Getting 'reservation conflict' errors while none of the applications on server are using SCSI reservations.

Solution Verified - Updated -

Issue

  • We have Oracle RAC setup of 2 nodes where each node uses its own set of EMC Symmetrix LUNs. These LUNs are not shared with cluster nodes, and Oracle RAC is also not setup to use SCSI reservations, but still the database restoration activity results in following SCSI reservation conflict errors:

    Dec  1 21:36:03 host1 kernel: [1842327.927058] sd 12:0:3:12: reservation conflict
    Dec  1 21:36:03 host1 kernel: [1842327.927065] sd 12:0:3:12: [sdbe] tag#27 Done: SUCCESS Result: hostbyte=DID_NEXUS_FAILURE driverbyte=DRIVER_OK
    Dec  1 21:36:03 host1 kernel: [1842327.927068] sd 12:0:3:12: [sdbe] tag#27 CDB: Write(10) 2a 00 37 2c aa e0 00 00 10 00
    Dec  1 21:36:03 host1 kernel: [1842327.927070] blk_update_request: critical nexus error, dev sdbe, sector 925674208
    Dec  1 21:36:03 host1 kernel: [1842327.927850] sd 12:0:1:12: reservation conflict
    Dec  1 21:36:03 host1 kernel: [1842327.927852] sd 12:0:1:12: [sdaq] tag#36 Done: SUCCESS Result: hostbyte=DID_NEXUS_FAILURE driverbyte=DRIVER_OK
    Dec  1 21:36:03 host1 kernel: [1842327.927853] sd 12:0:1:12: [sdaq] tag#36 CDB: Write(10) 2a 00 37 41 66 f0 00 00 10 00
    Dec  1 21:36:03 host1 kernel: [1842327.927855] blk_update_request: critical nexus error, dev sdaq, sector 927033072
    Dec  1 21:36:03 host1 kernel: sd 12:0:3:12: reservation conflict
    Dec  1 21:36:03 host1 kernel: sd 12:0:3:12: [sdbe] tag#27 Done: SUCCESS Result: hostbyte=DID_NEXUS_FAILURE driverbyte=DRIVER_OK
    Dec  1 21:36:03 host1 kernel: sd 12:0:3:12: [sdbe] tag#27 CDB: Write(10) 2a 00 37 2c aa e0 00 00 10 00
    Dec  1 21:36:03 host1 kernel: blk_update_request: critical nexus error, dev sdbe, sector 925674208
    Dec  1 21:36:03 host1 kernel: sd 12:0:1:12: reservation conflict
    Dec  1 21:36:03 host1 kernel: sd 12:0:1:12: [sdaq] tag#36 Done: SUCCESS Result: hostbyte=DID_NEXUS_FAILURE driverbyte=DRIVER_OK
    Dec  1 21:36:03 host1 kernel: sd 12:0:1:12: [sdaq] tag#36 CDB: Write(10) 2a 00 37 41 66 f0 00 00 10 00
    Dec  1 21:36:03 host1 kernel: blk_update_request: critical nexus error, dev sdaq, sector 927033072
    [...]
    
  • Need RCA for SCSI reservation conflict errors.

Environment

  • Red Hat Enterprise Linux 7
  • EMC Symmetrix SAN
    • Product: VMAX3
    • HYPERMAX OS 5977.1125.1125

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.

Current Customers and Partners

Log in for full access

Log In

New to Red Hat?

Learn more about Red Hat subscriptions

Using a Red Hat product through a public cloud?

How to access this content