SABnzbd Forums

I am running SABnzbd in Docker on a Debian 13 server. It has been running fine for months. I recently updated from Debian 12 to 13. I installed a new, higher performance firewall. I pulled the latest image of SABnzbd (4.5.3).

The issue I'm having is for the first 10 minutes or so, downloads and my connections to news servers are working fine. Then, all of the sudden, my SSL connections start falling away and becoming extremely sporadic. Sporadic in the sense that my connections with a server will go from 3 to 0 to 5 to 2 to 7 to 0. Also, when viewing the current connections during a download, the SSL "check mark" will remain, but the TLS1.2 and cipher will blink on and off sporadically. I have a clip of what is happening, but am not allowed to post links. I know there are a lot of unknowns, but does this sporadic behavior scream anything in particular to anyone? I am also experiencing SSL connection timeout errors.

A restart of the SABnzbd container fixes this issue temporarily. The longer I wait after bringing the container back up, the longer the service will go before experiencing these issues.

If anyone has any idea what may be happening, please let me know. If anyone has any questions, don't hesitate to ask.

Thanks!

> SSL Connections Timing Out

What if you don't use SSL?

> I installed a new, higher performance firewall.

What if you deactivate the firewall? And check the logging of your firewall.

> I am running SABnzbd in Docker on a Debian 13 server

Is this on a home network? If so: what if your SABnzbd on another system

My leading suspicion is Debian, Docker, or the container -- rather than the network. Due to the simple fact that restarting the container temporarily solves the issue until the timeouts start again. But, killing the queue for a while, than starting a brand new download (to rule out the possibility that it is that particular download) does not make anything better.

Also note that this happens on connections to multiple news servers.

sander wrote: ↑October 9th, 2025, 11:13 am > SSL Connections Timing Out

What if you don't use SSL?

> I installed a new, higher performance firewall.

What if you deactivate the firewall? And check the logging of your firewall.

> I am running SABnzbd in Docker on a Debian 13 server

Is this on a home network? If so: what if your SABnzbd on another system

1. SSL turned off completely mitigates the issue.

2. Firewall is not showing anything outbound being blocked.

3. Yes, on a home network. Will need to set it up on another system at another time.

> 3. Yes, on a home network. Will need to set it up on another system at another time.

Preferrably straight on OS. And then try with and without SSL.

Okay, tried it on bare Windows 11. The issue did not happen at all. However, even with the same exact SABnzbd settings, my download speed was 15% of what it is in the docker container. Both of these devices are on the same network on the same switch.

Here is a snippet of the container log from docker when the issue happens:

2025-10-09T22:07:43.211764915Z 2025-10-09 17:07:43,211::INFO::[downloader:751] Connecting 9@news.newshosting.com finished
2025-10-09T22:07:43.280703876Z 2025-10-09 17:07:43,279::INFO::[downloader:616] 2@news.newshosting.com: Initiating connection
2025-10-09T22:07:43.288788525Z 2025-10-09 17:07:43,288::INFO::[downloader:949] Thread 7@news.newshosting.com: Timed out
2025-10-09T22:07:43.403912988Z 2025-10-09 17:07:43,403::INFO::[newswrapper:351] 2@news.newshosting.com: Connected using TLSv1.2 (AES128-SHA)
2025-10-09T22:07:43.508415925Z 2025-10-09 17:07:43,507::INFO::[downloader:751] Connecting 2@news.newshosting.com finished
2025-10-09T22:07:43.592599264Z 2025-10-09 17:07:43,591::INFO::[downloader:616] 4@news.newsgroup.ninja: Initiating connection
2025-10-09T22:07:43.718962587Z 2025-10-09 17:07:43,718::INFO::[newswrapper:351] 4@news.newsgroup.ninja: Connected using TLSv1.2 (AES128-SHA)
2025-10-09T22:07:43.835220379Z 2025-10-09 17:07:43,834::INFO::[downloader:751] Connecting 4@news.newsgroup.ninja finished
2025-10-09T22:07:44.338648041Z 2025-10-09 17:07:44,338::INFO::[downloader:949] Thread 1@news.newshosting.com: Timed out
2025-10-09T22:07:47.489632305Z 2025-10-09 17:07:47,489::INFO::[downloader:949] Thread 8@news.newshosting.com: Timed out
2025-10-09T22:07:48.337808688Z 2025-10-09 17:07:48,337::INFO::[downloader:616] 9@news.newshosting.com: Initiating connection
2025-10-09T22:07:48.464608106Z 2025-10-09 17:07:48,463::INFO::[newswrapper:351] 9@news.newshosting.com: Connected using TLSv1.2 (AES128-SHA)
2025-10-09T22:07:48.597220347Z 2025-10-09 17:07:48,596::INFO::[downloader:751] Connecting 9@news.newshosting.com finished
2025-10-09T22:07:48.730412837Z 2025-10-09 17:07:48,729::INFO::[downloader:616] 2@news.newshosting.com: Initiating connection
2025-10-09T22:07:48.856361368Z 2025-10-09 17:07:48,855::INFO::[newswrapper:351] 2@news.newshosting.com: Connected using TLSv1.2 (AES128-SHA)
2025-10-09T22:07:48.996386377Z 2025-10-09 17:07:48,995::INFO::[downloader:751] Connecting 2@news.newshosting.com finished
2025-10-09T22:07:49.053564966Z 2025-10-09 17:07:49,053::INFO::[downloader:616] 4@news.newsgroup.ninja: Initiating connection
2025-10-09T22:07:49.178453572Z 2025-10-09 17:07:49,178::INFO::[newswrapper:351] 4@news.newsgroup.ninja: Connected using TLSv1.2 (AES128-SHA)
2025-10-09T22:07:49.282345460Z 2025-10-09 17:07:49,281::INFO::[downloader:751] Connecting 4@news.newsgroup.ninja finished
2025-10-09T22:07:50.573453554Z 2025-10-09 17:07:50,573::INFO::[downloader:949] Thread 10@news.newsgroup.ninja: Timed out
2025-10-09T22:07:53.957791367Z 2025-10-09 17:07:53,957::INFO::[downloader:616] 9@news.newshosting.com: Initiating connection
2025-10-09T22:07:54.085881398Z 2025-10-09 17:07:54,084::INFO::[newswrapper:351] 9@news.newshosting.com: Connected using TLSv1.2 (AES128-SHA)
2025-10-09T22:07:54.170842220Z 2025-10-09 17:07:54,170::INFO::[downloader:751] Connecting 9@news.newshosting.com finished

