[Freeipmi-devel] Re: ipmi-2.0 calls on ipmi-1.x hardware
Al, it was bug in my code! sorry for the trouble :D avati On Fri, Jun 30, 2006 at 09:06:46AM -0700, Al Chu wrote: > > > Also when you say "subsequent ipmi calls fail" do you mean "return an > > > error" or "don't respond"? > > > > "return an error". ALL subsequent calls, both 2.0 calls and 1.x calls. > > I"m guessing probably the ipmi_device_t got 'closed' somehow? > > What's the completion code? > > Al > > -- > Albert Chu > [EMAIL PROTECTED] > 925-422-5311 > 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] Re: ipmi-2.0 calls on ipmi-1.x hardware
> > Also when you say "subsequent ipmi calls fail" do you mean "return an > > error" or "don't respond"? > > "return an error". ALL subsequent calls, both 2.0 calls and 1.x calls. > I"m guessing probably the ipmi_device_t got 'closed' somehow? What's the completion code? Al -- Albert Chu [EMAIL PROTECTED] 925-422-5311 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] Re: ipmi-2.0 calls on ipmi-1.x hardware
> Hey Anand, > > Are you referring to bmc-config? After bmc-config trys to get ipmi 2.0 > data (i.e. sol info) subsequent calls in bmc-config fail? Is this in- > band or out-of-band? Yes this is about bmc-config. I'm trying everything inband. Strangely the current bmc-config works fine. I"m rewriting bmc-config in C. So i'm guessing i'm missing something. But i see no difference between the current bmc-config and the new one w.r.t ipmi calls. > Also when you say "subsequent ipmi calls fail" do you mean "return an > error" or "don't respond"? "return an error". ALL subsequent calls, both 2.0 calls and 1.x calls. I"m guessing probably the ipmi_device_t got 'closed' somehow? > Also what kind of machine is this? xeon dual core 32bit 3ghz. its the zresearch.com host. > If you are doing this out of band, I'm going to guess that the sequence > numbers are being handled incorrectly after a failure. Subsequent later > packets have duplicated sequence numbers, so the BMC drops the duplicate > packets and never responds. no, i'm doing only inband. > In band, I dunno. I could be something vendor specific. Some vendors > implemented sudo ipmi 2.0 functionality on their ipmi 1.5 machines, so > you might be making semi-successful ipmi calls. the current bmc-config works fine. Most probably its an issue with the code and not the hardware. avati ___ Freeipmi-devel mailing list Freeipmi-devel@gnu.org http://lists.gnu.org/mailman/listinfo/freeipmi-devel
[Freeipmi-devel] Re: ipmi-2.0 calls on ipmi-1.x hardware
Hey Anand, Are you referring to bmc-config? After bmc-config trys to get ipmi 2.0 data (i.e. sol info) subsequent calls in bmc-config fail? Is this in- band or out-of-band? Also when you say "subsequent ipmi calls fail" do you mean "return an error" or "don't respond"? Also what kind of machine is this? If you are doing this out of band, I'm going to guess that the sequence numbers are being handled incorrectly after a failure. Subsequent later packets have duplicated sequence numbers, so the BMC drops the duplicate packets and never responds. In band, I dunno. I could be something vendor specific. Some vendors implemented sudo ipmi 2.0 functionality on their ipmi 1.5 machines, so you might be making semi-successful ipmi calls. Al On Fri, 2006-06-30 at 04:42 -0700, Anand Avati wrote: > Hi, > I observed that when i make a call to any ipmi-2 functionality (like > vlan or sol) running on an older ipmi hardware, then any subsequent > (unrelated) ipmi calls fail. > > if i remove the ipi-2 functionaly calls and run the code again, then the > subsequent functions work successfully, which were previously failing. > > is this expected behaviour? am i supposed to do some kind of check > before making ipmi-2 calls? (if so can the checks be moved to within the > ipmi-2 functinoality calls?) why are subsequent ipmi calls failing? > > regards, > avati > -- Albert Chu [EMAIL PROTECTED] 925-422-5311 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
