Description of problem: adding RHGS 3.1.3 and 3.2 nodes to RHV-Manager 4.1 throws an error ""Failed to install Host Host1. Failed to execute stage 'Package installation': Package collectd cannot be found." Version-Release number of selected component (if applicable): "Failed to install Host Host1. Failed to execute stage 'Package installation': Package collectd cannot be found." How reproducible: Always Steps to Reproduce: 1. Install RHEV-Manager 4.1 2. Create a dc and cluster with compatibility version 3.6 3. Now try adding RHGS 3.1.3 node and RHGS 3.1.2 node Actual results: Addition of both fails with error "Failed to install Host Host1. Failed to execute stage 'Package installation': Package collectd cannot be found." Expected results: 3.1.3 and 3.1.2 nodes should be added successfully. Additional info: 2017-04-07 05:59:54 DEBUG otopi.plugins.otopi.packagers.yumpackager yumpackager.verbose:76 Yum queue package collectd for install/update 2017-04-07 05:59:54 ERROR otopi.plugins.otopi.packagers.yumpackager yumpackager.error:85 Yum Cannot queue package collectd: Package collectd cannot be found 2017-04-07 05:59:54 DEBUG otopi.context context._executeMethod:142 method exception Traceback (most recent call last): File "/tmp/ovirt-Y71upa1y6R/pythonlib/otopi/context.py", line 132, in _executeMethod method['method']() File "/tmp/ovirt-Y71upa1y6R/otopi-plugins/ovirt-host-deploy/collectd/packages.py", line 53, in _packages 'collectd-write_http', File "/tmp/ovirt-Y71upa1y6R/otopi-plugins/otopi/packagers/yumpackager.py", line 307, in installUpdate ignoreErrors=ignoreErrors File "/tmp/ovirt-Y71upa1y6R/pythonlib/otopi/miniyum.py", line 883, in installUpdate **kwargs File "/tmp/ovirt-Y71upa1y6R/pythonlib/otopi/miniyum.py", line 500, in _queue package=package, RuntimeError: Package collectd cannot be found
ovirt-host-deploy log can be found in the link below: ======================================================= http://rhsqe-repo.lab.eng.blr.redhat.com/sosreports/1440111/
Is collectd a required package? Is there a way to not require collectd for RHGS nodes as no metrics are collected from gluster?
(In reply to Sahina Bose from comment #2) > Is collectd a required package? yes > Is there a way to not require collectd for > RHGS nodes as no metrics are collected from gluster? No, nothing I'm aware of. Redirecting to Shirly for metrics team evaluation.
Does it happen with 3.2?
(In reply to Yaniv Kaul from comment #4) > Does it happen with 3.2? By 3.2 did you mean RHGS 3.2 ? If yes, it does happen.
Are older nodes expected to have collectd packages when adding to a 3.6 cluster? This seems to break backward compatibility?
(In reply to Sahina Bose from comment #6) > Are older nodes expected to have collectd packages when adding to a 3.6 > cluster? No. > This seems to break backward compatibility? BTW, I'm quite sure you cannot add old nodes to RHV as well.
What's the final decision on this? - Are we making the collectd packages optional when being added to 3.6 cluster? - Adding collectd packages in RHGS optional
Alok?
(In reply to Sahina Bose from comment #8) > What's the final decision on this? > - Are we making the collectd packages optional when being added to 3.6 > cluster? > - Adding collectd packages in RHGS optional We are looking to see what's the complexity of being flexible, per cluster version, of adding or not the collectd. This is tracked in a different bug - need to find it.
Seems like we decided to fix the missing packages issue, see bug 1443508. Can we close current as duplicate?
Closing as duplicate of bug #1443508 *** This bug has been marked as a duplicate of bug 1443508 ***
Hi sandro, I understand that bug fix might be same for both the bugs bug #1443508 and 1440111. But i would still like to keep this bug open as this needs to be tested and verified. Can we reopen this bug or would some one who is verifying the bug #1443508 will cover this flow as well? can you please clarify? Thanks kasturi
I personally think they are the same bug. You can add a comment there asking QE to verify it also with/for RHGS. I don't mind also reopening current, moving it to MODIFIED and set it to depend on 1443508, if you think that's better.
(In reply to Yedidyah Bar David from comment #14) > I personally think they are the same bug. You can add a comment there asking > QE to verify it also with/for RHGS. I don't mind also reopening current, > moving it to MODIFIED and set it to depend on 1443508, if you think that's > better. second options looks good to me.
With the latest ovirt-host package made available for RHGS el7 based systems, the RHGS servers could be added 4.1 compatibility cluster and managed
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. https://access.redhat.com/errata/RHEA-2018:1512
BZ<2>Jira Resync