Re: [Freeipmi-devel] tweaked Sun workaround text
Al Chu ch...@llnl.gov writes: A. Ok ... then in that case I should put: Sun Fire 2200/4150/4450 with ELOM I didn't understand that it was an ELOM thing. I can't actually demonstrate that on the other systems, but I think it must be the case; I assume ELOM has the same lan bugs on any system that runs it. At least one of the 4150 and 4450 -- I can't recall what I looked up -- is upgradable to ILOM, in which case the workaround will change. I generally dislike documenting version numbers in b/c they can be out of date very quickly. Yes. That was just `it's not fixed as of that version'. Also, how are you getting the firmware version number? B/c the IPMI way to get version numbers doesn't have that much detail. Yes, it's annoying. SSH to the SP says it on login, and `version' has further details: - version SP firmware 2.0.2.5 SP firmware build number: 32265 SP firmware date: Fri Apr 25 20:35:59 PDT 2008 SP filesystem version: 0.1.14 Of course, you can't just do `ssh host version', sigh. The web interface has the same info. On the other hand, ELOM says `Version 3.15' in the login banner and the IPMI info, but this from the CLI: /SP - version SM CLP Version v1.0.0 SM ME Addressing Version v1.0.0 /SP - show -d properties AgentInfo FWVersion /SP/AgentInfo Properties: FWVersion = 3.15 Don't get me started on the inconsistency, and the fact that our x2200s can't be upgraded to ILOM! ___ Freeipmi-devel mailing list Freeipmi-devel@gnu.org http://lists.gnu.org/mailman/listinfo/freeipmi-devel
Re: [Freeipmi-devel] tweaked Sun workaround text
Al Chu ch...@llnl.gov writes: Hey Dave, Thanks for some of the clarifications. I'd like to keep the ordering the same b/c it's actually split up IPMI 1.5 vs. IPMI 2.0. Ah, that wasn't immediately obvious. And some of the text was from folks at Sun. So I've done the below patch instead. Look good? I think it's misleading to list `Sun Fire X2200/X4150/X4450' other than as examples. As far as I can tell, this is a generic ELOM thing, and the should say `ELOM'. Also, the `reported to be fixed in a later release' is from at least a year ago, according to CVS, and it hasn't happened yet as far as I can tell; I think it's at least worth saying it's not fixed as of 2.0.2.5. I could see if my hardware-oriented HPC Sun contact knows any more if that would be helpful. ___ Freeipmi-devel mailing list Freeipmi-devel@gnu.org http://lists.gnu.org/mailman/listinfo/freeipmi-devel
Re: [Freeipmi-devel] tweaked Sun workaround text
Hey Dave, On Wed, 2009-06-10 at 15:06 +0100, Dave Love wrote: Al Chu ch...@llnl.gov writes: Hey Dave, Thanks for some of the clarifications. I'd like to keep the ordering the same b/c it's actually split up IPMI 1.5 vs. IPMI 2.0. Ah, that wasn't immediately obvious. And some of the text was from folks at Sun. So I've done the below patch instead. Look good? I think it's misleading to list `Sun Fire X2200/X4150/X4450' other than as examples. As far as I can tell, this is a generic ELOM thing, and the should say `ELOM'. A. Ok ... then in that case I should put: Sun Fire 2200/4150/4450 with ELOM I didn't understand that it was an ELOM thing. Also, the `reported to be fixed in a later release' is from at least a year ago, according to CVS, and it hasn't happened yet as far as I can tell; Fair point. It's a phrase that I've added to many workarounds as long as the vendor says that they are working on it. I think I'll remove it. I think it's at least worth saying it's not fixed as of 2.0.2.5. I generally dislike documenting version numbers in b/c they can be out of date very quickly. Also, how are you getting the firmware version number? B/c the IPMI way to get version numbers doesn't have that much detail. # /usr/sbin/bmc-info | grep sion Device Revision: 1 Firmware Revision: 2.00 IPMI Version: 2.0 Aux Firmware Revision Info: 302h I could see if my hardware-oriented HPC Sun contact knows any more if that would be helpful. Ok, hopefully we can get more clarifications. Al -- Albert Chu ch...@llnl.gov Computer Scientist High Performance Systems Division Lawrence Livermore National Laboratory ___ Freeipmi-devel mailing list Freeipmi-devel@gnu.org http://lists.gnu.org/mailman/listinfo/freeipmi-devel
