Posting Tickets to CORE.SaM


CORE.SAM (this website) is the online support platform for the users of SAM clusters. "CORE" stands for Collaborative Online Resource Exchange. With the word "Resource" we mean knowledge, information, help and support of all sorts. CORE.SAM's objective is to create a collaborative/conversational online environment for the users of our HPC clusters.

The site contains documentation, application forms for SAM accounts, ticketing, and forum based discussion. Users can search for help, request support, and even provide help and share knowledge on a wide range of topics such as basic usage of the cluster or Linux command line in general, caveats and tips for any (installed or not) domain specific software application package, parallel code development and debugging, optimization tips, etc.


When can you post a ticket?

After you have checked your favorite search engine, previous tickets, and with your fellow group members. Posting a ticket should NEVER be your first line of action!

What should I include in my ticket?

As much information as possible! Ask yourself some questions:

  • Does it matter which cluster this should be completed on? (frank, htc, mpi)
  • If your job is failing, where is the job located? (pwd) What cluster? (frank, htc, mpi)
  • Is my title descriptive enough? (Our database is searchable!)
  • Would my PI's name help? (ticket can likely be answered more quickly)
  • Have I provided enough information for someone to reproduce my error?

How should I format my ticket?

We use the markdown language to format tickets, use it! It makes reading your tickets much easier! Reference for the markdown syntax. I will highlight some basics below:

  • Inline code: `command`, looks like: command
  • Code or text blocks:

Some text

looks like:

Some text

echo "Some code"

looks like:

echo "Some code"

Who sees when I post?

See here for the visibility/access settings.

How will I know if my question is answered?

(notification emails, RSS)

Killed message


I'm trying to run DHSVM model from /ihome/xliang/fjp9/2HTGCM_TESTED/dhsvm_srcw and after a few minutes it stops and show a "killed" message:

Initializing dump procedures Accepting dump command for pixel named plot1 Accepting dump command for pixel named MW1 Accepting dump command for pixel named US Accepting dump command for pixel named DS total number of accepted dump pixels 4 Initializing snow map Restoring model state Killed

I'm not sure if this is a server problem. I'd appreciate your help.