Bug 1060524
| Summary: | rhev 3.3 does not work with vdsm 3.4 : Command PollVDS execution failed | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Virtualization Manager | Reporter: | Ohad Basan <obasan> |
| Component: | vdsm | Assignee: | Douglas Schilling Landgraf <dougsland> |
| Status: | CLOSED ERRATA | QA Contact: | sefi litmanovich <slitmano> |
| Severity: | high | Docs Contact: | |
| Priority: | urgent | ||
| Version: | 3.4.0 | CC: | aberezin, bazulay, danken, dougsland, eedri, gklein, iheim, lbopf, lpeer, obasan, pstehlik, talayan, yeylon |
| Target Milestone: | --- | Keywords: | AutomationBlocker |
| Target Release: | 3.4.0 | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | infra | ||
| Fixed In Version: | av1 | Doc Type: | Bug Fix |
| Doc Text: |
Previously, connecting a 3.3 Manager with a host running VDSM 4.14.1 would fail. Now, host deploys successfully.
|
Story Points: | --- |
| Clone Of: | Environment: | ||
| Last Closed: | 2014-06-09 13:28:53 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | Infra | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
|
Description
Ohad Basan
2014-02-02 11:16:17 UTC
something wrong with vdsm there, it restarts in a loop by getting SIGTERM, not sure why yet. is it reproducible or relates only to that specific host? Could you do `su - vdsm -s /bin/bash` and then run /usr/share/vdsm/vdsm to see why vdsm resets itself so often? BTW, is sysvinit-tools installed on your host? it just gets stuck like this: [root@cinteg07 ~]# su - vdsm -s /bin/bash -bash-4.1$ /usr/share/vdsm/vdsm (PID: 26271) I am the actual vdsm 4.14.1-21.git291df0d.el6 cinteg07.ci.lab.tlv.redhat.com (2.6.32-431.el6.x86_64) Run and protect: registerDomainStateChangeCallback(callbackFunc=<bound method clientIF.contEIOVms of <clientIF.clientIF instance at 0x284b320>>) Run and protect: registerDomainStateChangeCallback, Return response: None Starting up MOM Setting channels' timeout to 30 seconds. Starting VM channels listener thread. trying to connect libvirt and regarding to your question > yes, it is installed. Would you make sure that `service vdsmd stop` before running a new vdsm instance? When you run Vdsm manually, is it suddenly responsive to Engine's polling? I stopped vdsmd but it could be that supervdsm resumed it. how can I disable supervdsm? supervdsm does not restart vdsm. I do not understand your report then: with `service vdsmd start`, vdsm constantly crashes, and from the command line all is well? When you run Vdsm manually, is it suddenly responsive to Engine's polling? you get alot of:
libvir: XML-RPC error : authentication failed: Authorization requires authentication but no agent is available.
libvir: XML-RPC error : authentication failed: Authorization requires authentication but no agent is available.
libvir: XML-RPC error : authentication failed: Authorization requires authentication but no agent is available.
libvir: XML-RPC error : authentication failed: Authorization requires authentication but no agent is available.
libvir: XML-RPC error : authentication failed: Authorization requires authentication but no agent is available.
Vm's recovery failed
Traceback (most recent call last):
File "/usr/share/vdsm/clientIF.py", line 390, in _recoverExistingVms
caps.CpuTopology().cores())
File "/usr/share/vdsm/caps.py", line 127, in __init__
self._topology = _getCpuTopology(capabilities)
File "/usr/lib64/python2.6/site-packages/vdsm/utils.py", line 832, in __call__
value = self.func(*args)
File "/usr/share/vdsm/caps.py", line 147, in _getCpuTopology
capabilities = _getCapsXMLStr()
File "/usr/lib64/python2.6/site-packages/vdsm/utils.py", line 832, in __call__
value = self.func(*args)
File "/usr/share/vdsm/caps.py", line 141, in _getCapsXMLStr
return libvirtconnection.get().getCapabilities()
File "/usr/lib64/python2.6/site-packages/vdsm/libvirtconnection.py", line 146, in get
conn = utils.retry(libvirtOpenAuth, timeout=10, sleep=0.2)
File "/usr/lib64/python2.6/site-packages/vdsm/utils.py", line 949, in retry
return func()
File "/usr/lib64/python2.6/site-packages/libvirt.py", line 102, in openAuth
if ret is None:raise libvirtError('virConnectOpenAuth() failed')
libvirtError: authentication failed: Authorization requires authentication but no agent is available.
trying to connect libvirt
libvir: XML-RPC error : authentication failed: Authorization requires authentication but no agent is available.
familiar?
