freeipmi-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Freeipmi-devel] internal IPMI error on successful ipmi_cmd_dcmi_set


From: Albert Chu
Subject: Re: [Freeipmi-devel] internal IPMI error on successful ipmi_cmd_dcmi_set_asset_tag
Date: Tue, 28 Jul 2015 09:42:23 -0700

Hey Curtis,

Using the original FreeIPMI you found the error with, could you run your
command below and give me the full --debug output.  The below makes
sense given how I programmed the workaround.  I'm betting there is
something earlier in the --debug stream that I missed.

Al

On Mon, 2015-07-27 at 23:34 -0700, Curtis Doty wrote:
> OK, this is curious...
> 
> 
> When I --commit with this new Release-1_4_0_branch it doesn't exit
> with an error status anymore. However...
> 
> 
> If I --commit --key-pair=DCMI_Conf:Asset_Tag=ABCD it only replaces the
> first 4 characters with ABCD. So if the previous Asset_Tag was
> 123456789, then I have a new/corrupt Asset_Tag of ABCD56789.
> 
> 
> ../C
> 
> 
> 
> 
> 
> On Mon, Jul 27, 2015 at 10:20 AM, Albert Chu <address@hidden> wrote:
>         Hey Curtis,
>         
>         I believe I have a fix in this branch if you could try it out.
>         
>         svn co
>         http://svn.savannah.gnu.org/svn/freeipmi/branches/Release-1_4_0_branch
>         freeipmi
>         cd freeipmi
>         ./autogen.sh
>         ./configure
>         make
>         ipmi-config/ipmi-config
>         
>         Thanks,
>         Al
>         
>         
>         On Mon, 2015-07-27 at 08:19 -0700, Albert Chu wrote:
>         > Hi Curtis,
>         >
>         >
>         > The "internal IPMI error", appears to come from the fact
>         that the
>         > Supermicro board doesn't return the length of the tag
>         written.
>         >
>         > DCMI Set Asset Tag Response
>         > ---------------------------
>         >
>         > FIID Template: tmpl_cmd_dcmi_set_asset_tag_rs
>         >
>         >     { 8, "cmd", REQUIRED, LENGTH-FIXED,
>         MAKES-PACKET-SUFFICIENT }
>         >     { 8, "comp_code", REQUIRED, LENGTH-FIXED,
>         MAKES-PACKET-SUFFICIENT }
>         >     { 8, "group_extension_identification", REQUIRED,
>         LENGTH-FIXED }
>         >     { 8, "total_asset_tag_length_written", REQUIRED,
>         LENGTH-FIXED }
>         >
>         > It just so happens that the write actually succeeds as a
>         side effect.
>         >
>         > I could write a patch to workaround this by specifically
>         checking for this condition and not outputting the error
>         message??
>         >
>         > Could you tell me the Supermicro motherboard you're using so
>         I can document it?
>         >
>         > Al
>         >
>         > On Sun, Jul 26, 2015 at 2:39 PM, Curtis Doty
>         <address@hidden>
>         > wrote:
>         >         The asset tag is actually changed successfully. But
>         >         ipmi-config still exits with status 2.
>         >
>         >
>         >         Is this a bug in FreeIPMI or the SuperMicro BMC I'm
>         using?
>         >
>         >
>         >
>         >         ../C
>         >
>         >
>         >         $ ipmi-config -vv --driver-type=LAN_2_0
>         --hostname=192.0.2.42
>         >         --username=foo --password=bar --category=dcmi
>         --commit
>         >         --key-pair=DCMI_Conf:Asset_Tag=ABCD --debug
>         >          ...
>         >         192.0.2.42:
>         >
>          =====================================================
>         >         192.0.2.42: IPMI 2.0 Group Extension - Set Asset Tag
>         Response
>         >         192.0.2.42:
>         >
>          =====================================================
>         >         192.0.2.42: RMCP Header:
>         >         192.0.2.42: ------------
>         >         192.0.2.42: [               6h] = version[ 8b]
>         >         192.0.2.42: [               0h] = reserved[ 8b]
>         >         192.0.2.42: [              FFh] =
>         sequence_number[ 8b]
>         >         192.0.2.42: [               7h] =
>         message_class.class[ 5b]
>         >         192.0.2.42: [               0h] =