Here is an example of what you see from the web GUI:

> Okay, tried it on bare Windows 11. The issue did not happen at all. However, even with the same exact SABnzbd settings, my download speed was 15% of what it is in the docker container.

What speed did you get on Windows?

And, on Windows:

While SABnzbd is not doing anything: In SABnzbd's upper right corner, click on the wrench symbol ("Status and interface options"), then click on first tab Status, and there click on the Refresh Arrow. Post the values of the lower part here.

Then, still on Windows:

In SABnzbd's upper right corner, click on the wrench symbol ("Status and interface options"), then click on first tab Status, click Test download 1GB. After downloading, in history, at the download in the drop down, what was the resulting speed? Example "Download Downloaded in 55 seconds at an average of 22.9 MB/s"





> Both of these devices are on the same network on the same switch.

And what are the differences? That firewall of you: on the docker system (and thus EDIT: NOT affecting Windows), or on your LAN (thus affecting Windows)?

> What speed did you get on Windows?
Must've just needed to restart it yesterday. Today, I ran 3 10GB tests. All 3 kept a STEADY 73MB/s speed with very little fluctuation. No errors.

My Windows machine is MUCH better in terms of hardware specs:
System performance (Pystone) 1004030 AMD Ryzen 5 9600X 6-Core Processor
Download folder speed 1458.2 MB/s C:\Downloads\incomplete
Complete folder speed 1404.5 MB/s C:\Downloads\complete
Internet Bandwidth 72.14 MB/s 577.12 Mbps
Platform Windows-11

My Linux machine that's having the SSL issues:
Download speed limited by Disk speed (27x)
System performance (Pystone) 246900 AMD Ryzen 5 PRO 2400G with Radeon Vega Graphics
Download folder speed 57.6 MB/s /data/usenet/incomplete
Complete folder speed 64.5 MB/s /data/usenet/complete
Internet Bandwidth 71.19 MB/s 569.52 Mbps

Speed is not a concern here really. Like I said, Docker (on Linux) has been working fine for me for months. No SSL issues whatsoever. So I don't believe it is a hardware limitation. However, thanks to your great questions, sander, it is extremely obvious now that it is my Linux/Docker machine that is causing the issue. Where the issue lies within my configuration is less obvious.

So here's a bit about my configuration: SABnzbd is in a Docker Compose container, that container is on a bridge network with the starr applications. All web applications are accessed via reverse proxy thru NGINX Proxy Manager.

>And what are the differences? That firewall of you: on the docker system (and thus affecting Windows), or on your LAN (thus affecting Windows)?
Not really sure what you mean here. Neither the firewall or switch are doing anything different between the two devices. There are no rules, routes, or services applied to either device specifically.

