Sabnzbd hanging?? or something...

[neogrinch]

in the past month or so, sonarr and radarr are on occasion not able to reach sabnzbd. And, If I go to the URL on my lan, it can't be found. Though it is running in task bar. I can then right click the icon in task bar and shutdown the server, then restart it, and everything is fine again. I am thinking that it may be doing this when it is checking RSS feeds. Not every time, however, but this is happening to me 2-3 times a week, I'd guesstimate.

When I downloaded the log this is all I see:

2022-05-03 20:55:58,649::INFO::[rss:473] Starting scheduled RSS read-out for "PC World"
2022-05-03 20:56:16,347::INFO::[rss:473] Starting scheduled RSS read-out for "OK Magazine USA"
2022-05-03 20:56:34,845::INFO::[rss:473] Starting scheduled RSS read-out for "National Geo"
2022-05-03 20:56:54,676::INFO::[rss:473] Starting scheduled RSS read-out for "popular mechanics"
2022-05-03 20:57:13,083::INFO::[rss:489] Finished scheduled RSS read-outs
2022-05-04 11:18:48,817::INFO::[__init__:450] [N/A] Performing SABnzbd shutdown
2022-05-04 11:18:48,837::INFO::[__init__:357] SABnzbd shutting down...
2022-05-04 11:18:49,002::INFO::[ssdp:102] Stopping SSDP
2022-05-04 11:18:49,030::INFO::[directunpacker:542] Aborting all DirectUnpackers
2022-05-04 11:18:49,079::INFO::[notifier:123] Sending notification: SABnzbd - Shutting down (type=startup, job_cat=None)
2022-05-04 11:18:49,096::INFO::[downloader:691] Shutting down
2022-05-04 11:18:49,130::INFO::[decoder:128] Shutting down decoder Thread-3
2022-05-04 11:18:49,138::INFO::[decoder:128] Shutting down decoder Thread-4
2022-05-04 11:18:49,209::INFO::[decoder:128] Shutting down decoder Thread-2
2022-05-04 11:18:49,215::INFO::[assembler:69] Shutting down
2022-05-04 11:18:49,224::INFO::[nzbqueue:237] Saving queue
2022-05-04 11:18:49,418::INFO::[postproc:128] Saving postproc queue
2022-05-04 11:18:49,512::INFO::[__init__:436] All processes stopped


So basically the last activity I see before it becomes unresponsive is the RSS read-outs. Then you can see the next day where I shut it down after realizing it was hanging. Any ideas? I am using version 3.5.3 [bca9f3b]

[safihre]

And if you try to access it locally through localhost?

[neogrinch]

If I right click the Sabnzbd icon in the windows task bar, select "show interface" the page cannot be contacted (only when the issue is happening). I've been using sabnzbd for years, never seen anything like this happen before. I typically type in the local IP address in the address bar, ie: http://192.168.0.10:8080/sabnzbd/

When you say localhost, do you mean going to http://localhost:8080? I have never accessed it that way, however trying to right now, page cannot be found, though I can get to it fine the way I usually do, by internal IP. I can get to Radarr using http://localhost:7878, can do the same for sonarr and others.

Under Sabnzbd settings I have 192.168.0.10 as the Sabnzbd Host, and I have this IP address reserved for this machine on the router. Like I said, once it starts up it works fine, and can go for maybe a day or two, then this issue will happen and I have to shut down and restart Sab. The last couple of times it has happened, it seems like it might be related to RSS feed checking, however that might just be a coincidence.

If I need to, I can disable the RSS feeds for a few days and see if it happens again.

[safihre]

You could disable RSS feeds just to test.
I would also suggest to try to change the Host setting to 0.0.0.0 instead. Just for testing.

[neogrinch]

I changed host to 0.0.0.0 as suggested, going to try that first. I see that I can get to http://localhost:8080 now as well as the internal IP. If it hangs again, I will then disable RSS to test that.

[neogrinch]

it did it again today. I restarted machine on Friday, so it was up for two days before it hang. I have now disabled all of the RSS and see if it will stay up the next few days.

[neogrinch]

It did it again this morning, where I couldn't load the interface, though it was running in task bar. I had the RSS feeds disabled this time. But when I tried to load up localhost, or right click the icon and "show interface" the page couldn't be found. Likewise, my other programs such as Radarr cannot contact the sabnzbd instance either, "All download clients are unavailable due to failures". I believe this started happening shortly after the last time I updated Sabnzbd. Might it be a bug of some sort, or likely something else? Anyway, I restart the server, and everything is groovy for a day or two. Thanks.