[Freeipmi-devel] tweaked Sun workaround text
As far as I can tell, the Sun workarounds are generic in ELOM and ILOM, and haven't been fixed in the current versions. I have x4100 with ILOM 1.0 and 2.0, x2200M2 with ELOM 2.0, and x4200m2 and x4500 with ILOM 2.0. All seem to need the workarounds, and have the latest firmware apart from ILOM 1.0. I think this change to the text makes it clearer, and it's worth keeping the Sun bits together. By the way, ELOM is EOLed -- we're stuck with its problems on most of our kit, sigh -- but there is an ILOM 3 in the works; perhaps that fixes the bugs, and I'll try to check if other people here get newer kit with it on. 2009-06-09 Dave Love f...@gnu.org * common/man/manpage-common-workaround-text.man: Tweak the Sun stuff. *** manpage-common-workaround-text.man 08 Apr 2009 21:47:05 +0100 1.6 --- manpage-common-workaround-text.man 09 Jun 2009 14:40:00 +0100 *** *** 41,54 username invalid or k_g invalid errors to occur. In order to work around this issue, the authcap workaround must be specified. .LP ! Sun ILOM 2.0: The session sequence numbers returned for IPMI 1.5 ! sessions are the wrong endian on some systems running ILOM 2.0. The incorrect endian depends on the service processor endianness. This will likely cause session timeout errors to occur. In order to work ! around this issue, the endianseq workaround must be specified. This ! problem is reported to be fixed in a later release. .LP ! Sun Fire X4150/X4450: The motherboard does not properly report username capabilities. This will likely cause username invalid errors to occur. In order to work around this issue, the authcap workaround must be specified. --- 41,63 username invalid or k_g invalid errors to occur. In order to work around this issue, the authcap workaround must be specified. .LP ! Sun ILOM 1.0 and 2.0: The session sequence numbers returned for IPMI 1.5 ! sessions are the wrong endian on some systems running ILOM 1.0 and 2.0 ! (at least up to 2.0.2.5). The incorrect endian depends on the service processor endianness. This will likely cause session timeout errors to occur. In order to work ! around this issue, the endianseq workaround must be specified. ! .\ This problem is reported to be fixed in a later release. .LP ! Sun ILOM (at least up to 2.0.2.5) with IPMI 2.0: ! There are several Sun IPMI 2.0 bugs. These ! problems may cause password invalid or bmc error errors to occur. ! They can be worked around by specifying the sun20 workaround. The ! workarounds include handling invalid lengthed hash keys and invalid ! cipher suite records. ! .LP ! Sun ELOM (seen on at least X2200/X4150/X4450): The motherboard does ! not properly report username capabilities. This will likely cause username invalid errors to occur. In order to work around this issue, the authcap workaround must be specified. *** *** 68,79 cause password invalid errors to occur. These compliance bugs are confirmed to be fixed on newer firmware. .LP - Sun Fire 4100 with ILOM: There are several Sun IPMI 2.0 bugs. These - problems may cause password invalid or bmc error errors to occur. - They can be worked around by specifying the sun20 workaround. The - workarounds include handling invalid lengthed hash keys and invalid - cipher suite records. - .LP Inventec 5441: The privilege level sent during the Open Session stage of an IPMI 2.0 connection is used for hashing keys instead of the privilege level sent during the RAKP1 connection stage. This may --- 77,82 ___ Freeipmi-devel mailing list Freeipmi-devel@gnu.org http://lists.gnu.org/mailman/listinfo/freeipmi-devel
Re: [Freeipmi-devel] tweaked Sun workaround text
Hey Dave, Thanks for some of the clarifications. I'd like to keep the ordering the same b/c it's actually split up IPMI 1.5 vs. IPMI 2.0. And some of the text was from folks at Sun. So I've done the below patch instead. Look good? Al On Tue, 2009-06-09 at 15:02 +0100, Dave Love wrote: As far as I can tell, the Sun workarounds are generic in ELOM and ILOM, and haven't been fixed in the current versions. I have x4100 with ILOM 1.0 and 2.0, x2200M2 with ELOM 2.0, and x4200m2 and x4500 with ILOM 2.0. All seem to need the workarounds, and have the latest firmware apart from ILOM 1.0. I think this change to the text makes it clearer, and it's worth keeping the Sun bits together. By the way, ELOM is EOLed -- we're stuck with its problems on most of our kit, sigh -- but there is an ILOM 3 in the works; perhaps that fixes the bugs, and I'll try to check if other people here get newer kit with it on. ___ Freeipmi-devel mailing list Freeipmi-devel@gnu.org http:// lists.gnu.org/mailman/listinfo/freeipmi-devel -- Albert Chu ch...