Re: [Freeipmi-users] problem with freeipmi 0.5.3 and intel SE7221BK1-E

[Al Chu]

Hey Chris,

Well, there are a few interesting things.

A) everything that returned "C2" indicates that lun != 0x0 is a bad lun.
(C2 == command invalid for given lun). So my initial guess was wrong,
but that's ok, that removes one possibility.

B) 

address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x0 0x0A 0x20
> ipmi-ssif-api-udm.c: 158: ipmi_ssif_cmd_raw_udm: errnum (4): expression failed
> ipmi_cmd_raw: permission denied

This is strange, because this is supposed to be the same as:

> ================================================================
> [ VALUE               TAG NAME:LENGTH                          ]
> ================================================================
> [              20h] = cmd[ 8b]
> ================================================================
> ================================================================
> [ VALUE               TAG NAME:LENGTH                          ]
> ================================================================
> [               0h] = cmd[ 8b]
> [               0h] = comp_code[ 8b]
> [               0h] = sdr_version_major[ 4b]
> [               0h] = sdr_version_minor[ 4b]
> [               0h] = record_count[16b]
> [               0h] = free_space[16b]
> [               0h] = most_recent_addition_timestamp[32b]
> [               0h] = most_recent_erase_timestamp[32b]
> [               0h] =
get_sdr_repository_allocation_info_command_supported[ 1b]
> [               0h] = reserve_sdr_repository_command_supported[ 1b]
> [               0h] = partial_add_sdr_command_supported[ 1b]
> [               0h] = delete_sdr_command_supported[ 1b]
> [               0h] = reserved[ 1b]
> [               0h] =
> modal_non_modal_sdr_repository_update_operation_supported[ 2b]
> [               0h] = overflow_flag[ 1b]
> ================================================================

which you sent me earlier.  It could be that ipmi-raw doesn't work b/c
the raw commands desire to read all bytes available rather than a
specific number of bytes.  So it could confuse the underlying driver.
No biggie in the end.

Could you run ipmi-sensors just one more time for kicks to see if we get
the same output (run w/ -f one time first, to flush the cache that may
have been created beforehand.)

Now, the question we now ask is why are most of the tools working (bmc-
info, bmc-config, pef-config, ipmi-chassis, i'm not sure if you've tried
ipmipower or ipmiping yet), but anything related to the SDR/sensors is
not working b/c the IPMI commands return junk.  I suppose its possible
on this motherboard they just aren't supported?  Do you know if your
vendor has asserted that ipmi based sensors are available on this
motherboard or other tools do work?

I'd also like to see how ipmi-sensors, ipmi-fru and ipmi-sel work w/o
using the SDR.  I will send you a tar.gz (out of this thread) later on
in the day to try this idea out with some new options in the tools.

