https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=224496

--- Comment #45 from Matthias Pfaller <matthias.pfal...@familie-pfaller.de> ---
(In reply to Sharad Ahlawat from comment #44)
[root@nyx ~]# sysctl kern.cam.da.retry_count
kern.cam.da.retry_count: 4

Previously we had the drives connected to the onboard SATA/SAS ports (hoping
the mps problems would get solved). The machine was running for month without
showing any disk related problems. At the moment we are looking for a
replacement for the SAS2008 (for other machines without that many onboard
SATA/SAS channels) and are testing the following config:

ahci0: <Marvell 88SE9215 AHCI SATA controller> port
0x3050-0x3057,0x3040-0x3043,0x3030-0x3037,0x3020-0x3023,0x3000-0x301f mem
0xdf940000-0xdf9407ff irq 32 at device 0.0 on pci5
ahci0: AHCI v1.00 with 4 6Gbps ports, Port Multiplier supported with FBS
ahcich0: <AHCI channel> at channel 0 on ahci0
ahcich1: <AHCI channel> at channel 1 on ahci0
ahcich2: <AHCI channel> at channel 2 on ahci0
ahcich3: <AHCI channel> at channel 3 on ahci0
ahci1: AHCI v1.00 with 4 6Gbps ports, Port Multiplier supported with FBS
ahcich4: <AHCI channel> at channel 0 on ahci1
ahcich5: <AHCI channel> at channel 1 on ahci1
ahcich6: <AHCI channel> at channel 2 on ahci1
ahcich7: <AHCI channel> at channel 3 on ahci1
ada0 at ahcich0 bus 0 scbus0 target 0 lun 0
ada0: <ST12000VN0007-2GS116 SC60> ACS-3 ATA SATA 3.x device
ada0: Serial Number ZJV3ZYSX
ada0: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes)
ada0: Command Queueing enabled
ada0: 11444224MB (23437770752 512 byte sectors)
ada1 at ahcich1 bus 0 scbus1 target 0 lun 0
ada1: <ST12000VN0007-2GS116 SC60> ACS-3 ATA SATA 3.x device
ada1: Serial Number ZJV1VVWX
ada1: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes)
ada1: Command Queueing enabled
ada1: 11444224MB (23437770752 512 byte sectors)
ada2 at ahcich2 bus 0 scbus2 target 0 lun 0
ada2: <ST12000VN0007-2GS116 SC60> ACS-3 ATA SATA 3.x device
ada2: Serial Number ZJV1WLXM
ada2: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes)
ada2: Command Queueing enabled
ada2: 11444224MB (23437770752 512 byte sectors)
ada3 at ahcich3 bus 0 scbus3 target 0 lun 0
ada3: <ST12000VN0007-2GS116 SC60> ACS-3 ATA SATA 3.x device
ada3: Serial Number ZJV2YY9A
ada3: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes)
ada3: Command Queueing enabled
ada3: 11444224MB (23437770752 512 byte sectors)
ada4 at ahcich4 bus 0 scbus4 target 0 lun 0
ada4: <ST12000VN0007-2GS116 SC60> ACS-3 ATA SATA 3.x device
ada4: Serial Number ZJV3ZJNA
ada4: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes)
ada4: Command Queueing enabled
ada4: 11444224MB (23437770752 512 byte sectors)
ada5 at ahcich5 bus 0 scbus5 target 0 lun 0
ada5: <ST12000VN0007-2GS116 SC60> ACS-3 ATA SATA 3.x device
ada5: Serial Number ZJV3ZXN5
ada5: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes)
ada5: Command Queueing enabled
ada5: 11444224MB (23437770752 512 byte sectors)
ada6 at ahcich6 bus 0 scbus6 target 0 lun 0
ada6: <ST12000VN0007-2GS116 SC60> ACS-3 ATA SATA 3.x device
ada6: Serial Number ZJV2MWZ0
ada6: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes)
ada6: Command Queueing enabled
ada6: 11444224MB (23437770752 512 byte sectors)
ada7 at ahcich7 bus 0 scbus7 target 0 lun 0
ada7: <ST12000VN0007-2GS116 SC60> ACS-3 ATA SATA 3.x device
ada7: Serial Number ZCH0HHJ6
ada7: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes)
ada7: Command Queueing enabled
ada7: 11444224MB (23437770752 512 byte sectors)
ada8: <SAMSUNG MZ7KM240HMHQ-00005 GXM5304Q> ACS-2 ATA SATA 3.x device
ada8: Serial Number S3F5NX0M400493
ada8: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes)
ada8: Command Queueing enabled
ada8: 228936MB (468862128 512 byte sectors)
ada8: quirks=0x1<4K>
ada9: <SAMSUNG MZ7KM240HMHQ-00005 GXM5304Q> ACS-2 ATA SATA 3.x device
ada9: Serial Number S3F5NX0M400461
ada9: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes)
ada9: Command Queueing enabled
ada9: 228936MB (468862128 512 byte sectors)
ada9: quirks=0x1<4K>

I reverted the "camcontrol tags <DEV> -N 1" and the changed sysctls. So with
this setup the system runs with the same drives and without any changed sysctl
settings or cam settings. 

Up to now there are no problems showing up. Operating conditions are the same
as with the last mps test (a zpool scrub is running and our systems did a
backup to the machine last night).

As the drives behave well with at least three different controllers

  ahci0: <Intel Patsburg AHCI SATA controller>
  ahci1: <Marvell 88SE9215 AHCI SATA controller>
  isci0: <Intel(R) C600 Series Chipset SAS Controller (SATA mode)>

I can't imagine that changing parameters on the cam/zfs side (NCQ TRIM, NCQ,
timeouts) will help in our case.

Probably we will have to decommission the LSI9211 boards :-(


regards, Matthias

-- 
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"

Reply via email to