@llnl.gov Computer Scientist High Performance Systems Division Lawrence Livermore National Laboratory ? Makefile ? Makefile.in ? aclocal.m4 ? autom4te.cache ? conf27923.sh ? config.log ? config.status ? configure ? cscope.files ? cscope.out ? freeipmi-0.8.0.alpha0.tar.gz ? freeipmi-may-29-head.patch ? freeipmi.spec ? libtool ? bmc-config/Makefile ? bmc-config/Makefile.in ? bmc-config/bmc-config.8 ? bmc-config/bmc-config.8.pre ? bmc-config/bmc-config.conf.5 ? bmc-config/bmc-config.conf.5.pre ? bmc-config/src/.deps ? bmc-config/src/.libs ? bmc-config/src/Makefile ? bmc-config/src/Makefile.in ? bmc-config/src/bmc-config ? bmc-config/src/foo ? bmc-device/Makefile ? bmc-device/Makefile.in ? bmc-device/bmc-device.8 ? bmc-device/bmc-device.8.pre ? bmc-device/src/.deps ? bmc-device/src/.libs ? bmc-device/src/Makefile ? bmc-device/src/Makefile.in ? bmc-device/src/bmc-device ? bmc-info/Makefile ? bmc-info/Makefile.in ? bmc-info/bmc-info.8 ? bmc-info/bmc-info.8.pre ? bmc-info/src/.deps ? bmc-info/src/.libs ? bmc-info/src/Makefile ? bmc-info/src/Makefile.in ? bmc-info/src/bmc-info ? bmc-info/src/woprjr0-lt-bmc-info-28708.core ? bmc-info/src/woprjr0-lt-bmc-info-28969.core ? bmc-watchdog/Makefile ? bmc-watchdog/Makefile.in ? bmc-watchdog/bmc-watchdog.8 ? bmc-watchdog/bmc-watchdog.8.pre ? bmc-watchdog/src/.deps ? bmc-watchdog/src/.libs ? bmc-watchdog/src/Makefile ? bmc-watchdog/src/Makefile.in ? bmc-watchdog/src/bmc-watchdog ? common/Makefile ? common/Makefile.in ? common/man/Makefile ? common/man/Makefile.in ? common/man/foo.patch ? common/man/sun.patch ? common/src/Makefile ? common/src/Makefile.in ? common/src/configtool/.deps ? common/src/configtool/.libs ? common/src/configtool/Makefile ? common/src/configtool/Makefile.in ? common/src/configtool/libconfigtoolcommon.la ? common/src/configtool/libconfigtoolcommon_la-config-tool-argp.lo ? common/src/configtool/libconfigtoolcommon_la-config-tool-checkout.lo ? common/src/configtool/libconfigtoolcommon_la-config-tool-comment.lo ? common/src/configtool/libconfigtoolcommon_la-config-tool-commit.lo ? common/src/configtool/libconfigtoolcommon_la-config-tool-diff.lo ? common/src/configtool/libconfigtoolcommon_la-config-tool-parse.lo ? common/src/configtool/libconfigtoolcommon_la-config-tool-pef-conf-section.lo ? common/src/configtool/libconfigtoolcommon_la-config-tool-section.lo ? common/src/configtool/libconfigtoolcommon_la-config-tool-utils.lo ? common/src/configtool/libconfigtoolcommon_la-config-tool-validate.lo ? common/src/debugutil/.deps ? common/src/debugutil/.libs ? common/src/debugutil/Makefile ? common/src/debugutil/Makefile.in ? common/src/debugutil/libdebugutil.la ? common/src/debugutil/libdebugutil_la-debug-util.lo ? common/src/miscutil/.deps ? common/src/miscutil/.libs ? common/src/miscutil/Makefile ? common/src/miscutil/Makefile.in ? common/src/miscutil/cbuf.lo ? common/src/miscutil/conffile.lo ? common/src/miscutil/error.lo ? common/src/miscutil/fd.lo ? common/src/miscutil/hash.lo ? common/src/miscutil/hostlist.lo ? common/src/miscutil/libmiscutils.la ? common/src/miscutil/list.lo ? common/src/miscutil/secure.lo ? common/src/miscutil/thread.lo ? common/src/miscutil/timeval.lo ? common/src/pingtool/.deps ? common/src/pingtool/.libs ? common/src/pingtool/Makefile ? common/src/pingtool/Makefile.in ? common/src/pingtool/libpingtoolcommon.la ? common/src/pingtool/libpingtoolcommon_la-ping-tool-common.lo ? common/src/portability/.deps ? common/src/portability/.libs ? common/src/portability/Makefile ? common/src/portability/Makefile.in ? common/src/portability/freeipmi-portability.lo ? common/src/portability/libfreeipmiportability.la ? common/src/toolcommon/.deps ? common/src/toolcommon/.libs ? common/src/toolcommon/Makefile ? common/src/toolcommon/Makefile.in ? common/src/toolcommon/libtoolcommon.la ? common/src/toolcommon/libtoolcommon_la-tool-cmdline-common.lo ?