[sander]

Can you tell about your system on which SAB is running?

Which OS?
What interfaces?
VPN?
Firewall?

PS: can you set SAB's logging to +DEBUG?

[neogrinch]

I have sab installed on a Windows 10 Desktop computer. I've been using Sabnzbd w/out issue on this particular machine since 2017. It is allowed through windows firewall. Other programs that ulitize sabnzbd like radarr, sonarr, etc are also installed on the same machine. I do not use VPN on this machine. the machine is directly connected by ethernet to my Netgear AX1800 6700ax router.

I believe that I last updated Sabnzb around the beginning of April, a week or two after I noticed the update was available. In reflection, it was around that time that the issue started occurring.

I turned on debugging. Once it happens again should I paste in the relevant portion of the log here with the debugging log info, or do you need anything from that now?

Below is the normal log during the time that the issue seemed to happen. There is nothing there really. An item was queued and downloaded at 4:55 am, then cleaned up and finished. and when i went to look at sab this morning I could not reach it. Radarr/Sonarr also reported it could not communicate with download client. So I shut down and restarted Sabnzbd. Everything works fine again.

Code: Select all

2022-05-10 04:55:12,887::INFO::[postproc:128] Saving postproc queue
2022-05-10 04:55:12,889::INFO::[downloader:409] Post-processing finished, resuming download
2022-05-10 04:55:16,895::INFO::[nzbqueue:237] Saving queue
2022-05-10 04:55:16,907::INFO::[postproc:128] Saving postproc queue
2022-05-10 04:55:16,916::INFO::[notifier:123] Sending notification: SABnzbd - Queue finished (type=queue_done, job_cat=None)
2022-05-10 08:48:17,825::INFO::[panic:239] Launching browser with http://127.0.0.1:8080/sabnzbd
2022-05-10 08:49:53,627::INFO::[panic:239] Launching browser with http://127.0.0.1:8080/sabnzbd
2022-05-10 08:50:07,594::INFO::[__init__:450] [N/A] Performing SABnzbd shutdown
2022-05-10 08:50:07,594::INFO::[__init__:357] SABnzbd shutting down...
2022-05-10 08:50:07,801::INFO::[ssdp:102] Stopping SSDP
2022-05-10 08:50:07,802::INFO::[directunpacker:542] Aborting all DirectUnpackers
2022-05-10 08:50:07,847::INFO::[notifier:123] Sending notification: SABnzbd - Shutting down (type=startup, job_cat=None)
2022-05-10 08:50:07,848::INFO::[downloader:691] Shutting down
2022-05-10 08:50:07,872::INFO::[decoder:128] Shutting down decoder Thread-3
2022-05-10 08:50:07,887::INFO::[decoder:128] Shutting down decoder Thread-4
2022-05-10 08:50:07,887::INFO::[decoder:128] Shutting down decoder Thread-2
2022-05-10 08:50:07,888::INFO::[assembler:69] Shutting down
2022-05-10 08:50:07,889::INFO::[nzbqueue:237] Saving queue
2022-05-10 08:50:07,901::INFO::[postproc:128] Saving postproc queue
2022-05-10 08:50:07,911::INFO::[__init__:436] All processes stopped
2022-05-10 08:50:07,927::INFO::[_cplogging:213] [10/May/2022:08:50:07] ENGINE Bus STOPPING
2022-05-10 08:50:08,114::INFO::[_cplogging:213] [10/May/2022:08:50:08] ENGINE HTTP Server cherrypy._cpwsgi_server.CPWSGIServer(('0.0.0.0', 8080)) shut down
2022-05-10 08:50:08,115::INFO::[_cplogging:213] [10/May/2022:08:50:08] ENGINE Bus STOPPED
2022-05-10 08:50:08,115::INFO::[_cplogging:213] [10/May/2022:08:50:08] ENGINE Bus EXITING
2022-05-10 08:50:08,116::INFO::[_cplogging:213] [10/May/2022:08:50:08] ENGINE Bus EXITED
2022-05-10 08:50:08,116::INFO::[notifier:123] Sending notification: SABnzbd - SABnzbd shutdown finished (type=startup, job_cat=None)
2022-05-10 08:50:08,116::INFO::[SABnzbd:1607] Leaving SABnzbd

[sander]

So no special stuff with your OS or network. Good.

