AzureStack TP3Refresh. Storage pool corrupted / missing. VMs bar DC01 now missing RRS feed

  • Question

  • My AzureStack Tp3 Refresh install is un-useable at present due to an issue with the Storage it would appear.

    Over night the VMs (with the exception of MAS-DC01) disappeared. Now this had happened before but this was different this time.

    the event log reported Event ID 129 LSI_SAS Reset to \device\raidport0 was issued

    Event id 153 Disk The IO Operation at logical addresss 0x4052x40 for disk 3 was retried

    there were multiple similar events following this

    then Event id 134 REFS. The file system was unable to write meta data to the media backing volume_1.A write failed with status the device is unresponsive. REFS will take the volume offline. It may be mounted again automatically.

    I've tried to look at the storage pools via server manager but this never loads, I just see "display will be complete when the server inventory has finished.

    I have reluctantly restarted the host but no change in behaviour.

    Anyone seen this issue before - any suggestion to fix what I believe is the problem with the underlying storage.

    Wednesday, June 7, 2017 2:01 PM


All replies

  • additional information.

    I've found the VMs which are in c:\clusterstorage\volume1\shares\SU_Infrastructure_1\1.0.170331.1

    But it leaves me with what is the best way to get these import back in again - I've had a similar issue before and when I imported the disks, the networking was not set up correctly . Importing the VMs again was not a supported scenario so I'm a bit stuck.

    Id rather not have to spend realistically another 2 days getting this reinstalled . (will I get this problem again?)

    Any suggestions , whether supported or not?

    Wednesday, June 7, 2017 3:10 PM
  • Its possible you have a disk issue that needs fixing.

    Once fixed you will likely have to redeploy your POC.



    Wednesday, June 7, 2017 11:19 PM
  • I can't find anything wrong with the disk following a reboot. the storage is all visible.

    I managed to recover all Vms bar one (MAS-ACS01) which si think was down to human error in the way I was recovering the VMs. still working through the recovery exercise but cannot get to the admim portal from mas-con01 at present so working through that issue at present.

    I suspect though that I will need to rebuild the environment (again).

    Friday, June 9, 2017 8:31 AM