Ah! .. I see. Seems like I started with a wrong assumption. I didn't not do the capture image process yet. I wanted to see if I can make a reservation for the already existing VM, that I built with the name 'vmwarewinxp-base7-v0' and placed it under the Data-Store-Path on the VMhost that built it.
Yes you were right I was missing the IP address in the Data-Store Path: I will put in the VMhost IP here. In that case shouldn't I generate-node-key's for all the VMhost's because it will try to SSH to it to get to the Data-Store-Path? Thank you Andy for the help ... I will give image capture a try first. Amit > -----Original Message----- > From: Andy Kurth [mailto:andy_ku...@ncsu.edu] > Sent: Thursday, March 18, 2010 10:56 AM > To: vcl-dev@incubator.apache.org > Subject: Re: Reservation Error: > > > 376 |16651|5:5|new| ---- WARNING ---- ^M > > 377 |16651|5:5|new| 2010-03-17 > 13:02:45|16651|5:5|new|DataStructure.pm:_automethod(675)|corresponding data > has not > been initialized for get_computer_pref erredimage_id: > $self->request_data- > >{reservation}{5}{computer}{preferredimage}{id}^M > > 378 |16651|5:5|new| ( 0) utils.pm, notify (line: 737)^M > > 379 |16651|5:5|new| (-1) DataStructure.pm, _automethod (line: 675)^M > > 380 |16651|5:5|new| (-2) Autoload.pm, __ANON__ (line: 80)^M > > 381 |16651|5:5|new| (-3) new.pm, process (line: 106)^M > > 382 |16651|5:5|new| (-4) vcld, make_new_child (line: 594)^M > > 383 |16651|5:5|new| (-5) vcld, main (line: 341)^M > > > > #>>> Not sure if I am missing something here. > > This isn't a problem. It occurs because computer.preferredimageid isn't set > in > the database. It should get set automatically after a successful reservation > is > made on the computer. > > > 475 2010-03-17 > > 13:02:46|16651|5:5|new|utils.pm:run_ssh_command(6180)|executing > SSH command on resxi-2:^M > > 476 |16651|5:5|new| /usr/bin/ssh -i /etc/vcl/vcl.key -l root -p 22 -x > > resxi-2 'cat > currentimage.txt' 2>&1^M > > 477 2010-03-17 > 13:02:47|16651|5:5|new|utils.pm:run_ssh_command(6262)|run_ssh_command > output:^M > > 478 |16651|5:5|new| cat: currentimage.txt: No such file or directory^M > > > > #>>> currentimage.txt : Is this something that we have to create at > > anytime? I am > thinking it is created automatically when a image is loaded for the first > time. > > This appears to be a problem. You never have to manually create/alter > currentimage.txt. It should be created automatically during the image capture > process and saved within the image. This file is used when a reservation is > made to determine if the correct image is loaded on the computer in order to > know if the computer needs to be reloaded or not. The computer will be > reloaded > every time a reservation is made if the file is missing or incorrect, even if > the correct image is loaded. > > Check if currentimage.txt exists in root's home directory on that computer. > You > should be able to shell into it from the management node via: > ssh -i /etc/vcl/vcl.key -l root -p 22 -x resxi-2 > > Then run "ls -l ~". What do you see? > > If you still have the vcld.log output from the image capture process, search > it > for "currentimage.txt" to see if anything went wrong. > > > 516 2010-03-17 > > 13:02:47|16651|5:5|new|utils.pm:run_ssh_command(6180)|executing > SSH command on /install/image/nfs1:^M > > 517 |16651|5:5|new| /usr/bin/ssh -i /etc/vcl/vcl.key -l root -p 22 -x > > /install/image/nfs1 > 'ls -1 /golden 2>&1' 2>&1^M > > > > #>>> This is where my real issues are: I realized reading the README for > > esx and > esxthin, that I need to have a VCL mount point and a subdirectory /golden for > my images > > But what I am missing here is why is it trying to ssh -x > > /install/image/nfs1 which is > actually the Data Store path. Isn't it supposed to ssh to my VMhost machine. > > Also when does VCL code use the esxthin.pm, as opposed to esx.pm? > > The previous two questions are related. I'm not too familiar with the esx.pm > module but you cannot ignore this. From looking at the code, the warning is > being displayed because there's a problem with your VM host profile's data > store > path value. It's looking for this setting to be in the format: > <IP address>:<datastore path> > > Check this setting: > Virtual Hosts > VM Host Profiles > Configure Profile > Data Store Path > > I'm guessing it doesn't include the IP address where the golden images reside > or > a colon. The "ls -1" SSH command above is being incorrectly assembled because > of this. It's inserting "/install/image/nfs1" where the IP address should go. > Once the data store path is changed to include "<IP address>:", the command > should be assembled correctly as: > /usr/bin/ssh -i /etc/vcl/vcl.key -l root -p 22 -x <IP address> 'ls -1 > /install/image/nfs1/golden 2>&1' > > The rest of the SSH errors are occurring because it's attempting to connect to > the computer with the address "/install/image/nfs1". > > > 518 ^M > > 519 |16651|5:5|new| ---- WARNING ---- ^M > > 520 |16651|5:5|new| 2010-03-17 > 13:02:47|16651|5:5|new|utils.pm:run_ssh_command(6252)|attempt 1/3: failed to > execute > SSH command on /install/image/nfs1: l s -1 /golden 2>&1, exit status: > 255, SSH exits > with the exit status of the remote command or with 255 if an error occurred, > output:^M > > 521 |16651|5:5|new| ssh output (ls -1 /gol...): ssh: > > /install/image/nfs1: Name or service > not known^M > > 522 |16651|5:5|new| ( 0) utils.pm, notify (line: 737)^M > > 523 |16651|5:5|new| (-1) utils.pm, run_ssh_command (line: 6252)^M > > 524 |16651|5:5|new| (-2) esx.pm, does_image_exist (line: 910)^M > > 525 |16651|5:5|new| (-3) new.pm, reload_image (line: 618)^M > > 526 |16651|5:5|new| (-4) new.pm, process (line: 266)^M > > 527 |16651|5:5|new| (-5) vcld, make_new_child (line: 594)^M > > 528 |16651|5:5|new| (-6) vcld, main (line: 341)^M > > > > Thank you, > > Amit > > > > > > Hope this helps, > Andy
