Currently the engine supports external fencing devices that are used in 2 different use cases: a. Storage/Clustering related b. None-responsive host This bug is about flow b (none responsive) and is a result of issues found and discussed on Bug 924801. The problem: Host becomes none-responsive due to an unexpected problem (Causing the VDSM to not respond over time) but the VMs themselves continue to live and be accessible. A simple restart of VDSM can solve the issue Current status: The host will be fenced by external fencing device and all the VMs will die (some might be sensitive) The solution: Add a preliminary stage to none-responsive fencing flow to try and restart the VDSM over SSH before doing the actual external fencing. Failing to perform this preliminary stage should mean continue to the external fencing, succeeding doing the restart should give the host additional grace time to recover, but if not recovered will also lead to the external fencing. Open Issues: - what should we implement for hosts without an external fencing device ? should we add a host restart vis SSH ?
I'd consider a checkbox of 'try to soft fence via ssh first' to power management or something like that.
*** Bug 590370 has been marked as a duplicate of this bug. ***
Open Issues: - what should we implement for hosts without an external fencing device ? should we add a host restart vis SSH ? What about this issue?
Please look at following URL to see how exactly SSH Soft Fencing is supposed to work in oVirt 3.3: http://www.ovirt.org/Automatic_Fencing#Automatic_Fencing_in_oVirt_3.3 http://lists.ovirt.org/pipermail/engine-devel/2013-July/005080.html
Created attachment 770831 [details] engine.log
Tests under: rhevm - is4 host with power management - vdsm-4.10.2-19.0.el6ev host without power management - vdsm-4.11.0-89.git8bf916a.el6 No ssh fencing Via host without power management, after manually stopped vdsmd, host in state non-responding and not try start vdsmd via ssh. Via host with power management, host do fence but no commands with ssh in engine.log. See engine log for more details.
I've looked at source code for RHEVM is5 and SSH Soft Fencing is included. Please retest.
Tests under: rhevm - is5 host with power management - vdsm-4.11.0-121.git082925a.el6 host without power management - vdsm-4.11.0-121.git082925a.el6 Verified Via host without power management, after manually stopped vdsmd, host 1-2 minutes in connecting state and after this host in up state. Via host with power management, host do sshSoftFencing if ssh enabled, if not do fence.
This bug is currently attached to errata RHEA-2013:15231. If this change is not to be documented in the text for this errata please either remove it from the errata, set the requires_doc_text flag to minus (-), or leave a "Doc Text" value of "--no tech note required" if you do not have permission to alter the flag. Otherwise to aid in the development of relevant and accurate release documentation, please fill out the "Doc Text" field above with these four (4) pieces of information: * Cause: What actions or circumstances cause this bug to present. * Consequence: What happens when the bug presents. * Fix: What was done to fix the bug. * Result: What now happens when the actions or circumstances above occur. (NB: this is not the same as 'the bug doesn't present anymore') Once filled out, please set the "Doc Type" field to the appropriate value for the type of change made and submit your edits to the bug. For further details on the Cause, Consequence, Fix, Result format please refer to: https://bugzilla.redhat.com/page.cgi?id=fields.html#cf_release_notes Thanks in advance.
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/RHSA-2014-0038.html