lost device (ada0 cache SSD) FreeNAS-8.0.3-RELEASE_MULTIMEDIA-p1-x64 (9591)

[Rogla]

Why do I get this in my kernel log:

freenas.WORKGROUP kernel log messages:
+++ /tmp/security.EcgshnBP 2012-02-08 03:01:01.000000000 +0100
+(ada0:ahcich0:0:0:0): lost device
+(ada0:ahcich0:0:0:0): removing device entry
+ada0 at ahcich0 bus 0 scbus2 target 0 lun 0
+ada0: <Corsair Force GT 1.3.3> ATA-8 SATA 3.x device
+ada0: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes)
+ada0: Command Queueing enabled
+ada0: 57241MB (117231408 512 byte sectors: 16H 63S/T 16383C)

I don't understand why it is lost and then found again several times a day.
Is this normal or something wrong in my configuration?


Some info about my system:
View attachment my_camcontrol.txt
View attachment my_dmesg.txt

The ada0 SSD is the cache disk, 60 GB.

FreeNAS Build FreeNAS-8.0.3-RELEASE_MULTIMEDIA-p1-x64 (9591) (fresh install, not an upgrade)
Platform AMD A8-3870 APU with Radeon(tm) HD Graphics
Memory 15837MB

The storage volume is ada1..ada6.

This "lost device" issue make me question if I can trust the setup.
Despite this, the read and write performance on the local volume is very good:

[Roger@freenas] ~/vol0> dd if=/dev/zero of=./tmp.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 97.723123 secs (1098759222 bytes/sec)

[Roger@freenas] ~/vol0> dd if=./tmp.dat of=/dev/null bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 104.558536 secs (1026928902 bytes/sec)

[Roger@freenas] ~/vol0> ls -l ./tmp.dat
-rw-r--r-- 1 Roger wheel 107374182400 Feb 7 22:01 ./tmp.dat
[Roger@freenas] ~/vol0>

 

[gcooper]

Why do I get this in my kernel log:

freenas.WORKGROUP kernel log messages:
+++ /tmp/security.EcgshnBP 2012-02-08 03:01:01.000000000 +0100
+(ada0:ahcich0:0:0:0): lost device
+(ada0:ahcich0:0:0:0): removing device entry
+ada0 at ahcich0 bus 0 scbus2 target 0 lun 0
+ada0: <Corsair Force GT 1.3.3> ATA-8 SATA 3.x device
+ada0: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes)
+ada0: Command Queueing enabled
+ada0: 57241MB (117231408 512 byte sectors: 16H 63S/T 16383C)

I don't understand why it is lost and then found again several times a day.
Is this normal or something wrong in my configuration?


Some info about my system:
View attachment 616
View attachment 617

The ada0 SSD is the cache disk, 60 GB.

FreeNAS Build FreeNAS-8.0.3-RELEASE_MULTIMEDIA-p1-x64 (9591) (fresh install, not an upgrade)
Platform AMD A8-3870 APU with Radeon(tm) HD Graphics
Memory 15837MB

The storage volume is ada1..ada6.

This "lost device" issue make me question if I can trust the setup.
Despite this, the read and write performance on the local volume is very good:

[Roger@freenas] ~/vol0> dd if=/dev/zero of=./tmp.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 97.723123 secs (1098759222 bytes/sec)

[Roger@freenas] ~/vol0> dd if=./tmp.dat of=/dev/null bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 104.558536 secs (1026928902 bytes/sec)

[Roger@freenas] ~/vol0> ls -l ./tmp.dat
-rw-r--r-- 1 Roger wheel 107374182400 Feb 7 22:01 ./tmp.dat
[Roger@freenas] ~/vol0>

Check the disk, controller, or cabling. If it drops on and off the channel it's a sign that the link is potentially going south; there's an offhand chance that it might be a driver bug, but bad hardware is more likely. It would be helpful to note when the device drops off the channel and see if it corresponds with another item being done with the controller at that given point in time (zfs scrub, periodic jobs iterating a large directory tree, etc).

 

[Rogla]

I was able to get rid of this issue.

I disconected the OPROM and moved the SSD cache disk to another SATA controller.

There are 2 SATA controllers on the ASUS F1A74-V EVO MB:
ahci0: AHCI v1.20 with 2 6Gbps ports, Port Multiplier supported
ahci1: AHCI v1.30 with 6 6Gbps ports, Port Multiplier supported

Before:
6 ports controller = 5 disk RAIDZ + SSD cache.
2 ports controller = OPROM + eSATA (not used).

Now:
6 port controller = 5 disk RAIDZ + 1 port not used.
2 port controller = SSD cache + eSATA (not used).

The strange thing is that when I look int the old dmsg output, only one of the controllers is reported:
ahci0: AHCI v1.30 with 6 6Gbps ports, Port Multiplier supported

I don't need the OPROM so I put this to a rest now, I'm happy!

 

[gcooper]

I was able to get rid of this issue.

I disconected the OPROM and moved the SSD cache disk to another SATA controller.

There are 2 SATA controllers on the ASUS F1A74-V EVO MB:
ahci0: AHCI v1.20 with 2 6Gbps ports, Port Multiplier supported
ahci1: AHCI v1.30 with 6 6Gbps ports, Port Multiplier supported

Before:
6 ports controller = 5 disk RAIDZ + SSD cache.
2 ports controller = OPROM + eSATA (not used).

Now:
6 port controller = 5 disk RAIDZ + 1 port not used.
2 port controller = SSD cache + eSATA (not used).

The strange thing is that when I look int the old dmsg output, only one of the controllers is reported:
ahci0: AHCI v1.30 with 6 6Gbps ports, Port Multiplier supported

I don't need the OPROM so I put this to a rest now, I'm happy!

That's interesting. Thanks for the feedback :)!

 

[Rogla]

A little late, but FYI...

Several months later than I opened this thread, a new BIOS version was released for the ASUS F1A74-V EVO MB.
All problems disapeard after installation of the new BIOS and the OPROM is working since then.
No problem with FreeNAS!