Lastly, after waiting about 24hrs, I started another test download on the Docker instance and sure enough the issue was present immediately. After restarting the container, the issue went away for a while then the issue began again. See the screenshot below of another error that is thrown every once in a while within the GUI when the issues arise:

https://imgur.com/a/7TLbDkL

So, the issue seems to be not necessarily that the connections are timing out, but that they are quickly being started and finished. I thought this was normal behavior, but it is not. It is normal for these connections to last much longer, sometimes even for the entirety of the download.
However, this is what is happening on my machine:
2025-10-10T18:25:55.999095232Z 2025-10-10 13:25:55,998::INFO::[downloader:616] 6@news.newshosting.com: Initiating connection
2025-10-10T18:25:56.051729992Z 2025-10-10 13:25:56,051::INFO::[downloader:751] Connecting 1@news.newshosting.com finished
2025-10-10T18:25:56.181834565Z 2025-10-10 13:25:56,181::INFO::[newswrapper:351] 6@news.newshosting.com: Connected using TLSv1.2 (AES128-SHA)
2025-10-10T18:25:56.273135962Z 2025-10-10 13:25:56,272::INFO::[downloader:751] Connecting 6@news.newshosting.com finished
2025-10-10T18:25:57.086019030Z 2025-10-10 13:25:57,085::INFO::[downloader:616] 3@news.newshosting.com: Initiating connection
2025-10-10T18:25:57.086705088Z 2025-10-10 13:25:57,086::INFO::[downloader:616] 6@news.newsgroup.ninja: Initiating connection
2025-10-10T18:25:57.207141291Z 2025-10-10 13:25:57,206::INFO::[newswrapper:351] 6@news.newsgroup.ninja: Connected using TLSv1.2 (AES128-SHA)
2025-10-10T18:25:57.210100218Z 2025-10-10 13:25:57,209::INFO::[newswrapper:351] 3@news.newshosting.com: Connected using TLSv1.2 (AES128-SHA)
2025-10-10T18:25:57.289918400Z 2025-10-10 13:25:57,289::INFO::[downloader:616] 1@news.newsgroup.ninja: Initiating connection
2025-10-10T18:25:57.373094445Z 2025-10-10 13:25:57,372::INFO::[downloader:751] Connecting 3@news.newshosting.com finished
2025-10-10T18:25:57.374567972Z 2025-10-10 13:25:57,374::INFO::[downloader:751] Connecting 6@news.newsgroup.ninja finished
2025-10-10T18:25:57.415524099Z 2025-10-10 13:25:57,415::INFO::[newswrapper:351] 1@news.newsgroup.ninja: Connected using TLSv1.2 (AES128-SHA)
2025-10-10T18:25:57.468286251Z 2025-10-10 13:25:57,467::INFO::[downloader:616] 7@news.newshosting.com: Initiating connection
2025-10-10T18:25:57.481704230Z 2025-10-10 13:25:57,481::INFO::[downloader:616] 5@news.newsgroup.ninja: Initiating connection
2025-10-10T18:25:57.557643525Z 2025-10-10 13:25:57,557::INFO::[downloader:751] Connecting 1@news.newsgroup.ninja finished
2025-10-10T18:25:57.587183695Z 2025-10-10 13:25:57,586::INFO::[newswrapper:351] 7@news.newshosting.com: Connected using TLSv1.2 (AES128-SHA)
2025-10-10T18:25:57.607795401Z 2025-10-10 13:25:57,607::INFO::[newswrapper:351] 5@news.newsgroup.ninja: Connected using TLSv1.2 (AES128-SHA)
2025-10-10T18:25:57.681766199Z 2025-10-10 13:25:57,681::INFO::[downloader:751] Connecting 7@news.newshosting.com finished
2025-10-10T18:25:57.697179983Z 2025-10-10 13:25:57,696::INFO::[downloader:751] Connecting 5@news.newsgroup.ninja finished

Once again, I will restate that restarting the container fixes the issue for a few downloads. Then they will start back up again.
Even pausing the current job, then restarting the container (then the job continues automatically) fixes the issue.

On the docker setup: what if you limit SABnzbd to one (1) connection to 1 newserver ... is that stable?

Something between sabnzbd-on-docker and the newsservers is interfering. The host on which docker is running? A firewall? Overloaded router? ... ?

As SAB-on-Windows is running OK for you, it's not your ISP nor newshosting / ninja.

>On the docker setup: what if you limit SABnzbd to one (1) connection to 1 newserver ... is that stable?
No, this does not change anything. Honestly, not a single network change I've made resolves the issue once it has started. The only thing that gets SAB working properly again is a restart of the service/container.

>Something between sabnzbd-on-docker and the newsservers is interfering. The host on which docker is running? A firewall? Overloaded router? ... ?
It seems like something is tripping up the connection, then SAB can't get a grip back on steady connections with the servers. I've noticed a few times now through my testing that my whole internal network takes a slight hit right before the SAB SSL issues begin. So, maybe the firewall is indeed being overloaded... I will try removing the firewall from the equation tomorrow.

I transplanted my linux drive to another computer and the issue is not happening... Which tells me the issue lies with the hardware that my Linux machine was running on. I am guessing either the NIC or CPU, as both are fairly low-grade components. I am going to just keep my server running on the new hardware for a while, and I may get to more testing in the future.

For now, my best guess is an issue between Debian 13 and my computer (HP Elitedesk) hardware.

Thanks for the really great prompts and questions, sander!! May God bless you and your future interactions on this forum.