Description of problem: RHV manager events spams with warning "Used memory of host hostname [98%] exceeded defined threshold [95%]", for hosts having huge memory. e.g. for host having 1.5 TB memory, 98% means approximately 30 GB memory still available so no need to report this host but there is no mechanism to skip particular host/cluster so that it will not be spammed in events. Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1. Check events tab in rhv environment having hosts with memory in TBs in one cluster and having hosts with memory in GBs in another cluster. 2. 3. Actual results: Hosts with memory in TBs spam events. Expected results: Set different memory thresholds for different clusters/hypervisors. Additional info: Currently adjusting memory threshold percentage by changing LogMaxPhysicalMemoryUsedThresholdInPercentage is applicable for every host.
The threshold can be changed using: engine-config -s LogMaxPhysicalMemoryUsedThresholdInPercentage=<VALUE> where <VALUE> is the percentage, when above message is logged. By default it's set to 95%
(In reply to Martin Perina from comment #2) > The threshold can be changed using: > > engine-config -s LogMaxPhysicalMemoryUsedThresholdInPercentage=<VALUE> > > where <VALUE> is the percentage, when above message is logged. By default > it's set to 95% The request is: 1. Also in absolute numbers 2. Per cluster I think the 1st request is far more important than the first 2nd one.
Ravi, shouldn't be those new LogMaxMemoryUsedThresholdType and LogMaxMemoryUsedThreshold be visible in engine-config? # engine-config -a | grep MemoryUsed LogMaxPhysicalMemoryUsedThresholdInPercentage: 95 version: general [root@10-37-140-204 ~]# yum list ovirt-engine Last metadata expiration check: 0:13:06 ago on Thu 20 Sep 2018 08:05:54 PM CEST. Installed Packages ovirt-engine.noarch 4.3.0-0.0.master.20180919131647.git219e0985a3b.fc28 @ovirt-master-snapshot
(In reply to Pavol Brilla from comment #7) > Ravi, > > shouldn't be those new LogMaxMemoryUsedThresholdType and > LogMaxMemoryUsedThreshold be visible in engine-config? > > > # engine-config -a | grep MemoryUsed > LogMaxPhysicalMemoryUsedThresholdInPercentage: 95 version: general > [root@10-37-140-204 ~]# yum list ovirt-engine > Last metadata expiration check: 0:13:06 ago on Thu 20 Sep 2018 08:05:54 PM > CEST. > Installed Packages > ovirt-engine.noarch > 4.3.0-0.0.master.20180919131647.git219e0985a3b.fc28 > @ovirt-master-snapshot No, engine-config still contains the default in percentage with 95 as value. But users can change that value for each cluster when you create new or update existing cluster.
UI elements found, will test further with host to check correct triggers
Verified on ovirt-engine-4.3.0-0.8.rc2.el7.noarch
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-2019:1085