Hide Forgot
Created attachment 776422 [details] ## Logs rhevm, vdsm, libvirt Description of problem: Command “vdsClient -s 0 getVdsStats” succeed run although /rhev/data-center/ folder is empty from links to Storage Pool and Storage Domain Version-Release number of selected component (if applicable): RHEVM 3.3 - IS6 environment: RHEVM: rhevm-3.3.0-0.9.master.el6ev.noarch VDSM: vdsm-4.12.0-rc1.12.git8ee6885.el6.x86_64 LIBVIRT: libvirt-0.10.2-18.el6_4.9.x86_64 QEMU & KVM: qemu-kvm-rhev-0.12.1.2-2.355.el6_4.5.x86_64 SANLOCK: sanlock-2.6-2.el6.x86_64 How reproducible: 100% Steps to Reproduce: Create iSCSI DC with 3 hosts Create 150 VM’s Power on all those VM’s Actual results: 1. All VM’s run only on one host, in my case it was on Host-02. 2. After I power off same VM’s and try run them again with changed parameter “Start Running On: Specific host_name“ (Host-01 & Host-03) action failed and VM continue run on same host as described in previous step (Host-02). After investigation we found following. 1. Following Commands work OK: vdsClient -s 0 getVdsCaps vdsClient -s 0 getVdsStats vdsClient -s 0 getVdsHardwareInfo vdsClient -s 0 getDeviceList 2. “Vdsmd” daemons in status up, but “SuperVdsmd” daemons down 3. There is no links under /rhev/data-center/ folder (folder /rhev/data-center/ empty) Expected results: 1. Command “vdsClient -s 0 getVdsStats” should failed if “SuperVdsmd” daemons down or failed run. 2. Command “vdsClient -s 0 getVdsStats” should failed there is no links under /rhev/data-center/ folder (folder /rhev/data-center/ empty) Impact on user: Failed run VM’s on specific host. Workaround: 1. Kill all “Vdsmd” service and run it again with verification that both “Vdsmd” and “SuperVdsmd” running 2. Manually enter in each VM configuration and changed parameter “Start Running On: Specific host_name“ Additional info: /var/log/ovirt-engine/engine.log /var/log/vdsm/vdsm.log
Created attachment 776423 [details] ## Logs superVdsm
*** Bug 989131 has been marked as a duplicate of this bug. ***
1. the issue with running the VM is solved by bug 960952 2. Yaniv, are you familiar with Supervdsmd not running issues?
How did you come up with: "Expected results: 1. Command “vdsClient -s 0 getVdsStats” should failed if “SuperVdsmd” daemons down or failed run." ? If an API call needs supervdsmd service to run and the service is down or after a crash, you will see an exception and restart of vdsm in the log.
I am closing that bug based on comment 4 (the storage part) and comment 5 (the infra part).
Created attachment 832646 [details] missing links logs I managed to reproduce, attaching logs
Elad, reproduction steps?
(In reply to Ayal Baron from comment #8) > Elad, reproduction steps? It happened when I blocked connection to storage server which one of the domains in the pool is located on (not the master)
(In reply to Elad from comment #9) > (In reply to Ayal Baron from comment #8) > > Elad, reproduction steps? > > It happened when I blocked connection to storage server which one of the > domains in the pool is located on (not the master) while it was not active, right? and then you activated it at some point? and it did not recreate the links?
(In reply to Ayal Baron from comment #10) > (In reply to Elad from comment #9) > > (In reply to Ayal Baron from comment #8) > > > Elad, reproduction steps? > > > > It happened when I blocked connection to storage server which one of the > > domains in the pool is located on (not the master) > while it was not active, right? > and then you activated it at some point? and it did not recreate the links? For focusing on one bug, as this bug is already closed, I answered to your question in https://bugzilla.redhat.com/show_bug.cgi?id=1026697
*** This bug has been marked as a duplicate of bug 1026697 ***