesday, February 08, 2012 2:54 PM To: Mike Burns Cc: Li, David; users@ovirt.org Subject: Re: [Users] Pxeboot On 02/08/2012 05:03 PM, Mike Burns wrote: On Wed, 2012-02-08 at 13:46 -0800, Li, David wrote: Mike, If I understand this correctly, today I should be able to pxeboot and nfs mount the root fs from a remote server. Apart from setting up the pxe stuff, I 'd have to populate the ovirt node root fs on the server - perhaps steal it from a disk install. In other words I am concerned about the point from which the kernel starts to execute /init script (in the initramfs) to the point /init is able to mount the final root fs from a remote server. No, there is no way to set this up currently in ovirt-node. You could install using a remote iscsi lun if you have a hardware iscsi HBA, but there isn't a way to mount a remote nfs share as the root fs. Supporting a remote NFS share as the root fs isn't even something that requested as an RFE at this point or on the roadmap as far as I'm aware. It sounds like what you're really looking for is a shared root fs that multiple hosts could use. This is something that we will probably look into eventually, but it's not on the immediate roadmap. Given that the rootfs of oVirt Node is fairly small and in a truly stateless environment would just run out of system RAM, there's no real reason to try to do a shared NFS based rootfs. It's an unnecessary complication I think, if the end goal is to move to truly stateless. For larger systems where the rootfs is on the order of GB's, shared root may make more sense. ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Pxeboot
Ok, I think we are on the same page now regarding stateless node. It looks like the achipel diff you just pulled in might be able to do this. But I can't quite make it out what the kernel option should be. Is it stateless=1? David -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Wednesday, February 08, 2012 3:27 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Wed, 2012-02-08 at 15:19 -0800, Li, David wrote: Perry, I am glad you brought this up. I 'd be perfectly happy to run the root fs entirely in RAM. In fact that would be ideal for me. It's just I am not quite sure how to use pxeboot to achieve this. I am doing a test now using the tftpboot files created from the iso. As far as I see, the kernel boot options (pxelinux.cfg/default) has: root=live:/ovirt-node-image-2.2.2-1.1.fc16.iso With this I can only pxeboot to the intall screen. What should I use to let the kernel mount the root fs in memory? Something like root=/dev/ram0? David It's not currently possible with ovirt-node. That's the whole stateless feature that we outlined. Once we actually have it implemented, it will be something along the lines of adding stateless to the kernel commandline. Mike -Original Message- From: Perry Myers [mailto:pmy...@redhat.com] Sent: Wednesday, February 08, 2012 2:54 PM To: Mike Burns Cc: Li, David; users@ovirt.org Subject: Re: [Users] Pxeboot On 02/08/2012 05:03 PM, Mike Burns wrote: On Wed, 2012-02-08 at 13:46 -0800, Li, David wrote: Mike, If I understand this correctly, today I should be able to pxeboot and nfs mount the root fs from a remote server. Apart from setting up the pxe stuff, I 'd have to populate the ovirt node root fs on the server - perhaps steal it from a disk install. In other words I am concerned about the point from which the kernel starts to execute /init script (in the initramfs) to the point /init is able to mount the final root fs from a remote server. No, there is no way to set this up currently in ovirt-node. You could install using a remote iscsi lun if you have a hardware iscsi HBA, but there isn't a way to mount a remote nfs share as the root fs. Supporting a remote NFS share as the root fs isn't even something that requested as an RFE at this point or on the roadmap as far as I'm aware. It sounds like what you're really looking for is a shared root fs that multiple hosts could use. This is something that we will probably look into eventually, but it's not on the immediate roadmap. Given that the rootfs of oVirt Node is fairly small and in a truly stateless environment would just run out of system RAM, there's no real reason to try to do a shared NFS based rootfs. It's an unnecessary complication I think, if the end goal is to move to truly stateless. For larger systems where the rootfs is on the order of GB's, shared root may make more sense. ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Pxeboot
Ok, I built an iso out of archipel and pxebooted it with stateless option. Now I got a localhost login prompt. Any idea what should I put in as user/pwd? David -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Wednesday, February 08, 2012 4:31 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: RE: [Users] Pxeboot On Wed, 2012-02-08 at 15:43 -0800, Li, David wrote: Ok, I think we are on the same page now regarding stateless node. It looks like the achipel diff you just pulled in might be able to do this. But I can't quite make it out what the kernel option should be. Is it stateless=1? stateless=1 or just stateless, both will work. David -Original Message- From: Mike Burns [mailto:mbu...@redhat.com] Sent: Wednesday, February 08, 2012 3:27 PM To: Li, David Cc: Perry Myers; users@ovirt.org Subject: Re: [Users] Pxeboot On Wed, 2012-02-08 at 15:19 -0800, Li, David wrote: Perry, I am glad you brought this up. I 'd be perfectly happy to run the root fs entirely in RAM. In fact that would be ideal for me. It's just I am not quite sure how to use pxeboot to achieve this. I am doing a test now using the tftpboot files created from the iso. As far as I see, the kernel boot options (pxelinux.cfg/default) has: root=live:/ovirt-node-image-2.2.2-1.1.fc16.iso With this I can only pxeboot to the intall screen. What should I use to let the kernel mount the root fs in memory? Something like root=/dev/ram0? David It's not currently possible with ovirt-node. That's the whole stateless feature that we outlined. Once we actually have it implemented, it will be something along the lines of adding stateless to the kernel commandline. Mike -Original Message- From: Perry Myers [mailto:pmy...@redhat.com] Sent: Wednesday, February 08, 2012 2:54 PM To: Mike Burns Cc: Li, David; users@ovirt.org Subject: Re: [Users] Pxeboot On 02/08/2012 05:03 PM, Mike Burns wrote: On Wed, 2012-02-08 at 13:46 -0800, Li, David wrote: Mike, If I understand this correctly, today I should be able to pxeboot and nfs mount the root fs from a remote server. Apart from setting up the pxe stuff, I 'd have to populate the ovirt node root fs on the server - perhaps steal it from a disk install. In other words I am concerned about the point from which the kernel starts to execute /init script (in the initramfs) to the point /init is able to mount the final root fs from a remote server. No, there is no way to set this up currently in ovirt-node. You could install using a remote iscsi lun if you have a hardware iscsi HBA, but there isn't a way to mount a remote nfs share as the root fs. Supporting a remote NFS share as the root fs isn't even something that requested as an RFE at this point or on the roadmap as far as I'm aware. It sounds like what you're really looking for is a shared root fs that multiple hosts could use. This is something that we will probably look into eventually, but it's not on the immediate roadmap. Given that the rootfs of oVirt Node is fairly small and in a truly stateless environment would just run out of system RAM, there's no real reason to try to do a shared NFS based rootfs. It's an unnecessary complication I think, if the end goal is to move to truly stateless. For larger systems where the rootfs is on the order of GB's, shared root may make more sense. ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
