Created attachment 919404 [details] engine.log Description of problem: Can't add RHEL7.0 host to av10.3 engine via UI. Version-Release number of selected component (if applicable): rhevm-3.4.1-0.30.el6ev.noarch How reproducible: 100% Steps to Reproduce: 1.Image to fresh RHEL7 two hosts and edit their repo files accordingly to https://mojo.redhat.com/docs/DOC-949006 2.Try to attach to av10.3 engine both hosts via UI. 3.Receive errors via UI. Actual results: Unable to add RHEL7.0 hosts to engine via UI. Expected results: Hosts have to be added to setup via av10.3 engine's UI. Additional info: log files attached.
There seem to be quite a few host installation failures in the engine log, earlier ones having to do with management network, but presumably this was fixed because later failures seem unrelated. Setting whiteboard to infra. Nikolai, several questions: 1. What happened with the earlier failures and how was that fixed? 2. You mentioned that installation via the GUI fails, does this mean it succeeds via REST? 3. Could you please attach the relevant host deploy log? Usually when an installation fails, the audit log contains entries that point you to an installation log file.
Created attachment 919563 [details] Both engine's log and host deployment log.
(In reply to Lior Vernia from comment #3) > There seem to be quite a few host installation failures in the engine log, > earlier ones having to do with management network, but presumably this was > fixed because later failures seem unrelated. Setting whiteboard to infra. > > Nikolai, several questions: > 1. What happened with the earlier failures and how was that fixed? > 2. You mentioned that installation via the GUI fails, does this mean it > succeeds via REST? > 3. Could you please attach the relevant host deploy log? Usually when an > installation fails, the audit log contains entries that point you to an > installation log file. 1.They weren't fixed, I did fresh image installation on both hosts and after tried to add both hosts to engine without success via UI. 2.Not relevant to my test cases steps, I'm doing only manual testing, REST isn't part of the test steps, user not working with it using UI. 3.Attached both engine.log and host deployment log.
looking in the host deploy logs, file /usr/lib64/python2.7/site-packages/vdsm/tool/vdsm-id.pyo conflicts between attempted installs of vdsm3-python-4.14.6-0.el7ev.x86_64 and vdsm-python-4.14.11-1.el7ev.x86_64 file /usr/lib64/python2.7/site-packages/vdsm/utils.pyc conflicts between attempted installs of vdsm3-python-4.14.6-0.el7ev.x86_64 and vdsm-python-4.14.11-1.el7ev.x86_64 file /usr/lib64/python2.7/site-packages/vdsm/utils.pyo conflicts between attempted installs of vdsm3-python-4.14.6-0.el7ev.x86_64 and vdsm-python-4.14.11-1.el7ev.x86_64 file /usr/lib64/python2.7/site-packages/vdsm/vdscli.py conflicts between attempted installs of vdsm3-python-4.14.6-0.el7ev.x86_64 and vdsm-python-4.14.11-1.el7ev.x86_64 file /usr/lib64/python2.7/site-packages/vdsm/vdscli.pyc conflicts between attempted installs of vdsm3-python-4.14.6-0.el7ev.x86_64 and vdsm-python-4.14.11-1.el7ev.x86_64 file /usr/lib64/python2.7/site-packages/vdsm/vdscli.pyo conflicts between attempted installs of vdsm3-python-4.14.6-0.el7ev.x86_64 and vdsm-python-4.14.11-1.el7ev.x86_64 file /usr/libexec/vdsm/libvirt_configure.sh conflicts between attempted installs of vdsm3-python-4.14.6-0.el7ev.x86_64 and vdsm-python-4.14.11-1.el7ev.x86_64 file /usr/lib/python2.7/site-packages/zombiereaper/__init__.pyc conflicts between attempted installs of vdsm3-python-zombiereaper-4.14.6-0.el7ev.noarch and vdsm-python-zombiereaper-4.14.11-1.el7ev.noarch file /usr/lib/python2.7/site-packages/zombiereaper/__init__.pyo conflicts between attempted installs of vdsm3-python-zombiereaper-4.14.6-0.el7ev.noarch and vdsm-python-zombiereaper-4.14.11-1.el7ev.noarch 2014-07-21 11:24:12 DEBUG otopi.transaction transaction.abort:131 aborting 'Yum Transaction' I don't know where the vdsm3 package is comming from ... it looks like this image was an experimental one done a long time ago. Please reinstall from scratch a new RHEL 7 host and retest adding it.
only repo i can think of is baseurl=http://download.lab.bos.redhat.com/rel-eng/repos/rhevm-3.4-rhel-7-candidate/x86_64/ try removing this repo and use only av10.3/el7 and see if it works.