Al

On Wed, 2008-01-23 at 13:30 -0500, Chris Morgan wrote:
> Hi Al.
> 
> Thanks for sticking with this.
> 
> 
> On Jan 23, 2008 12:34 PM, Al Chu <address@hidden> wrote:
> > Hey Chris,
> >
> > Well, its very evident that a lot of stuff does work.  Lets try this
> > one:
> >
> > ipmi-raw --no-probing -D SSIF --driver-address=0x42 --driver-
> > device=/dev/i2c-0 --register-spacing=1 --debug 0x0 0x04 0x2D 0x1
> >
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x0 0x04 0x2D
> 0x1
> ipmi-ssif-api-udm.c: 158: ipmi_ssif_cmd_raw_udm: errnum (4): expression failed
> ipmi_cmd_raw: permission denied
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x0 0x04 0x2D
> 0x0
> ipmi-ssif-api-udm.c: 158: ipmi_ssif_cmd_raw_udm: errnum (4): expression failed
> ipmi_cmd_raw: permission denied
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x0 0x04 0x2D
> 0x2
> ipmi-ssif-api-udm.c: 158: ipmi_ssif_cmd_raw_udm: errnum (4): expression failed
> ipmi_cmd_raw: permission denied
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x0 0x04 0x2D
> 0x3
> ipmi-ssif-api-udm.c: 158: ipmi_ssif_cmd_raw_udm: errnum (4): expression failed
> ipmi_cmd_raw: permission denied
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x0 0x04 0x2D
> 0x4
> ipmi-ssif-api-udm.c: 158: ipmi_ssif_cmd_raw_udm: errnum (4): expression failed
> ipmi_cmd_raw: permission denied
> 
> 
> 
> 
> > ipmi-raw --no-probing -D SSIF --driver-address=0x42 --driver-
> > device=/dev/i2c-0 --register-spacing=1 --debug 0x1 0x04 0x2D 0x1
> >
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x1 0x04 0x2D
> 0x1
> rcvd: 2D C2
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x1 0x04 0x2D
> 0x0
> rcvd: 2D C2
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x1 0x04 0x2D
> 0x2
> rcvd: 2D C2
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x1 0x04 0x2D
> 0x3
> rcvd: 2D C2
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x1 0x04 0x2D
> 0x4
> rcvd: 2D C2
> 
> 
> 
> > ipmi-raw --no-probing -D SSIF --driver-address=0x42 --driver-
> > device=/dev/i2c-0 --register-spacing=1 --debug 0x2 0x04 0x2D 0x1
> >
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x2 0x04 0x2D
> 0x1
> rcvd: 2D C2
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x2 0x04 0x2D
> 0x0
> rcvd: 2D C2
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x2 0x04 0x2D
> 0x2
> rcvd: 2D C2
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x2 0x04 0x2D
> 0x3
> rcvd: 2D C2
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x2 0x04 0x2D
> 0x4
> rcvd: 2D C2
> 
> 
> 
> > ipmi-raw --no-probing -D SSIF --driver-address=0x42 --driver-
> > device=/dev/i2c-0 --register-spacing=1 --debug 0x3 0x04 0x2D 0x1
> >
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x3 0x04 0x2D
> 0x1
> rcvd: 2D C2
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x3 0x04 0x2D
> 0x0
> rcvd: 2D C2
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x3 0x04 0x2D
> 0x2
> rcvd: 2D C2
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x3 0x04 0x2D
> 0x3
> rcvd: 2D C2
> 
> 
> 
> > (On the above, alter the last number (0x1) a few up and down to see if
> > you get different results if the output looks strange originally).
> >
> > then
> >
> > ipmi-raw --no-probing -D SSIF --driver-address=0x42 --driver-
> > device=/dev/i2c-0 --register-spacing=1 --debug 0x0 0x0A 0x20
> >
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x0 0x0A 0x20
> ipmi-ssif-api-udm.c: 158: ipmi_ssif_cmd_raw_udm: errnum (4): expression failed
> ipmi_cmd_raw: permission denied
> 
> 
> 
> > ipmi-raw --no-probing -D SSIF --driver-address=0x42 --driver-
> > device=/dev/i2c-0 --register-spacing=1 --debug 0x1 0x0A 0x20
> >
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x1 0x0A 0x20
> rcvd: 20 C2
> 
> 
> > ipmi-raw --no-probing -D SSIF --driver-address=0x42 --driver-
> > device=/dev/i2c-0 --register-spacing=1 --debug 0x2 0x0A 0x20
> >
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x2 0x0A 0x20
> rcvd: 20 C2
> 
> 
> > ipmi-raw --no-probing -D SSIF --driver-address=0x42 --driver-
> > device=/dev/i2c-0 --register-spacing=1 --debug 0x3 0x0A 0x20
> >
> address@hidden:~# ipmi-raw --no-probing -D SSIF --driver-address=0x42
> --driver-device=/dev/i2c-0 --register-spacing=1 --debug 0x3 0x0A 0x20
> rcvd: 20 C2
> 
> 
> > The above are the hex commands to get a sensor reading on different
> > logical units and get SDR info on different logical units.  My guess is
> > that the SDR/sensor info may be on a different LUN, which is why the
> > other commands aren't working by default.  That's just a guess at this
> > point :|
> >
> > Al
> 
> Chris
-- 
Albert Chu
address@hidden
925-422-5311
Computer Scientist
High Performance Systems Division
Lawrence Livermore National Laboratory