You are not logged in.
I wasn't trying to tell you how to fix it. That was in case anyone had a similar problem as in "for future reference". Also not suggesting that XSI make my decisions. I was just filling in details. I do think that the error condition should NOT flag the backup as failed. On the plus side, it did cause me to look into what caused the failure.
Just to add a little more information, I only back up a list of VM's, not everything. The error message occurred after each VM backup completed and it listed each VM backup as failed. In spite of the backup being listed as failed, all the backups were successful.
for future reference
vim-cmd /vmsvc/unregister ID# removed the orphaned VM
I had a VM that was listed in the VM database but no longer existed. It caused the backup to show as failed.
The error message is as below. The orphaned VM is VM '140'. In spite of the error messages, the backup was successful and once the offending VM was removed from the database the backup completed and shows success. Just thought you should be aware.
ERROR CLDELSN3, details: [SSL] error: cannot delete snapshot, details: Skipping invalid VM '140' Skipping invalid VM '140'
[img=Error]https://drive.google.com/open?id=1pMS0ytMIV9p_O_LeCLYln8rbRUWTE-7P[/img]
Just to update this post for future reference. In the end, it was just the SMTP port. I had created the scripts and copied the scripts and config files to both servers. This is my home lab and some time elapsed between deploying the script on the first server and the second. I am used to using port 587 rather than the less common 465. I had obviously changed the SMTP port on the first server. Since so much time had elapsed, I didn't think about the port change. Anyway, Gmail port 465 a must.
I have been going through those things. Just wondered if someone had seen the issue and found the answer. Since the same script works with one ESXi host and not the other and both use the same firewall rules, there is most likely a difference at the ESXi level. I will post back when I solved it.
The esxi machine has a hostname. If I open the port and run
openssl s_client -connect smtp.gmail.com:587 -servername smtp.gmail.com
it responds the same way, so I assume is on my side. I just don't know what is wrong and how to solve it. I do believe you are correct that openssl isn't presenting a certificate but which certificate. The contents of the /etc/vmware/ssl directories show the same content so I don't think there is anything missing, but one of them could be broken.
I am testing 2 servers. Using the same script on each server, one sends an email the other does not. The one that sends includes the Gmail TLS session ticket etc. The one that doesn't send shows the following.
Using stored SMTP server info...
Found conf/smtpsrvs file...
2018-03-22T20:50:11| Opening port 587 for SMTPout-587 service...
CONNECTED(00000003)
---
no peer certificate available
---
No client certificate CA names sent
---
SSL handshake has read 7 bytes and written 305 bytes
---
New, (NONE), Cipher is (NONE)
Secure Renegotiation IS NOT supported
Compression: NONE
Expansion: NONE
No ALPN negotiated
SSL-Session:
Protocol : TLSv1.2
Cipher : 0000
Session-ID:
Session-ID-ctx:
Master-Key:
Key-Arg : None
PSK identity: None
PSK identity hint: None
SRP username: None
Start Time: 1521751835
Timeout : 300 (sec)
Verify return code: 0 (ok)
---
Firewall rule SMTPout-587 closed.
Backup finished
Both servers are in the same network, different vlans, and both can reach smtp.gmail.com. Anyone have any ideas.
Thanks
Could you consider making the test mode acknowledgment come up sooner in the script before anything actually begins? There have been several times where I intended to test a change before implementing and missed adding test-mode=true A CTRL C usually kills the script but it would be easier to kill before it has started to create it's a snapshot.
Thank you.
So the two servers must be remotely accessible, which may not be possible. There are two back to back firewalls at one location. I have been launching the free version from a Linux server using a crontab. Can I use the --host option to connect to both locations if I can figure out how to connect to the one remote location.
I have 2 ESXi hosts in two separate sites, 1 server in each site. Do I need a separate license for each of those sites?
Add nohup in front of the ssh command
At least on the Linux side I am able to do this from cron by preceding the ssh command with nohup and using -tt
so
nohup ssh -tt root@server_ip '/vmfs/volumes/backup/XSI/backup-daily.sh'
I don't know if there is anything equivalent in Windows and plink.exe. You might consider the Linux subsystem in Win 10????