Red Hat Bugzilla – Bug 1165269
[RHSC] Adding a host to Console using FQDN fails, after it is once added using IP address and then removed.
Last modified: 2015-07-29 01:26:50 EDT
Created attachment 958666 [details] engine logs Description of problem: ------------------------ A host is added to a cluster managed by the Console using the IP address of the host and after it comes up, it is removed from the cluster. The same host is then imported to the Console by importing it into a cluster. This time the FQDN of the host is given as the address. The host fails to come up and remains in the non-responsive state. Following is from in engine logs - ----------------------------------- <snip> 2014-11-18 15:20:30,809 ERROR [org.ovirt.engine.core.bll.InstallVdsCommand] (org.ovirt.thread.pool-4-thread-3) [4a6edfb8] Host installation failed for host 038f10c5-76b 7-4af7-a431-2cc851495066, dhcp37-100.lab.eng.blr.redhat.com.: org.ovirt.engine.core.bll.InstallVdsCommand$VdsInstallException: Network error during communication with the host </snip> The following log is seen repeatedly in the engine.log file - --------------------------------------------------------------- 2014-11-18 15:32:07,251 ERROR [org.ovirt.engine.core.vdsbroker.vdsbroker.GetCapabilitiesVDSCommand] (DefaultQuartzScheduler_Worker-57) Command GetCapabilitiesVDSCommand(HostName = dhcp37-100.lab.eng.blr.redhat.com, HostId = 038f10c5-76b7-4af7-a431-2cc851495066, vds=Host[dhcp37-100.lab.eng.blr.redhat.com,038f10c5-76b7-4af7-a431-2cc851495066]) execution failed. Exception: VDSNetworkException: javax.net.ssl.SSLHandshakeException: server certificate change is restrictedduring renegotiation Version-Release number of selected component (if applicable): -------------------------------------------------------------- rhsc-3.0.3-1.17.el6rhs.noarch How reproducible: ------------------ Consistently. Steps to Reproduce: -------------------- 1. Create a cluster via Console and a node to it using the IP address in the 'address' field of the New Host dialog. 2. After the host comes up, move the host to maintenance and remove it from the cluster. 3. Create a new cluster and import this host as an existing cluster. Provide the FQDN of the host in the address field this time. Actual results: ---------------- The host does not come up after installation and ends up in non-responsive state. The events log shows - Host <hostname> installation failed. Network error during communication with the host. Expected results: ------------------ The host is expected to come up. Additional info: ------------------ Find logs attached.
Created attachment 958667 [details] host-deploy-FQDN
Created attachment 958668 [details] host-deploy-IP
Created attachment 958669 [details] vdsm logs
Re-installing the non-responsive host does not cause it to be up. Removing and adding again using FQDN also results in the host being non-responsive. Removing the host and adding using IP address works.
There might be an issue in generating /etc/pki/vdsm/certs/vdsmcert.pem when the host is re-added using FQDN.
Hi Kanagaraj, Can you please review the edited doc text for the known issue and sign off?
Corrected a typo
looks good.
Verified the below scenarios against ovirt-master and ovirt-engine-3.5-gluster branches and it works fine - SCENARIO-1 1. Created a cluster say cluster-1 and added the node-1 (IP: x1.x1.x1.x1) to the cluster 2. Once node-1 git UP, brought it to maintenance mode and removed from the cluster 3. Created a new cluster say cluster-2 and added the same node-1, to the cluster-2 but this time using FQDN (e.q. <name>.<domain>...) 4. Node got added successfully to the cluster-2 SCENARIO-2 1. Created a cluster say cluster-1 and added the node-1 (IP: x1.x1.x1.x1) to the cluster 2. Once node-1 git UP, brought it to maintenance mode and removed from the cluster 3. Again added the same node-1, to the cluster-1 but this time using FQDN (e.q. <name>.<domain>...) 4. Node got added successfully to the cluster-1
Please provide the fixed in version.
Verified and works fine with build rhsc-3.1.0-0.57.master.el6.noarch. Performed following steps to verify the bug: 1. Created a cluster say cluster-1 and added the node-1 (IP: x1.x1.x1.x1) to the cluster 2. Once node-1 is UP, brought it to maintenance mode and removed from the cluster 3. Again added the same node-1, to the cluster-1 but this time using FQDN (e.q. <name>.<domain>...) 4. Node got added successfully to the cluster-1
Shubhendu, Could you review and sign-off the edited doc text.
doc-text edited to correct names as "Red Hat Gluster Storage" and "Red Hat Gluster Storage Console". Looks fine now.
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://rhn.redhat.com/errata/RHEA-2015-1494.html