Hi All,
I'm trying to perform a scripted installation of VMware vCSA 6.0 using the new vcsa-deploy but I'm hitting a strange error that I'm not quite sure how to resolve (the last entry in the log says permission denied which is weird).
The process I have been through is:
Overview:
Error:
Start vCSA command line installer to deploy vCSA "Server01", an embedded node.
Please see /tmp/vcsa-cli-installer-P2jqzslog for logging information.
Run installer with "-v" or "--verbose" to log detailed information.
The SSO password meets the installation requirements.
Traceback (most recent call last):
File "/build/toolchain/lin64/cx-Freeze-4.3.3/lib/python2.7/site-packages/cx_Freeze/initscripts/Console.py", line 27, in <module>
File "install.py", line 198, in <module>
File "install.py", line 106, in check_database
File "/build/toolchain/lin64/python-2.7.1/lib/python2.7/subprocess.py", line 672, in __init__
File "/build/toolchain/lin64/python-2.7.1/lib/python2.7/subprocess.py", line 1202, in _execute_child
OSError: [Errno 13] Permission denied
Note 1: I have tried running this from under both the vi-admin and root accounts with the same result.
Note 2: I appreciate you can run this from either an OSX or Windows box but I need to deploy this from a Linux server and SUSE11SP3 seems to be the only officially supported platform hence you probably guessed that I am running this on the vMA 6 server.
Any thoughts or suggestions would be appreciated ![]()
Cheers
Andy
Looks like this last issue was the DNS test which was failing, so lesson learned from this adventure, if you are deploying using the new scripted install, validate your parameters first in the GUI.
Case Closed; hope this adventure helps someone else
In follow up to this when enabling verbose logs and tailing what it has written, there are no obvious error messages however the last command executed was:
vCSASCliInstallLogger - DEBUG - Run command: /home/vi-admin/vmware-vcsa/vcsa-cli-installer/lin64/jre/bin/java -jar /home/vi-admin/vmware-vcsa/vcsa-cli-installer/dbchecker/dbchecker.jar servername/VCDB vpxuser ******
Taking this into account:
The SSO password meets the installation requirements.
Error: failed to read from file: vmware-vcsa
Completed with errors
Gathering VC support log bundle. This can take a few minutes.
Failed to gather VC support bundle. Details: Failed to find VM: ServerName
Checking my log file didn't reveal anything useful but searching for the above file which failed pointed me to what is/was the OVA specified in my json file, looks like I fat fingered the path hence the error message.
Re-running the command after making the relevant changes and off it went, ovftool kicked in and started to deploy; in theory I should have a deployed vCenter server soon ![]()
Root Cause: human error
With that said, the scripted installer successfully deploys the vCSA appliance, powers it up and starts to configure it but at 95% it fails; error log below:
Progress: 5%. Setting up storage
Progress: 55%. Installed VMware-unixODBC-2.3.1.vmw.2-6.0.0.x86_64.rpm
Progress: 62%. Installed oracle-instantclient11.2-odbc-11.2.0.2.0.x86_64.rpm
Progress: 77%. Installed VMware-invsvc-6.0.0-2562558.x86_64.rpm
Progress: 79%. Installed VMware-vpxd-6.0.0-2559267.x86_64.rpm
Progress: 83%. Installed VMware-vpxd-sim-6.0.0-2559267.x86_64.rpm
Progress: 86%. Installed VMware-sps-6.0.0-2559267.x86_64.rpm
Progress: 89%. Installed VMware-vsm-6.0.0-2559267.x86_64.rpm
Progress: 95%. Configuring the machine
Gathering VC support log bundle. This can take a few minutes.
VC Support log bundle path: /tmp/vc-support-bundle.tgz
Looking through the vc-support-bundle logs, the only thing that stands out is:
cmd /opt/VMware/bin/sso-config.sh get_identity_sources $VMWARE_LOG_DIR timed out after 60.001019001 seconds due to lack of progress in last 3 seconds (0 bytes read)
Investigation continues...
Taking a step back and re-running my deployment using the GUI and the same input params resulted in similar effects but instead provided a bit more information that may or may not be related to the scripted deployment but I'm assuming it's the former at the moment; error message stated that
Firstboot script execution error.
The supplied system name servername.domain.com is not valid
If the supplied system name is an FQDN, then make sure the DNS forward lookup results in at least one valid IP address in the system. If the supplied system name is an IP Address, then it should be one of the valid IP Address(es) in the system.
Time for a re-run with the system name as an IP Address and then have a poke around at DNS.
Looks like this last issue was the DNS test which was failing, so lesson learned from this adventure, if you are deploying using the new scripted install, validate your parameters first in the GUI.
Case Closed; hope this adventure helps someone else
hello all
i have the same error. and DNS server its ok ts.loc and appliance is vapp.ts.loc
its a mistety ...
![]()
Firstboot script execution error.
The supplied System Name vcenter.ts.loc is not valid.
If the supplied system name is a FQDN, then make sure the DNS forward lookup results in at least one valid IP address in the system. If the supplied system name is an IP address, then it should be one of the valid IP address(es) in the system.
Hi Tsistemas,
I haven't done any more investigation into this issue, instead I just worked around it using the IP Address as above; at some point I will need to replace it with the FQDN of the box though.
If I get chance next week I will try and repro with the relevant DNS Server behind it and see if the issue still occurs.
Regards
Andy