Yes, hopefully / maybe with DEBUG we see more.

My hypothesis: SAB is still running, but you can't access the webinterface anymore. Ah, a first check for that is: set "RSS Checking Interval" (via Config -> RSS) to 15 minutes. Then your logging should show a RSS message each 15 minutes. If so, SAB is still running. If not, SAB is not running ...

Code: Select all

2022-05-10 17:59:35,694::INFO::[rss:489] Finished scheduled RSS read-outs
2022-05-10 18:15:06,879::INFO::[rss:489] Finished scheduled RSS read-outs
2022-05-10 18:30:38,030::INFO::[rss:489] Finished scheduled RSS read-outs
2022-05-10 18:46:09,225::INFO::[rss:489] Finished scheduled RSS read-outs
2022-05-10 19:01:40,397::INFO::[rss:489] Finished scheduled RSS read-outs

or (more lines):

Code: Select all

2022-05-10 17:59:04,517::INFO::[rss:473] Starting scheduled RSS read-out for "nzb.su cart"
2022-05-10 17:59:20,391::INFO::[rss:473] Starting scheduled RSS read-out for "Drunkenslug cart (???)"
2022-05-10 17:59:35,694::INFO::[rss:489] Finished scheduled RSS read-outs
2022-05-10 18:14:35,696::INFO::[rss:473] Starting scheduled RSS read-out for "nzb.su cart"
2022-05-10 18:14:51,576::INFO::[rss:473] Starting scheduled RSS read-out for "Drunkenslug cart (???)"
2022-05-10 18:15:06,879::INFO::[rss:489] Finished scheduled RSS read-outs
2022-05-10 18:30:06,881::INFO::[rss:473] Starting scheduled RSS read-out for "nzb.su cart"
2022-05-10 18:30:22,735::INFO::[rss:473] Starting scheduled RSS read-out for "Drunkenslug cart (???)"
2022-05-10 18:30:38,030::INFO::[rss:489] Finished scheduled RSS read-outs
2022-05-10 18:45:38,032::INFO::[rss:473] Starting scheduled RSS read-out for "nzb.su cart"
2022-05-10 18:45:53,900::INFO::[rss:473] Starting scheduled RSS read-out for "Drunkenslug cart (???)"
2022-05-10 18:46:09,225::INFO::[rss:489] Finished scheduled RSS read-outs
2022-05-10 19:01:09,230::INFO::[rss:473] Starting scheduled RSS read-out for "nzb.su cart"
2022-05-10 19:01:25,104::INFO::[rss:473] Starting scheduled RSS read-out for "Drunkenslug cart (???)"
2022-05-10 19:01:40,397::INFO::[rss:489] Finished scheduled RSS read-outs

[neogrinch]

Ok. I will report back.

Yes, before I disabled the RSS feeds, it would show the RSS every time it checked, and at some point, would just stop. i will turn it on again at 15 minute interval so that can be seen via log.

[sander]

Yes, before I disabled the RSS feeds, it would show the RSS every time it checked, and at some point, would just stop.

... hmm. That is a signal that SAB is more hanging than 'just' the SAB interface. Hopefully DEBUG will tell more.

[neogrinch]

So, i believe I might have figured out what was happening. I am now on day 3 of it being up without a hang.

I had the sabnzbd port exposed on the router. It has been that way for a long time. I had never had any issue with it. in addition, I have Netgear Armor (powered by Bitdefender) on my router.

looking at my Bitdefender logs, I have had several incidents where an attack was blocked. These occurred around the time when Sabnzbd would hang up:

"A directory traversal exploit attempt was made by mypublicip:8080. We blocked the connection to prevent it from gaining unauthorized access and altering critical system data."
"An attempt to inject a command towards your system through a dangerous URL was made by 45.9.20.101. We blocked the connection to prevent malicious commands from being executed on the device and across the network."

These two above happened one right after another about the time it started hanging last time. Sabnzbd is on port 8080.

Not sure why this only just started happening recently for the past month, but never prior, maybe automated attacks of some sort. Anyway, I stop exposing sabnzbd on the router, and will just access it internally or via VPN from here on out. much safer that way anyway -- was probably unwise to have the ports open on the router to begin with.

It has only been three days, but I feel like this was the most likely cause. I haven't had any hangs yet, and no more warnings from BitDefender. But if it happens again I will update.