I have CSPC collector version 2.8 and it tries to auto-update to 220.127.116.11. It didn't complete the update and since then, I am unable to connect to the CSPC using SSH (admin).
When I tried, the server just close the connection without any error message.
From the web interface, I can see the status is still "Apply-in-progress".
Anyone could guide me troubleshooting that issue?
Are you able to leverage the console to gain access to the CSPC CLI? If so, can you please login via 'collectorlogin' user, 'su' and input the credentials for 'root' user. As 'root' user, kindly send me the printed output of the LCM apply/install logs with the following command:
#tail -f /opt/LCM/logs/* -- this will provide continuous output (if any) - use Ctrl + C to terminate
#tail -n 25 /opt/LCM/logs/* -- this will provide printed output for last 25 lines of each log
Note: if you have not enabled the 'collectorlogin' or 'root' users by way of the adminshell via 'admin' user, then you will need to do so prior to the aforementioned as it is necessary to enable the respective user shell.
I am also interested in viewing the behavior (printed output) of the following, as 'root' user:
#service sshd restart
And lastly can you send my your sshd_config configuration - full path is:
Please provide the requested info via private message.
unfortunately, I didn't enabled the 'collectorlogin' and 'root' account.
I only have the 'admin' one. Could you please tell me how to proceed now, since I can't access with 'admin'.
As 'admin' user via the CSPC CLI, you can enable the 'collectorlogin' and 'root' users per the following command:
#pwdreset <user> 90 --- where <user> input 'collectorlogin' or 'root'
The system will generate a token credential for use -- please record the temporary password for each newly enabled user. If performed correctly you should be able to login as mentioned in my first reply.
Let me know if you get logged in and send me the requested info. If for any reason you are unable to login and are locked out - you may have to wait 30minutes for each failed login attempt (after the 3rd).
this is the issue, I can't access the server using admin, because the server automatically close the connection.
Therefore, how to run any of the commands you've sent so far?
It is possible that you did not understand my first post -- having access via SSH (remotely) is your issue which I understood. I am asking you to leverage console access via the VM hypervisor and or via the console port of the physical box. In theory, this shoudl still be a viable method to gain access. Let me know if you need additional clarification.
sorry, I missed that part, anyway, I am not the owner of the vsphere, but this morning, I asked the owner to reboot from vsphere, and after rebooting, I tried to connect to ssh from vsphere straightly, and it was the same.
I am able to enter username (admin) and password, but just after the welcome message, it automatically disconnect.
So I don't understand clearly what you were meaning by leverage console access.
If you do not have access to the VM console via Vsphere client then our efforts to troubleshoot this issue will be severely hindered but kindly clarify the following:
You mention that you are not the owner of the vshpere, but that you tried to connect to ssh from vshpere straightly AND you were able to get as far as inputting the username/pw for the 'admin' user but the session terminates immediately -- have I understood correctly?
Assuming this is the case, you are leveraging the remote console functionality of the vsphere client/VM - that said, what may be happening is the following:
You mentioned having restarted the VM after instantiating the upgrade and having found it to be unresponsive with 'apply-in-progress' message via the WebUI. What may be happening is after having restarted while the CSPC was upgrading, it could have caused the adminshell to break in that the upgrade is in two parts - the first deals with adminshell/JeOS components; the second is the CSPC/Linux. Since we cannot verify, per the logs, what the issue is per the install script/LCM apply & install logs, this is speculative at this point.
I will try to replicate this issue and will update you on any find.
thanks a lot for your support.
Yes, you summarized exactly what happened.
I will be waiting for your updates, and bottom line, if we couldn't solve it, I will export all the data and ask our infra team to rebuild the vm using the 2.7.4 OVA file.