Bug 253954 - kernel: g_access(958): provider da8 has error 6 set
Summary: kernel: g_access(958): provider da8 has error 6 set
Status: New
Alias: None
Product: Base System
Classification: Unclassified
Component: kern (show other bugs)
Version: 12.1-STABLE
Hardware: amd64 Any
: --- Affects Some People
Assignee: freebsd-bugs (Nobody)
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2021-03-02 11:52 UTC by Willem Vermeylen
Modified: 2021-03-24 16:32 UTC (History)
3 users (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Willem Vermeylen 2021-03-02 11:52:22 UTC
Hey,

I am encountering a (possible) bug with FreeBSD which we tried to solve for the past few years with no success.

When a disk fails and a new disk is being inserted in our server, it is not correctly being detected by the system leaving the following results in geom:

```
Geom name: da8
Providers:
1. Name: da8
   Mediasize: 8001563222016 (7.3T)
   Sectorsize: 512
   Stripesize: 4096
   Stripeoffset: 0
   Mode: r0w0e0
   wither: (null)
```

Weirdly enough when we reboot the system, the disk is fine and is fully being detected like normally.

While the disk is having this issue, we are left with a system that can't do anything with the disk. We can't turn on the disk indicator through the sas controller, we can't add it to zfs and trying to reset it with camcontrol is also not giving any results.

We did see the following pop up when we try to add the disk to zfs anyway:

```
kernel: g_access(958): provider da8 has error 6 set
```

to me it almost looks like it somehow holds on to the old disk and therefor can't fully detect the new one until you reboot and therefor flush everything.

From what I remember we started to have this bug after we upgraded from 10.3 to 11.2 and it continued on 12.1 and 12.2 now
Comment 1 Willem Vermeylen 2021-03-02 12:04:18 UTC
some info I forgot:

the sas controller we use:

```
r0@pci0:1:0:0:	class=0x010700 card=0x30e01000 chip=0x00971000 rev=0x02 hdr=0x00
    vendor     = 'Broadcom / LSI'
    device     = 'SAS3008 PCI-Express Fusion-MPT SAS-3'
    class      = mass storage
    subclass   = SAS
```

upgrading the controller makes no difference in this issue so far.
Comment 2 Tomasz "CeDeROM" CEDRO 2021-03-24 16:32:01 UTC
Hello world :-)

I have just encountered this problem using external usb dvd recorder. Here is the dmesg output:

ugen0.13: <Hitachi-LG Data Storage Inc Portable Super Multi Drive> at usbus0
umass0 on uhub1
umass0: <6238--Storage> on usbus0
umass0:  8070i (ATAPI) over Bulk-Only; quirks = 0x0100
umass0:2:0: Attached to scbus2
cd0 at umass-sim0 bus 0 scbus2 target 0 lun 0
cd0: <HL-DT-ST DVDRAM GP57EB40 PF00> Removable CD-ROM SCSI device
cd0: 40.000MB/s transfers
cd0: 0MB (1 0 byte sectors)
cd0: quirks=0x10<10_BYTE_ONLY>
GEOM_PART: integrity check failed (cd0, MBR)
GEOM_PART: integrity check failed (iso9660/Kali%20Live, MBR)
ugen0.13: <Hitachi-LG Data Storage Inc Portable Super Multi Drive> at usbus0 (disconnected)
umass0: at uhub1, port 1, addr 16 (disconnected)
cd0 at umass-sim0 bus 0 scbus2 target 0 lun 0
cd0: <HL-DT-ST DVDRAM GP57EB40 PF00>  detached
g_access(961): provider cd0 has error 6 set
g_access(961): provider cd0 has error 6 set
g_access(961): provider cd0 has error 6 set


May be caused by GEOM but also usbconfig hangs and no new devices can be attached.

Any hints welcome :-)
Tomek