>         message_class.reserved[ 2b]
>         >         192.0.2.42: [               0h] =
>         message_class.ack[ 1b]
>         >         192.0.2.42: IPMI RMCPPLUS Session Header:
>         >         192.0.2.42: -----------------------------
>         >         192.0.2.42: [               6h] =
>         authentication_type[ 4b]
>         >         192.0.2.42: [               0h] = reserved1[ 4b]
>         >         192.0.2.42: [               0h] = payload_type[ 6b]
>         >         192.0.2.42: [               1h] =
>         >         payload_type.authenticated[ 1b]
>         >         192.0.2.42: [               1h] =
>         payload_type.encrypted[ 1b]
>         >         192.0.2.42: [        DBCC3969h] = session_id[32b]
>         >         192.0.2.42: [               2h] =
>         session_sequence_number[32b]
>         >         192.0.2.42: [              20h] =
>         ipmi_payload_len[16b]
>         >         192.0.2.42: IPMI RMCPPLUS Payload:
>         >         192.0.2.42: ----------------------
>         >         192.0.2.42: [  BYTE ARRAY ... ] =
>         confidentiality_header[16B]
>         >         192.0.2.42: [ 81h FBh 54h DFh E1h BFh 56h 47h ]
>         >         192.0.2.42: [ 87h 88h EAh 7Bh A1h 54h 37h 5Bh ]
>         >         192.0.2.42: [  BYTE ARRAY ... ] = payload_data[ 9B]
>         >         192.0.2.42: [ 81h B4h CBh 20h 7Ch 08h 00h DCh ]
>         >         192.0.2.42: [ 80h ]
>         >         192.0.2.42: [   6060504030201h] =
>         confidentiality_trailer[56b]
>         >         192.0.2.42: IPMI Message Header:
>         >         192.0.2.42: --------------------
>         >         192.0.2.42: [              81h] = rq_addr[ 8b]
>         >         192.0.2.42: [               0h] = rq_lun[ 2b]
>         >         192.0.2.42: [              2Dh] = net_fn[ 6b]
>         >         192.0.2.42: [              CBh] = checksum1[ 8b]
>         >         192.0.2.42: [              20h] = rs_addr[ 8b]
>         >         192.0.2.42: [               0h] = rs_lun[ 2b]
>         >         192.0.2.42: [              1Fh] = rq_seq[ 6b]
>         >         192.0.2.42: IPMI Command Data:
>         >         192.0.2.42: ------------------
>         >         192.0.2.42: [               8h] = cmd[ 8b]
>         >         192.0.2.42: [               0h] = comp_code[ 8b]
>         >         192.0.2.42: [              DCh] =
>         >         group_extension_identification[ 8b]
>         >         192.0.2.42: IPMI Trailer:
>         >         192.0.2.42: -------------
>         >         192.0.2.42: [              80h] = checksum2[ 8b]
>         >         192.0.2.42: IPMI RMCPPLUS Session Trailer:
>         >         192.0.2.42: ------------------------------
>         >         192.0.2.42: [            FFFFh] = integrity_pad[16b]
>         >         192.0.2.42: [               2h] = pad_length[ 8b]
>         >         192.0.2.42: [               7h] = next_header[ 8b]
>         >         192.0.2.42: [  BYTE ARRAY ... ] =
>         authentication_code[12B]
>         >         192.0.2.42: [ 46h 09h 9Eh F3h 96h BDh 0Ah 22h ]
>         >         192.0.2.42: [ 9Eh E2h 21h 3Dh ]
>         >         ipmi_cmd_dcmi_set_asset_tag: internal IPMI error
>         >          ...
>         >
>         >
>         >
>         >         _______________________________________________
>         >         Freeipmi-devel mailing list
>         >         address@hidden
>         >
>          https://lists.gnu.org/mailman/listinfo/freeipmi-devel
>         >
>         >
>         >
>         > _______________________________________________
>         > Freeipmi-devel mailing list
>         > address@hidden
>         > https://lists.gnu.org/mailman/listinfo/freeipmi-devel
>         --
>         Albert Chu
>         address@hidden
>         Computer Scientist
>         High Performance Systems Division
>         Lawrence Livermore National Laboratory
>         
>         
> 
> 
-- 
Albert Chu
address@hidden
Computer Scientist
High Performance Systems Division
Lawrence Livermore National Laboratory





reply via email to

[Prev in Thread] Current Thread [Next in Thread]