I'm always amazed by how the most (otherwise) elegant and robust solutions can fail because of some silly configuration that should handled in a very different way.
* * *ASB* *http://XeeMe.com/AndrewBaker* *Harnessing the Advantages of Technology for the SMB market… * On Sat, Oct 29, 2011 at 2:31 AM, David Lum <[email protected]> wrote: > Remotely working on a %nightjob% client tonight, both a VM and > corresponding host unexpectedly drop my LogMeIn connection. I see from > LogMeIn that other systems in that room are online, so I know it wasn’t the > circuit that dropped. Oh joy, I get to drive in (thankfully a short 20 min > drive).**** > > ** ** > > I get onsite and the host server is halted at the POST screen for the eSATA > RAID controller, and the eSATA RAID controller reports a degraded disk on > one of the two volumes. Power everything off, pull the drives, > disconnect/reconnect the cables, etc. Power it back up and everything shows > good.**** > > ** ** > > So the host comes up (YAY ½ way there! Well…) and I log in and watch for > the VM to start…it gets to 50% then stops, and after 15 minutes (and you > know how long 15 minutes is when you’re waiting for a **VERY** critical > server to come up don’tcha?) the VM goes back to “stopped”. **** > > ** ** > > As I do full volume backups nightly to the eSATA I’m not too worried yet, > but even recovering to that this client would lose a day of work (Internet > backups start at 7pm, servers went offline at 5:13pm). A cursory look at the > event logs shows nothing exciting, so I change the VM “autostart” from 60 > seconds after host OS to 500 seconds and then reboot the host.**** > > ** ** > > No change. Joy.**** > > ** ** > > Thinking maybe it’s an issue on the host I pull a two week old DISK2VHD > file that was handier than the backups, I create a new VM on the host and > use this VHD. That VM fires up just fine, but it makes me wonder if I can > just create a new VM and point to the existing disk files for this critical > server. I file that away for plan B.**** > > ** ** > > I hit the event logs again, I went through both system and app logs for the > timeframe including 30 mins on either side of the start failures (and you > know I tried to start that VM more than just those two times…). Somehow I > stumbled upon one of Windows 2008’s 1 zillion new logs, under Windows > logs\Applicaitons and Services logs\Microsoft\Windows\Hyper-V Worker and I > found my golden nugget:**** > > ** ** > > Log Name: Microsoft-Windows-Hyper-V-Worker-Admin**** > > Source: Microsoft-Windows-Hyper-V-Worker**** > > Date: 10/28/2011 7:02:45 PM**** > > Event ID: 12140**** > > Task Category: None**** > > Level: Error**** > > Keywords: **** > > User: NETWORK SERVICE**** > > Computer: Host4.thehosed.one.local**** > > Description:**** > > 'thehosed.one': Failed to open attachment > '\\192.168.116.249\Inst-server\Windows 2008 > R2\SW_DVD5_Windows_Svr_DC_EE_SE_Web_2008R2_64-bit_English_X15-59754.ISO'. > Error: 'The specified network name is no longer available.' (0x80070040). > (Virtual machine 97527135-A765-4700-AF66-C6FE2143391D)**** > > Event Xml:**** > > ** ** > > Google-Fu then returned a thread to me where someone else was having the > same issue because about a VM not starting and it turned out to be a CD-ROM > driver issue. Was the VM was failing to start because I had the CD-ROM > mapped to a network location that was no longer valid? ARE YOU KIDDING ME? > Go into VM settings, remove the CD-ROM from the config and boot the VM. > Presto! Took me just over four hours to find the necessary 2-second config > change…**** > > ** ** > > I charge 1.5x my normal hourly rate to break my routine and drive onsite, > somehow I think just one hour is fairhere – sometimes the lesson and the > relief that there was zero data loss for the client is reward enough!**** > > *David Lum* > Systems Engineer // NWEATM > Office 503.548.5229 //* *Cell (voice/text) 503.267.9764**** > > ** ** > > ~ Finally, powerful endpoint security that ISN'T a resource hog! ~ > ~ <http://www.sunbeltsoftware.com/Business/VIPRE-Enterprise/> ~ > > --- > To manage subscriptions click here: > http://lyris.sunbelt-software.com/read/my_forums/ > or send an email to [email protected] > with the body: unsubscribe ntsysadmin > ~ Finally, powerful endpoint security that ISN'T a resource hog! ~ ~ <http://www.sunbeltsoftware.com/Business/VIPRE-Enterprise/> ~ --- To manage subscriptions click here: http://lyris.sunbelt-software.com/read/my_forums/ or send an email to [email protected] with the body: unsubscribe ntsysadmin
