Hi,
This morning our main DB server which is in production went into read only disk mode, similar to the following appeared in dmesg:
[102570.045689] hv_storvsc vmbus_0_1: cmd 0x2a scsi status 0x2 srb status 0x4
[102580.073255] hv_storvsc vmbus_0_1: cmd 0x2a scsi status 0x2 srb status 0x4
[102590.108656] hv_storvsc vmbus_0_1: cmd 0x2a scsi status 0x2 srb status 0x4
[102600.630709] hv_storvsc vmbus_0_1: cmd 0x2a scsi status 0x2 srb status 0x4
[102610.656294] hv_storvsc vmbus_0_1: cmd 0x2a scsi status 0x2 srb status 0x4
[102620.686627] hv_storvsc vmbus_0_1: cmd 0x2a scsi status 0x2 srb status 0x4
[102630.695776] sd 0:0:0:0: [sda] Unhandled error code
[102630.695786] sd 0:0:0:0: [sda] Result: hostbyte=invalid driverbyte=DRIVER_OK
[102630.695796] sd 0:0:0:0: [sda] CDB: Write(10): 2a 08 01 db 5e 20 00 00 08 00
[102630.695814] end_request: critical target error, dev sda, sector 31153696
[102630.695828] end_request: critical target error, dev sda, sector 31153696
[102630.695837] Buffer I/O error on device sda1, logical block 3893956
[102630.695843] lost page write due to I/O error on sda1
[102630.696154] Aborting journal on device sda1.
[102630.696946] EXT3-fs (sda1): error: ext3_journal_start_sb: Detected aborted journal
[102630.696953] EXT3-fs (sda1): error: remounting filesystem read-only
[102630.713427] journal commit I/O error
What would be the cause of this? How did it happen? Why did it happen? This is a production system, is there an SLA?
Thanks.