[jira] [Created] (CLOUDSTACK-7072) [VMware] Volume attachment intermittently fails with 'A specified parameter was not correct' error.

["Likitha Shetty (JIRA)" <ji...@apache.org>]

Likitha Shetty created CLOUDSTACK-7072:
------------------------------------------

             Summary: [VMware] Volume attachment intermittently fails with 'A specified parameter was not correct' error.
                 Key: CLOUDSTACK-7072
                 URL: https://issues.apache.org/jira/browse/CLOUDSTACK-7072
             Project: CloudStack
          Issue Type: Bug
      Security Level: Public (Anyone can view this level - this is the default.)
          Components: VMware, Volumes
    Affects Versions: 4.4.0
            Reporter: Likitha Shetty
            Assignee: Likitha Shetty
             Fix For: 4.5.0


Today, while attaching a new disk to an instance the unit number on the controller key chosen for the disk is 1 digit higher than the highest unit number that is currently in use.

Hence we could run into potential volume attach failures. If we try to attach 15 or more volumes to the same instance and call attach and detach in fast succession it will result in us calling disk attach for a disk before the detach of the volume on the controller key with the highest unit number. In this case every volume will get a unit number that is one more than the highest, even though the lower unit numbers are not in use. So CS will try to attach the 15th data volume with SCSI id 16 even though the lower unit numbers are not in use and this attach will fail because there are only 16 devices on a SCSI bus.




--
This message was sent by Atlassian JIRA
(v6.2#6252)