Panics on mps(4) firmware timeouts
Description
Problem/Justification
Impact
SmartDraw Connector
Katalon Manual Tests (BETA)
Activity
Alexander Motin February 12, 2021 at 3:12 PM
Thomas Kempf February 12, 2021 at 1:56 PM
Hi, im using an LSI HBA too. Form 12.0-U1 on my backups to a LTO-VI Library went smooth without shoeshining becaus of the increased io_size.
After the update to U2 my backups failed, because of the default io_size. I had no problems with my attached 16x8TB seagate Exos SAS-Drives.
Just to be sure, if i want to increase the io_size to 256k, i set the tunable to 64 ? Is that correct ?
dmesg | grep mps
mps0: <Avago Technologies (LSI) SAS2308> port 0x3000-0x30ff mem 0x8b440000-0x8b44ffff,0x8b400000-0x8b43ffff at device 0.0 on pci6
mps0: Firmware: 20.00.07.00, Driver: 21.02.00.00-fbsd
mps0: IOCCapabilities: 5a85c<ScsiTaskFull,DiagTrace,SnapBuf,EEDP,TransRetry,EventReplay,MSIXIndex,HostDisc>camcontrol negotiate sa0 -v
Current parameters:
(pass0:mps0:0:0:0): transfer speed: 600.000MB/s
(pass0:mps0:0:0:0): tagged queueing: enabled
mps0: SIM/HBA version: 1
mps0: supports tag queue messages
mps0: supports SDTR message
mps0: supports 16 bit wide SCSI
mps0: does its own scanning
mps0: unmapped I/O supported
mps0: user has disabled initial BUS RESET or controller is in target/mixed mode
mps0: HBA engine count: 0
mps0: maximum target: 1023
mps0: maximum LUN: 255
mps0: highest path ID in subsystem: 0
mps0: initiator ID: 1024
mps0: SIM vendor: FreeBSD
mps0: HBA vendor: Avago Tech
mps0: HBA vendor ID: 0x0000
mps0: HBA device ID: 0x0000
mps0: HBA subvendor ID: 0x0000
mps0: HBA subdevice ID: 0x0000
mps0: bus ID: 0
mps0: base transfer speed: 150.000MB/sec
mps0: maximum transfer size: 131072 bytesDavid Kowis February 2, 2021 at 3:39 PM
So I've been up for another 5-ish days, no errors or freakouts. So I guess that's good
Alexander Motin January 29, 2021 at 4:38 AM
> We found one bug that the tunable isn't happening when a comment is included.
Please create separate ticket for us to look at that another day.
> But, why did my panics go away when the tunable was set to a higher number anyway?
Was it happening just after boot before? Can't it be that it happen only later when memory fragmentation starts to grow, making addressation more complicated for HBA? Or may be it is specific to some specific I/O pattern, like scrub, that completed meanwhile. I can speculate all day long.
David Kowis January 29, 2021 at 3:35 AM
@Josh Karli Here's the drive information that I've got in the system:
Script used to get you this info:
#!/bin/bash
for foo in da0 da1 da2 da3 da4 da5 da6 da7 da8 da9 da10 da11; do
echo "Drive: /dev/$foo"
smartctl -a /dev/$foo | grep -E "(Device Model|Firmware Version)"
doneoutput:
root@freenas[~]# bash driveInfo.sh
Drive: /dev/da0
Device Model: WDC WD80EZAZ-11TDBA0
Firmware Version: 83.H0A83
Drive: /dev/da1
Device Model: WDC WD80EZAZ-11TDBA0
Firmware Version: 83.H0A83
Drive: /dev/da2
Device Model: WDC WD80EZAZ-11TDBA0
Firmware Version: 83.H0A83
Drive: /dev/da3
Device Model: WDC WD80EZAZ-11TDBA0
Firmware Version: 83.H0A83
Drive: /dev/da4
Device Model: WDC WD80EZAZ-11TDBA0
Firmware Version: 83.H0A83
Drive: /dev/da5
Device Model: WDC WD80EZAZ-11TDBA0
Firmware Version: 83.H0A83
Drive: /dev/da6
Device Model: ST2000VN004-2E4164
Firmware Version: SC60
Drive: /dev/da7
Device Model: ST2000VN004-2E4164
Firmware Version: SC60
Drive: /dev/da8
Device Model: ST2000VN004-2E4164
Firmware Version: SC60
Drive: /dev/da9
Device Model: WDC WD2001FFSX-68JNUN0
Firmware Version: 81.00A81
Drive: /dev/da10
Device Model: ST2000VN004-2E4164
Firmware Version: SC60
Drive: /dev/da11
Device Model: WDC WD2002FAEX-007BA0
Firmware Version: 05.01D05
@Thomas Kempf Correct.