Devolutions ForumDevolutions Forum

Getting randomly GATEWAY_OFFLINE messages

Getting randomly GATEWAY_OFFLINE messages

avatar
pvansluisveld

Last week I implemented a new DVLS gateway farm with 2 DVLS gateways on Debian 13.

My config:

{
  "Id": "d174211d-2e50-4d30-93cb-bc0dfec5a595",
  "Hostname": "off-dvls-gw02.linux.interconnect.services",
  "ProvisionerPublicKeyFile": "provisioner.pem",
  "ProvisionerPrivateKeyFile": null,
  "TlsCertificateFile": "server.crt",
  "TlsPrivateKeyFile": "server.key",
  "Listeners": [
    {
      "InternalUrl": "tcp://*:8181",
      "ExternalUrl": "tcp://*:8181"
    },
    {
      "InternalUrl": "https://*:7171",
      "ExternalUrl": "https://*"
    }
  ]
}


2 DVLS gateways are running without any issues and it its possible to connect with RDP and SSH.


For some reason I'm getting random GATEWAY_OFFLINE errors logs in my Devolutions Server.

In de attachment I added my gateway logs. I cannot find the issue, is there something I can do?

a20e6419-4421-4843-8262-e8e9de2cbc1a.png

adcbe990-6a10-4eb6-b881-0d7327e00072.png

gateway(2).log

All Comments (2)

avatar
Michel Audi

Hello pvansluisveld,

Thank you for reaching to our forum. The DVLS Scheduler is repeatedly logging GATEWAY_OFFLINE for the same node: off-dvls-gw02.linux.interconnect.services. Since the farm shows the gateways registered as https://<host>:71710, this typically means DVLS intermittently fails its health/status check to gw02 on port 71710 (timeout, connection reset, or TLS handshake/validation issue). This can still happen even if RDP/SSH sessions appear to work due to brief interruptions and/or failover to the other node.


Could you verify the following using 2–3 specific GATEWAY_OFFLINE timestamps for gw02:

  1. On gw02, check Gateway service logs at the same minute
  • Look for restart/startup messages, TLS/handshake errors, “connection refused”, or sudden disconnects.
  1. On gw02, check system logs at the same minute
  • Look for systemd restarts of the gateway service, OOM-killer events, disk space/filesystem errors, or kernel/network issues.
  1. From the DVLS server, test the exact endpoint DVLS uses
  • Run repeated checks to https://off-dvls-gw02.linux.interconnect.services:71710 and confirm whether you ever see timeouts/resets or TLS validation errors (certificate/SAN/chain/trust).


If you share the timestamps and what you find at those exact times (even if it’s “no restart/no errors”), we can quickly determine whether this is a gateway process interruption on gw02, a network/path issue to :71710, or a TLS/certificate validation problem.

Best regards,

Michel Audi

avatar
Richard Markiewicz

Hello

Just an observation but the shared configuration has the HTTP port on 7171, but the DVLS screenshot shows 71710? That appears wrong - not only it doesn't match the Gateway but it's also not a valid port number (it exceeds 65535).

Can you confirm if this is intentional (perhaps there are other pieces to your setup that we don't see) or a mistake?

Connections like RDP and SSH use the TCP port, so they can still function even with a misconfigured HTTP setup.

Thanks and kind regards,

Richard Markievicz