dev_vdisk: ***ERROR***: FLUSH bio failed: -121 (cmd ffff8801efd16ac8)

Esos-User123

unread,

Apr 19, 2017, 9:54:17 AM4/19/17

to esos-users

Hi,

we have an problem with our setup.
We are using the esos as a FC-target for just one Server.

-vdisk_blockio with default blocksize (512) ans settings
-1Target/2Initiators configured as Windows MPIO
-HBA: qle2462
-ESOS-Version:1.0.0

dmesg:
[76076.208996] sd 0:0:0:0: [sda] tag#311 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[76076.208999] sd 0:0:0:0: [sda] tag#311 Sense Key : Illegal Request [current]
[76076.209009] sd 0:0:0:0: [sda] tag#311 Add. Sense: Invalid command operation code
[76076.209015] sd 0:0:0:0: [sda] tag#311 CDB: Synchronize Cache(10) 35 00 00 00 00 00 00 00 00 00
[76076.209017] blk_update_request: critical target error, dev sda, sector 0
[76076.209018] dev_vdisk: ***ERROR***: FLUSH bio failed: -121 (cmd ffff8801efd14a48)
[76076.209417] sd 0:0:0:0: [sda] tag#311 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[76076.209420] sd 0:0:0:0: [sda] tag#311 Sense Key : Illegal Request [current]
[76076.209422] sd 0:0:0:0: [sda] tag#311 Add. Sense: Invalid command operation code
[76076.209425] sd 0:0:0:0: [sda] tag#311 CDB: Synchronize Cache(10) 35 00 00 00 00 00 00 00 00 00
[76076.209426] blk_update_request: critical target error, dev sda, sector 0
[76076.209428] dev_vdisk: ***ERROR***: FLUSH bio failed: -121 (cmd ffff8801efd15a88)
[76076.212353] sd 0:0:0:0: [sda] tag#311 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[76076.212355] sd 0:0:0:0: [sda] tag#311 Sense Key : Illegal Request [current]
[76076.212358] sd 0:0:0:0: [sda] tag#311 Add. Sense: Invalid command operation code
[76076.212361] sd 0:0:0:0: [sda] tag#311 CDB: Synchronize Cache(10) 35 00 00 00 00 00 00 00 00 00
[76076.212362] blk_update_request: critical target error, dev sda, sector 0
[76076.212363] dev_vdisk: ***ERROR***: FLUSH bio failed: -121 (cmd ffff8801efd16108)

Sometimes the connection get lost and the server is unable to reconnect to the FC-Target.
Do you have some ideas, what happends and who we can fix it.

Best Regards,
EU123

/etc/scst.conf:
__________________________________________________
# Automatically generated by SCST Configurator v3.2.0-pre1.

# Non-key attributes
max_tasklet_cmd 10
poll_us 0
setup_id 0x0
suspend 0
threads 4

HANDLER vdisk_blockio {
        DEVICE NAS5 {
                filename /dev/disk/by-id/scsi-20b9ca35e00d00000

                # Non-key attributes
                block "0 0"
                blocksize 512
                cluster_mode 0
                expl_alua 0
                nv_cache 0
                pr_file_name /var/lib/scst/pr/NAS5
                prod_id NAS5
                prod_rev_lvl " 321"
                read_only 0
                removable 0
                rotational 1
                size 7988639170560
                size_mb 7618560
                t10_dev_id 543d2380-NAS5
                t10_vend_id SCST_BIO
                thin_provisioned 0
                threads_num 1
                threads_pool_type per_initiator
                tst 1
                usn 543d2380
                vend_specific_id 543d2380-NAS5
                write_through 0
        }
}

TARGET_DRIVER copy_manager {
        # Non-key attributes
        allow_not_connected_copy 0

        TARGET copy_manager_tgt {
                # Non-key attributes
                addr_method PERIPHERAL
                black_hole 0
                cpu_mask f
                forwarding 0
                io_grouping_type auto
                rel_tgt_id 0

                LUN 1 NAS5 {
                        # Non-key attributes
                        read_only 0
                }
        }
}

TARGET_DRIVER iscsi {
        enabled 0
}

TARGET_DRIVER qla2x00t {
        TARGET XX:XX:XX:XX:XX:XX:XX:XX {
               HW_TARGET

                enabled 1
                rel_tgt_id 1

                # Non-key attributes
                addr_method PERIPHERAL
                black_hole 0
                cpu_mask f
                explicit_confirmation 0
                forwarding 0
                io_grouping_type auto
                node_name 20:00:00:24:ff:0c:7e:f9
                port_name 21:00:00:24:ff:0c:7e:f9

                GROUP nas {
                        LUN 0 NAS5 {
                                # Non-key attributes
                                read_only 0
                        }

                        INITIATOR XX:XX:XX:XX:XX:XX:XX:XX

                        INITIATOR XX:XX:XX:XX:XX:XX:XX:XX

                        # Non-key attributes
                        addr_method PERIPHERAL
                        black_hole 0
                        cpu_mask f
                        io_grouping_type auto
                }
        }
}

Marc Smith

unread,

Apr 19, 2017, 10:42:26 AM4/19/17

to esos-...@googlegroups.com

Looks like something is wrong with your '/dev/sda' block device... I
assume that is what /dev/disk/by-id/scsi-20b9ca35e00d00000 resolves
to? What is that block device? A SCSI disk from a hardware RAID
controller? Can you access it inside ESOS (eg, fdisk -l /dev/sda)?

--Marc

> --
> You received this message because you are subscribed to the Google Groups
> "esos-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to esos-users+...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

Esos-User123

unread,

Apr 20, 2017, 8:13:14 AM4/20/17

to esos-users

Hey,

yes we are using a Raid-Controller (ICP 5085BL) with 8 HDD (2TB).
The configuration is RAID 10.

fdisk -l /dev/sda says: "fdisk: device has more than 2^32 sectors, can't use all of them", because we are using a GPT-Partiontable.
But the device is visible and we are able to write more then 2TB on the Storage(as to the moment the OS is crashing).

Marc Smith

unread,

Apr 20, 2017, 9:04:55 AM4/20/17

to esos-...@googlegroups.com

On Thu, Apr 20, 2017 at 8:13 AM, Esos-User123 <bianco...@gmail.com> wrote:
>
> Hey,
>
> yes we are using a Raid-Controller (ICP 5085BL) with 8 HDD (2TB).
> The configuration is RAID 10.
>
> fdisk -l /dev/sda says: "fdisk: device has more than 2^32 sectors, can't use
> all of them", because we are using a GPT-Partiontable.
> But the device is visible and we are able to write more then 2TB on the
> Storage(as to the moment the OS is crashing).
>
>

That warning that fdisk displays is fine, what I was looking for was
if you could read data from the /dev/sda device inside of ESOS?
Double-check that your RAID controller is functioning properly and the
logical RAID10 device is accessible.

And checking for any other interesting kernel messages.

--Marc

Reply all

Reply to author

Forward