I'm hoping someone has seen this and can assist. This is more of an annoyance than a true issue.
I have a Dell R710 stacked with Debian 9.4 (amd64). The physical unit has the following characteristics:
- 2x Xeon X5550
- 32GB RAM
- Perc 6/i RAID Controller
- 2x 146GB 15k SAS
- 3x 146GB 10k SAS (combined as RAID disk)
- 3x 300GB 10k SAS (combined as RAID disk)
- SATA DVD Drive
I am booting off of the first 15k SAS drive currently and not using the other drives at present.
The physical box has an internal USB & SD card port. I had previously booted another system on the internal and opted just to go with a physical disk at this time so there are no items in the USB or SD card port.
BIOS is set to boot to the "C:" drive (their naming ... not mine)
Once up the box runs like a champ with no issues. During boot however is another story.
Kernel Installed:
Linux localhost 4.9.0-6-amd64 #1 SMP Debian 4.9.82-1+deb9u3 (2018-03-02) x86_64 GNU/Linux
There is a 2 min+ gap during startup that I've seen the following errors associated with in dmesg:
[ 2.513315] DMAR: DRHD: handling fault status reg 502
[ 2.513378] DMAR: [DMA Read] Request device [03:00.0] fault addr fff0d000 [fault reason 06] PTE Read access is not set
[ 2.518031] scsi 0:0:32:0: Wrong diagnostic page; asked for 7 got 0
[ 2.617616] device-mapper: uevent: version 1.0.3
[ 2.618173] device-mapper: ioctl: 4.35.0-ioctl (2016-06-23) initialised: dm-devel@redhat.com
[ 2.778729] DMAR: DRHD: handling fault status reg 602
[ 2.778793] DMAR: [DMA Read] Request device [03:00.0] fault addr fff20000 [fault reason 06] PTE Read access is not set
[ 2.783376] scsi 0:0:32:0: Wrong diagnostic page; asked for 7 got 0
[ 2.972320] clocksource: Switched to clocksource tsc
[ 3.042491] DMAR: DRHD: handling fault status reg 702
[ 3.042556] DMAR: [DMA Read] Request device [03:00.0] fault addr ffc83000 [fault reason 06] PTE Read access is not set
[ 3.047175] scsi 0:0:32:0: Wrong diagnostic page; asked for 7 got 0
[ 3.307839] DMAR: DRHD: handling fault status reg 2
[ 3.307915] DMAR: [DMA Read] Request device [03:00.0] fault addr ff3bb000 [fault reason 06] PTE Read access is not set
[ 3.312489] scsi 0:0:32:0: Wrong diagnostic page; asked for 7 got 0
[ 3.363044] random: crng init done
[ 3.577114] scsi 0:0:32:0: Wrong diagnostic page; asked for 7 got 0
[ 3.840869] scsi 0:0:32:0: Wrong diagnostic page; asked for 7 got 0
[ 4.105521] scsi 0:0:32:0: Wrong diagnostic page; asked for 7 got 0
[ 4.371579] scsi 0:0:32:0: Wrong diagnostic page; asked for 7 got 0
[ 4.635322] scsi 0:0:32:0: Wrong diagnostic page; asked for 7 got 0
[ 4.837103] ses 0:0:32:0: Attached Enclosure device
[ 131.523048] PM: Starting manual resume from disk
The line
[ 4.635322] scsi 0:0:32:0: Wrong diagnostic page; asked for 7 got 0
I believe is a known issue in the kernel (see here).
I had at one point done some lunacy and was running a stretch-backports kernel which was getting a different error (asked for 10 got 0).
Staying at the topic at hand - is there a way to tell the kernel to ignore device "0:0:32:0"?