(In reply to Barak from comment #6) > looking in the host deploy logs, > > file /usr/lib64/python2.7/site-packages/vdsm/tool/vdsm-id.pyo conflicts > between attempted installs of vdsm3-python-4.14.6-0.el7ev.x86_64 and > vdsm-python-4.14.11-1.el7ev.x86_64 > file /usr/lib64/python2.7/site-packages/vdsm/utils.pyc conflicts between > attempted installs of vdsm3-python-4.14.6-0.el7ev.x86_64 and > vdsm-python-4.14.11-1.el7ev.x86_64 > file /usr/lib64/python2.7/site-packages/vdsm/utils.pyo conflicts between > attempted installs of vdsm3-python-4.14.6-0.el7ev.x86_64 and > vdsm-python-4.14.11-1.el7ev.x86_64 > file /usr/lib64/python2.7/site-packages/vdsm/vdscli.py conflicts between > attempted installs of vdsm3-python-4.14.6-0.el7ev.x86_64 and > vdsm-python-4.14.11-1.el7ev.x86_64 > file /usr/lib64/python2.7/site-packages/vdsm/vdscli.pyc conflicts between > attempted installs of vdsm3-python-4.14.6-0.el7ev.x86_64 and > vdsm-python-4.14.11-1.el7ev.x86_64 > file /usr/lib64/python2.7/site-packages/vdsm/vdscli.pyo conflicts between > attempted installs of vdsm3-python-4.14.6-0.el7ev.x86_64 and > vdsm-python-4.14.11-1.el7ev.x86_64 > file /usr/libexec/vdsm/libvirt_configure.sh conflicts between attempted > installs of vdsm3-python-4.14.6-0.el7ev.x86_64 and > vdsm-python-4.14.11-1.el7ev.x86_64 > file /usr/lib/python2.7/site-packages/zombiereaper/__init__.pyc conflicts > between attempted installs of > vdsm3-python-zombiereaper-4.14.6-0.el7ev.noarch and > vdsm-python-zombiereaper-4.14.11-1.el7ev.noarch > file /usr/lib/python2.7/site-packages/zombiereaper/__init__.pyo conflicts > between attempted installs of > vdsm3-python-zombiereaper-4.14.6-0.el7ev.noarch and > vdsm-python-zombiereaper-4.14.11-1.el7ev.noarch > > 2014-07-21 11:24:12 DEBUG otopi.transaction transaction.abort:131 aborting > 'Yum Transaction' > > > I don't know where the vdsm3 package is comming from ... it looks like this > image was an experimental one done a long time ago. > > Please reinstall from scratch a new RHEL 7 host and retest adding it. No connectivity, it all worked just fine with 3.3.1, no connection to image, they all work fine.
Created attachment 919635 [details] logs from engine
This time the host deploy failed on systemctl starting the service which is entirely different from the initial log. I have asked nikolai to install from RHEL 7 GA bits (the version of current repo is not clear ... looks like pre GA bits) and do another deploy.
I also retested on alma04 now without http://download.lab.bos.redhat.com/rel-eng/repos/rhevm-3.4-rhel-7-candidate/x86_64/ on fresh image, works for me, the problem was in wrong repo.
so this can be closed NOTABUG?
(In reply to Itamar Heim from comment #14) > so this can be closed NOTABUG? We may close the bug, but you have to edit the HOW-TO accordingly, many people will fail the whole procedure the same way I failed, may be there is also a hidden bug there, as if I'll enable the broken repository, system still fails to add the host with weird errors... Please decide on this and make changes to How-To anyway.
mojo was updated. https://mojo.redhat.com/docs/DOC-962690 i removed the bad repo until rel-eng will fix it. for now people can use the pkg available from latest_av_dev/el7 repo.
mojo was updated. https://mojo.redhat.com/docs/DOC-962690 i removed the bad repo until rel-eng will fix it. for now people can use the pkg available from latest_av_dev/el7 repo. i also opened a ticket to rel-eng to fix it.:https://engineering.redhat.com/rt/Ticket/Display.html?id=307273 closing the bug.