NZB's stuck in FETCH status

[mainframesysprog]

Running SABnzbd 4.0.3. My RSS feeds from NZBGeek have been working fine until I made some changes. I found a post on Reddit about using switches to delete unwanted files from downloads. I added those extensions to my switches. I'm 95% certain that is all I changed. Now, SABnzbd reads my cart but all of the NZB entries in my queue are stuck in fetch status. I've looked in the log and also turned on debug but don't see anything meaningful

2023-09-03 10:56:46,910::INFO::[notifier:123] Sending notification: NZB added to queue <SNIP> Trying to fetch NZB from URL /api?t=get&id=<SNIP>&del=1 (type=download, job_cat=None)

I went back into switches and restored defaults but I'm still in the same state.

If I manually download the NZB and add it to the queue, it downloads just fine.

I would welcome any ideas or suggestions.

[safihre]

There's nothing before /api? Seems like broken url

[mainframesysprog]

New users are not allow to post links so I had to modify the post and remove the link. It's there in my log. I just changed it to URL. But it is a H T T P S to API dot NZBGEEK dot INFO

[safihre]

If you want those back, I think you best do a queue repair.
Otherwise, can you enable Debug logging in the Status window and then after about 5 min click Show Logging and send the logs to me at safihre@sabnzbd.org
Just so I can debug what's going on. It should fail jobs after 10 tries.

[mainframesysprog]

Log emailed.

[mainframesysprog]

Were you able to find anything in the log?

[safihre]

I'm currently on vacation until early next week.
I did have a quick peek at the logs, but could you pause the queue and then wait 5 min and click Show Logging?
It's a bit hard to read because of all the downloading going on.

[mainframesysprog]

Sorry, I didn't realize you were on vacation. This can wait until you are back and available.

Is there an option to clear the log and start fresh? I can easily recreate the issue without all the noise in there. Pausing didn't do much other than show pausing and then list the parms:

2023-09-08 11:41:38,900::INFO::[downloader:442] Pausing
2023-09-08 11:41:38,901::INFO::[notifier:123] Sending notification: SABnzbd - Paused (type=pause_resume, job_cat=None)
2023-09-08 11:41:38,901::INFO::[downloader:462] Forcing disconnect
__version__ = 19
__encoding__ = utf-8
[misc]
pre_script = None
...

[mainframesysprog]

I decided to start a new log file. I shutdown SABnzbd, renamed the current log to .OLD and restarted SABnzbd. Everything started working again. I'm not sure why. Thanks for your assistance and sorry to have bothered you when a restart seems to have fixed it.

[safihre]

Great

[mainframesysprog]

Well, I spoke to soon. Items from my RSS feed are now getting stuck in FETCH status again. I've even gone as far as reinstalling SAB and renaming the \appdata\SABNZBD folder to start with fresh settings. I am getting a little bit more information in the logs now. I'm seeing this:

2023-09-10 09:59:58,738::INFO::[urlgrabber:120] Grabbing URL https://nzbgeek.info/geekseek.php?guid=<HASH>dfa9c1d
2023-09-10 10:00:12,493::INFO::[urlgrabber:269] Unknown filetype when fetching N<USERNAME>B, retry after 30s https://nzbgeek.info/geekseek.php?guid=<HASH>dfa9c1d

[sander]

So SAB gets something from nzbgeek of which the extension is not a valid archive nor a valiz nzb.
See https://github.com/sabnzbd/sabnzbd/blob ... er.py#L245

Easy check: that URL https://nzbgeek.info/geekseek.php?guid=<HASH>dfa9c1d ... if you open that with a web browser (Chrome, Firefox, etc) ... what do you get?

[sander]

Update:

I created an account at nzbgeek, and the URL of the cart RSS is of this format:

https://api.nzbgeek.info/rss?t=-2&limit ... =<geek_key>

So I wonder what you've put into SAB's RSS settings?

FWIW: that "https://nzbgeek.info/geekseek.php?guid" on which your SAB reports "Unknown fileytpe" ... I don't think that can ever lead to a NZB or zipped NZB

[mainframesysprog]

This is my RSS feed setting (key removed after R=)

https://api.nzbgeek.info/rss?t=-2&limit ... 1&del=1&r=

When the fetch times out, the NZB link shows in in SAB. I can cut, paste and navigate to the link successfully in the browser.

[sander]

OK, that's good.

Then again: how it is possible that SABnzbd is trying to get anything at all from ...geekseek... ?
EDIT
in my nzbgeek cart RSS, there are URLs starting with https://nzbgeek.info/geekseek.php?guid (like below), but I have no idea why SAB would try to fetch stuff from that URL. Google hit for "guid isPermalink": "a url that can be opened in a Web browser, that points to the full item described by the <item> element." ... so web browser stuff, nothing usable for SAB

< guid isPermaLink="true">https://nzbgeek.info/geekseek.php?guid= ... aaaaaaaaaa< /guid >

/EDIT


From my logging:

Code: Select all

sander@zwart2204:~$ cat .sabnzbd/logs/sabnzbd.log | grep -i nzbgeek.info |  wc -l
101

... and all of them api.nzbgeek.info:

Code: Select all

sander@zwart2204:~$ cat .sabnzbd/logs/sabnzbd.log | grep -i nzbgeek.info | grep -i api.nzbgeek.info |  wc -l
101

and none with "geekseek" in them:

Code: Select all

sander@zwart2204:~$ cat .sabnzbd/logs/sabnzbd.log | grep -i nzbgeek.info | grep -i geekseek |  wc -l
0

What does your logging show?