Worker failed for Proxmox VE

Hi @bravo444 

Welcome to the Community!

I’m not too familiar with the implementation of Proxmox, as yet. Hoping to give it a try soon.

However, I came across this that might help:

Might be worth having a look. 

Additionally, do you have a firewall that could be blocking connections?

Hi dips, 

Thanks for your quick response!  I double checked with firewall, but it should be no problem since the worker host was added in Proxmox VE server successfully.  

Regards,

No problem :)

Have you checked here if your device meets requirements?

Also, I would try re-adding it again and see if that makes any difference. 

I followed an article that you mentioned as above, but I have no luck.

Yes, I checked for requirements and all of them seem to be no problem.  I tried to add and remove the Proxmox server over and over and also tried to configure worker, but none of them have any clues.

I would like to know if “worker” name should have already been exist in the DNS record or not.

Well, I tried to put worker name which exist in DNS and doesn’t exist.  None of them works…

I completey get stuck here…

Regards,

Sorry, I am not able to help any further. 

It might be worth opening a case with Support in the meantime. Additionally, someone might jump in and assist as well when they see this. 

No problem dips and thank you!!

In addition, I found that an article mentioned about “Proxmox Plug-in” setup.  Do I need to install plug-in?  If so, where can I get it?  I just installed Veeam backup and replication software and I didn’t get any plug-in setup window.

Regards,

It should be part of the installation process. However, it should be on the ISO you download. Let me take a look. 

Have a look in this location:

Thank you for your information!  I tried to install these 3 plugin, but all of them have already been installed.  So I think that the plugins should be OK…

Regards,

Welcome @bravo444 

I would recommend working with Veeam Support

Hi dips, 

I am using community edition and I will need to pay for support right?

I am not sure if I can get Veeam support for free or not.

Hey @bravo444 

You can get support for the Community Edition but it will be on a best-case basis so they may not be able to assist straight-away.

Wonder if @regnor has seen this before?

Hi dips,

Do you know where I can get support?  Please let me know about URL for this?

Appreciate it!!

Regards,

Here you go: https://www.veeam.com/support.html

Hello,

I found that the root cause was router setting.  The port 19000 was opened and then the issue was fixed.  However, I got following problem.

“Failed to connect to the worker core service.  Failed to connect to the bakcup appliance.”

I got some log files, but I am not sure which one should be for this error.  

Anyway, I will post this issue as new one.

Hi @bravo444! Please check this helpcenter article regarding the firewall ports necessary for Proxmox backup. 

https://helpcenter.veeam.com/docs/vbproxmoxve/userguide/used_ports.html?ver=1

If you need support, you can open a case at https://my.veeam.com/

Good news!  Finally, I found the root cause and worker was tested successfully!

I apologize that I forgot to restart named service after I added worker hostname in DNS.

After rebooting, the issue was fixed.

I would like to thank you so much for your quick help and resopond!

Regards,

Hi bravo444,
yesterday I have installed Veeam B&R at an environment and get the same issuse:
“...Failed to connect to the worker core service...”

After searching for an solution I found this contribution.

I try to setup DNS settings like you but this wasn´t the solution.

After long trying and reading the log files I see an entry at “C:\ProgramData\Veeam\Backup\Plugins\PVE\workers\worker...\worker_....zip\veeam_backup\test_connection_service\vmb_api\library.log”

There were lines like: “vmbcli   | Register backup server [169.254.245.196]:[10006] Failed.”

I have 4 NICs at this server and only the first one was connected. The others where not connectd but “active”.

I deactivate them and … tada … it works. 

Mybe some others found this while searching for this issue and my contribution would help.

Had this same issue. First step was setting a static IP for the worker but this only slightly improved the issue (at best 2 of 3 VM’s would backup).

After a bit of troubleshooting I found that changing the Bond mode in ProxMox network settings from balance-rr to LACP (802.3ad) resolved the issue.

So I have the same issue. 

I install the worker, it works for 2-3 days and then I get these: Failed to obtain an IP address issues. Upon troubleshooting, I try to start the VM manually, and it doesn’t boot (doesn’t even reach BIOS) and CPU is 90%+.

Console doesn’t even work. 

If I delete and re-install the worker, it works for 2-3 more days and then I receive the same. 

Hello guys.

It’s my first time posting here, to thank everyone but specially to ​@ICT-MT .

I have tried every suggestion here. Firewall, DNS, reinstall… I was about to open a case with Veeam when I saw your post and realized I have the same scenario. Worked like a charm after disabling the unused interfaces. Thank you!

Regards.

I have the same issue, I confirmed the Worker was using a NIC that wasnt connected….cheers for the heads up on that one.

Unfortunately though, deactivating unused NICs didnt work for me.

The Worker continued to using a deactivated NIC.

What did work however was while Veeam was trying to test the connection and the Worker is powered up, I edited the NIC on the fly to one that is in use.  This did the trick and the Worker kept the configuration after retesting.

What helped for me was changing the metric on the NIC that should talk to the worker to 1.

When a Windows machine has multiple network interfaces (NICs) then Windows needs to decide which IP address to use when you do things like: ping %COMPUTERNAME%

Each network interface has a metric value — a number representing its cost or priority.

So when Windows must choose between multiple active interfaces, it prefers the one with the lowest metric. That interface’s IP address becomes the primary one used for things like Outgoing traffic (default route selection), Self-hostname resolution (which IP to advertise) and DNS auto-registration.

I don’t know why in detail but changing this had a positive impact to the worker communication problem.