You can start a training with sccBmc -i from the command line and when you do you get the menu below. You are given four choices of initial frequencies.
Please select from the following possibilities:
INFO: (0) Tile533_Mesh800_DDR800
INFO: (1) Tile800_Mesh1600_DDR1066
INFO: (2) Tile800_Mesh1600_DDR800
INFO: (3) Tile800_Mesh800_DDR1066
INFO: (4) Tile800_Mesh800_DDR800
INFO: (others) Abort!
Make your selection:
If you use the GUI, click on the symbol that looks like a radiation hazard sign. We sometimes refer to this as the Nuke button.
It takes a few minutes ... not inordinately long. I haven't timed it.
i am trying to initialize the SCC but i keep getting this error message after training:
ERROR: Training unsuccessful: Too small optimal data eye: optimal delay setting is 063 in a 005 wide range, select slower/different SIF phy operation speed ***
ERROR: Failed to trim SIF! Please try to reload the driver or powercycle SCC board in case of repeated failure!
INFO: System initialization done.
i have done powercycle but it still gives me this error. Should i do something else beside this?
Do you have your own MCPC/RockyLake or are you using a machine in our SCC Data Center?
If you have your own system, I'd suggest trying the steps in http://communities.intel.com/docs/DOC-5946
This is taking the system all the way down ... no power to anything ... and then bringing it back up.
If you are using the Data Center, you can try the sccPowercycle in /opt/sccKit/current/bin. You may or may not have permission to run that command. If you run it and still see no improvement or you cannot run it, file a bug under the category "Marc Administration Needed" in our Bugzilla
http://marcbug.scc-dc.com/bugzilla3/ Bugs in thisProduct Category are treated as job tickets for the team that administers teh Data Center.
If yiou repeatedly cannot train, this may unfortunately be a hardware problem. Has this chip successfully trained in the past?
Oh, if you are using an external system and you repeatedly cannot train, please send me a private message and we can correspond offline about possible hardware replacement.
No there is no hardware problem the issue was that the MCPC lost the configurations of the BMC so i had to re-install them and now it is ok up and running. Thanks for the response and the help
Can you be more specific about the configurations you lost and how you restored them? I don't see how one can lose BMC configurations. Was this eth1 configuration? sccKit?
I tried to boot linux on the SCC but it said that re-initialization was needed. I tried to re-initialize the BMC but it was unsuccesful. I found out that the PCIe driver wasnt configured correctly so i had to reinstall it and follow the whole procedure as described in the documents. After that it is ok. Maybe i didnt described my problem correctly sorry
By reinstalling do you mean actually running install.csh again? In /opt/sccKit/current/firmware. When we do that, we usually remove crbif-dkms first as in
sudo apt-get remove crbif-dkms
The reason I'm asking is that we consider not being able to train pretty serious and would very much like to record how users fix that problem. Thanks.
Exactly i actually had to rerun the install.csh. I had some problems with the drkms and as you describe i had to unistall it and install it again (using the synaptic manager) in order to have a succesfull installation of the PCIe driver. I realised that the MCPC Ubuntu had crashed problems with the drkms and that was the reason why i had these dificulties.