Custom Query (98 matches)

Nodes which are successfully booted from the images FEDORA15-STD and FEDORA15-OPENVZ-STD in utah.geniracks.net, appear not to work with the iLO remote console. The java console app reports "No Video".

I've corroborated this just now, on:

• 155.98.34.104, iLO for pc3, which is running FEDORA15-OPENVZ-STD for the experiment emulab-ops/shared-nodes
• 155.98.34.105, iLO for pc4, which is running FEDORA15-STD for the experiment pgeni-gpolab-bbn-com/ecgtest

During testing of IG-ADM-1, i successfully used the remote consoles for pc3 and pc4 at other times. So i believe this is not working right now because of the fedora images, rather than because of a connectivity problem with those particular hosts.

I created the experiment peni-gpolab-bbn-com/ecgtest on Saturday morning, and used it on Saturday and (possibly) yesterday. Today, ​https://boss.utah.geniracks.net/showexp.php3?experiment=952 shows the node status of pcvm3-1 as "up (TBFAILED)", and i can't SSH into the VM. I get the error:

capybara,[~],14:23(0)$ sshtmp -p 30010 pc3.utah.geniracks.net
The authenticity of host 'pc3.utah.geniracks.net (155.98.34.13)' can't be established.
RSA key fingerprint is 6d:1d:76:53:a5:25:99:39:e2:89:ea:b0:99:e3:d3:b9.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added 'pc3.utah.geniracks.net,155.98.34.13' (RSA) to the list of known hosts.
Warning: no access to tty (Bad file descriptor).
                                                Thus no job control in this shell.

and then the shell hangs (sshtmp is an alias for "ssh -o UserKnownHostsFile?=/dev/null".)

Determine what happened to the experiment and why it got into a bad state.

Per ticket:24:1, the canonical locations of src and obj trees for the Emulab software used to build a given rack boss node, should be:

/usr/testbed/{src,obj}/emulab-stable

On boss.instageni.gpolab.bbn.com, /usr/testbed/src/emulab-stable is 10 months old, and /usr/testbed/obj/emulab-stable does not exist:

boss,[~],15:42(0)$ ls -l /usr/testbed/src/
total 8
drwxr-xr-x  53 root  wheel  2048 Mar  7  2012 emulab-devel/
drwxr-xr-x  53 root  wheel  2048 Mar  7  2012 emulab-stable/
drwxrwxr-x  54 282   wheel  3072 Mar  9  2012 testbed/

boss,[~],15:42(0)$ ls -l /usr/testbed/obj/
total 2
drwxr-xr-x  38 root  wheel  1024 Mar  9  2012 testbed/

Make sure the Emulab source used to build a given boss is located on that boss in a canonical location, and ideally that the obj tree for that compile is around somewhere too.