This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

VMs With More Allocated Space Than Used

Hey folks,

In VMan, in the "All Alerts" dashboard widget, there are always 5ish alerts flagged on topics more related to optimization than actual issues. While this awareness is all fine and dandy, the perfectionist in me really doesn't like having near-permanent alerts active. To me, if an alerted state is constant/acceptable/normal, it isn't an alert. That said, I do want to be able to see the information contained in the alert, so I'm averse to disabling it (it just would be nice if it was in an "Optimization" section/dashboard, but I digress...).

Anyways, the alert in question is "VMs With More Allocated Space Than Used". My question to the community is: How do you address this alert in your environment?

I'm primarily a Windows shop, running vSphere 5 (Enterprise Plus) w/ 3PAR storage and enough wiggle room to svMotion stuff around, if it is worth the effort. I've already adjusted the alert to assess on 10GB difference, rather than 1GB, because virtually ever guest of mine has that much excess due to Windows patching, etc.

In your realm, do you use sdelete.exe to zero the space and then Storage vMotion to re-thin? Or do you use a 3rd party utility? Or...

Let me know! It'd be great to have actionable steps associated with these alerts so we can all get to states where the "All Alerts" widget is green.

~Chris

  • I know this doesn't answer your question, but from my experience those alerts don't mean much anyway - I feel that they're only useful in report form, not so much in a "check every XX hours and email me" kind of way.

    In fact, I disabled those alerts altogether after significantly bumping up the amount of array-based deduplicated storage available to my VM guests...

    Just wanted to offer another way of looking at things...  :-)

  • I agree, especially about optimization data being shown as an alert.  WIth the built-in alerts, I think you really have to review them and add/change/delete based on your desired operational standards.  For example, the alert you mentioned (Over-allocated Space), it unfortunately includes all disks (e.g. RDMs) in it's summation, not just thin disks.  So my file server shows reclaimable space of 230+ GB, even though the thin-provisioned C drive is only 60 GB.  That extra "space" comes from my file share LUNs.  So I have to adjust it to exclude these RDMs.

    I've made similar adjustments on about 6 other alerts and continue to think over the remaining defaults.

    But as you said, the truly best thing to have in vMan would be a widget that can show "Best Practice" or "Optimization" items, separate from real, hair-starting-to-smoke alerts.


  • Thanks, guys. Yeah, the VMan alerts definitely call for a good amount of tweaking and even disabling on a few, depending on your environment.

    We'll see if anyone else out there has a favorite or magic recipe for reducing the count, especially of the automated variety emoticons_happy.png.