The Joys of Virtualisation

Discussion in 'Virtual and Cloud Computing' started by onoski, Nov 25, 2010.

  1. onoski

    onoski Terabyte Poster

    3,120
    51
    154
    We have a VMware virtual environment setup using ESX hosted via virtual centre servers. I had a database server on one of the ESX host that was getting packet drops and performance degradation for no apparent reason as the CPU, RAM and resource allocation was reasonably set to above average.

    The funny thing was that our ESX host are ten in total and this server was the only server on one host so couldn't quite figure out why it was down etc. The network link is 1GB and so far has faired well amongst two sites and replicating vice versa.

    The farm is also configured with DRM, distributed resource management, so don't know why this didn't kick in to tackle the above stated issue. Again, I would research and look into this and update as appropriate.

    Finally, after much VMware forum searches and research decided to move the VM to another host by migrating the VM to a different ESX host.

    To my surprise after carrying out the VM migration to another host that even has several other VM's, the above issues were eradicated.

    I still need to find out if there is sometime wrong with the host that the VM was originally hosted on.

    However, after looking at the tasks and events and system logs there was no logs as to why the issues were surfacing.

    I am still learning a lot and trying to get my head around storage, vdisk etc. Well would be going on an official troubleshooting course on Dec, 14th to 17th Dec, 2010 so would ask the lecturer as much questions as possible about the above etc.

    Well just glad a live server can be down for under ten minutes and back online again in almost less that five minutes. The joys of virtualisation I suppose, so can't complain:D

    Have you experienced a similar scenario? If you did how did you go about fixing or resolving it.
     
    Certifications: MCSE: 2003, MCSA: 2003 Messaging, MCP, HNC BIT, ITIL Fdn V3, SDI Fdn, VCP 4 & VCP 5
    WIP: MCTS:70-236, PowerShell
  2. danielno8

    danielno8 Gigabyte Poster

    1,306
    49
    92
    Have you checked the link between the ESX server and the switchport it is connected to? Duplex mismatch would cause packet loss. Also, the fact you moved the VM and the problem went away shows it to be something with the ESX host. You say you didn't expect it to fix it but that's probably the quickest way to narrow the issue down.
     
    Certifications: CCENT, CCNA
    WIP: CCNP
  3. SimonD
    Honorary Member

    SimonD Terabyte Poster

    3,681
    440
    199
    Just a couple of things for you, it's either DRS (Distributed Resource Scheduling) or DPM (Distributed Power Management).

    If the ESX host is now guestless then I would suggest running vendor diagnostics on it (CPU, RAM etc), I would perhaps also look at running ESXTOP from the service console to see if that's showing anything.
     
    Certifications: CNA | CNE | CCNA | MCP | MCP+I | MCSE NT4 | MCSA 2003 | Security+ | MCSA:S 2003 | MCSE:S 2003 | MCTS:SCCM 2007 | MCTS:Win 7 | MCITP:EDA7 | MCITP:SA | MCITP:EA | MCTS:Hyper-V | VCP 4 | ITIL v3 Foundation | VCP 5 DCV | VCP 5 Cloud | VCP6 NV | VCP6 DCV | VCAP 5.5 DCA
  4. onoski

    onoski Terabyte Poster

    3,120
    51
    154
    Thanks for the feedback and comments SimonD and Daniel as I would look into your suggestions.
     
    Certifications: MCSE: 2003, MCSA: 2003 Messaging, MCP, HNC BIT, ITIL Fdn V3, SDI Fdn, VCP 4 & VCP 5
    WIP: MCTS:70-236, PowerShell
  5. dales

    dales Terabyte Poster

    2,005
    51
    142
    Onoski, If I'm reading your post right DRM didn't kick in because there were no contention issues on a host, the performance degredation was not because the esx host had too many resources being chewed up only because there was something wrong with it the vm performance appeared to drop. So DRM had nothing to do.
     
    Certifications: vExpert 2014+2015+2016,VCP-DT,CCE-V, CCE-AD, CCP-AD, CCEE, CCAA XenApp, CCA Netscaler, XenApp 6.5, XenDesktop 5 & Xenserver 6,VCP3+5,VTSP,MCSA MCDST MCP A+ ITIL F
    WIP: Nothing
  6. onoski

    onoski Terabyte Poster

    3,120
    51
    154

    Hi Dales, yes you're correct as I too noticed it was nothing to do with DRM but still puzzled as to why the issues.

    I would look into DPM as that's an area I've investigated yet. Thanks for the comments:)
     
    Certifications: MCSE: 2003, MCSA: 2003 Messaging, MCP, HNC BIT, ITIL Fdn V3, SDI Fdn, VCP 4 & VCP 5
    WIP: MCTS:70-236, PowerShell
  7. dales

    dales Terabyte Poster

    2,005
    51
    142
    DPM again would not cause performance issues it just monitors all the hosts for load and moves vm's around when it calculates that it can put a one or more host's into standby. I guess you must have seen this video but I'll stick it up anyway http://www.youtube.com/watch?v=7CbRS0GGuNc its just shows DPM's awesomeness even if the video is a couple of versions out of date now.
     
    Certifications: vExpert 2014+2015+2016,VCP-DT,CCE-V, CCE-AD, CCP-AD, CCEE, CCAA XenApp, CCA Netscaler, XenApp 6.5, XenDesktop 5 & Xenserver 6,VCP3+5,VTSP,MCSA MCDST MCP A+ ITIL F
    WIP: Nothing
  8. onoski

    onoski Terabyte Poster

    3,120
    51
    154


    Thanks for your help Dales, much appreciated:)
     
    Certifications: MCSE: 2003, MCSA: 2003 Messaging, MCP, HNC BIT, ITIL Fdn V3, SDI Fdn, VCP 4 & VCP 5
    WIP: MCTS:70-236, PowerShell
  9. Rob1234

    Rob1234 Megabyte Poster Forum Leader

    940
    127
    114
    Could be a profile issue :dry
     
    Certifications: A few.

Share This Page

Loading...
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.