(In reply to Yaniv Bronhaim from comment #10) > you get alot of: > libvir: XML-RPC error : authentication failed: Authorization requires > authentication but no agent is available. > libvir: XML-RPC error : authentication failed: Authorization requires > authentication but no agent is available. > libvir: XML-RPC error : authentication failed: Authorization requires > authentication but no agent is available. > libvir: XML-RPC error : authentication failed: Authorization requires > authentication but no agent is available. > libvir: XML-RPC error : authentication failed: Authorization requires > authentication but no agent is available. > Vm's recovery failed > Traceback (most recent call last): > File "/usr/share/vdsm/clientIF.py", line 390, in _recoverExistingVms > caps.CpuTopology().cores()) > File "/usr/share/vdsm/caps.py", line 127, in __init__ > self._topology = _getCpuTopology(capabilities) > File "/usr/lib64/python2.6/site-packages/vdsm/utils.py", line 832, in > __call__ > value = self.func(*args) > File "/usr/share/vdsm/caps.py", line 147, in _getCpuTopology > capabilities = _getCapsXMLStr() > File "/usr/lib64/python2.6/site-packages/vdsm/utils.py", line 832, in > __call__ > value = self.func(*args) > File "/usr/share/vdsm/caps.py", line 141, in _getCapsXMLStr > return libvirtconnection.get().getCapabilities() > File "/usr/lib64/python2.6/site-packages/vdsm/libvirtconnection.py", line > 146, in get > conn = utils.retry(libvirtOpenAuth, timeout=10, sleep=0.2) > File "/usr/lib64/python2.6/site-packages/vdsm/utils.py", line 949, in retry > return func() > File "/usr/lib64/python2.6/site-packages/libvirt.py", line 102, in openAuth > if ret is None:raise libvirtError('virConnectOpenAuth() failed') > libvirtError: authentication failed: Authorization requires authentication > but no agent is available. > > trying to connect libvirt > libvir: XML-RPC error : authentication failed: Authorization requires > authentication but no agent is available. > > familiar? Hi, Can you please check if virsh tool works? Also, I see a report [1] about "Authorization requires authentication but no agent is available" although might be not related to this one. [1] https://bugzilla.redhat.com/show_bug.cgi?id=872166 Answering myself, virsh works:
# virsh
Welcome to virsh, the virtualization interactive terminal.
Type: 'help' for help with commands
'quit' to quit
virsh # list
Please enter your authentication name: vdsm@rhevh
Please enter your password:
Id Name State
----------------------------------------------------
virsh # list --all
Id Name State
----------------------------------------------------
virsh # net-list
Name State Autostart Persistent
--------------------------------------------------
default active no yes
vdsm-rhevm active yes yes
vdsm-virbr0 active yes yes
According with scenario looks like the polkit dir doesn't have any rule so vdsm will show us:
"trying to connect to libvirt"
# ls /var/lib/polkit-1/localauthority/10-vendor.d
#
# cat /etc/redhat-release
Red Hat Enterprise Linux Server release 6.5 (Santiago)
From our vdsm.spec:
=======================
# Required paths
%if 0%{?fedora} >= 18
%global _polkitdir %{_datadir}/polkit-1/rules.d
%else
%global _polkitdir %{_localstatedir}/lib/polkit-1/localauthority/10-vendor.d
%endif
and
%install
# Install the polkit for libvirt
%if 0%{?fedora} >= 18
install -Dm 0644 vdsm/vdsm-libvirt-access.rules \
%{buildroot}%{_polkitdir}/10-vdsm-libvirt-access.rules
%else
install -Dm 0644 vdsm/vdsm-libvirt-access.pkla \
%{buildroot}%{_polkitdir}/10-vdsm-libvirt-access.pkla
More evidences: # rpm -qa | grep -i vdsm vdsm-python-4.14.1-24.gitb4b06ab.el6.x86_64 vdsm-hook-fileinject-4.12.0-120.git83a430b.el6.noarch vdsm-xmlrpc-4.14.1-24.gitb4b06ab.el6.noarch vdsm-4.14.1-24.gitb4b06ab.el6.x86_64 vdsm-python-zombiereaper-4.14.1-24.gitb4b06ab.el6.noarch vdsm-cli-4.14.1-24.gitb4b06ab.el6.noarch # yum downloader vdsm # rpm2cpio ./vdsm-4.14.1-24.gitb4b06ab.el6.x86_64.rpm | cpio -div # cd vdsm* # find . -name *pkla* No polkit available in the package. - vdsm not starting Now: ========= - scp 10-vdsm-libvirt-access.pkla to host # cp 10-vdsm-libvirt-access.pkla /var/lib/polkit-1/localauthority/10-vendor.d - started vdsm manually # service vdsmd status VDS daemon server is running # vdsClient -s 0 getVdsCaps (works) # su - vdsm -s /bin/bash -bash-4.1$ /usr/share/vdsm/vdsm (PID: 17731) I am the actual vdsm 4.14.1-24.gitb4b06ab.el6 cinteg07.ci.lab.tlv.redhat.com (2.6.32-431.el6.x86_64) Run and protect: registerDomainStateChangeCallback(callbackFunc=<bound method clientIF.contEIOVms of <clientIF.clientIF instance at 0x2170320>>) Run and protect: registerDomainStateChangeCallback, Return response: None Starting up MOM Setting channels' timeout to 30 seconds. trying to connect libvirt (Not more blocking on this) Starting VM channels listener thread. still I don't understand how some managed to work with vdsm after the merge of the with_systemd patch [1].. its almost a year since then [1] http://gerrit.ovirt.org/#/c/12086 Verified on rhevm-3.4.0-0.3.master.el6ev.noarch,host deployment for host with vdsm-4.14.2-0.4.el6ev.x86_64.rpm works fine. Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHBA-2014-0504.html The needinfo request[s] on this closed bug have been removed as they have been unresolved for 1000 days |