Find content
Not Telling
Display name historyHi guys!
From a long time (few months) i see some bad behaviour of my main box. It's dead until i do hard restart (pull a plug).
If the problem occurs i see:
- no response to remote control
- no response to touching the power button
- no ping
- no ssh
- no open port 80 (webif)
- no open port 8888 (oscam)
- no video on HDMI (only port that i connect to TV)
Now, what i know so far:
1. Problem occurs from time to time. No more than once per day. There can be 20 days without problem, and there can be 5 days with problem everyday.
2. I see some odd behaviour with streaming status - without daily restarts, it was possible to see that one client box is streaming 40 times at the same time. 39 were without any particular program set, one was with a program chosen. So 39 of them were some kind of dead streams. Right now it's usual i have 2-3 from the same client and 1-2 of them are "dead".
In webif i see "Streaming Info: S192 (unknown service)"
3. The max of streaming clients is 6. Avarage 2-3.
4. The max of recording and streaming is 6.
5. Problem occured during my vacation so main box was only streaming and recording (no one used it on TV).
6. I have a zabbix collecting some data on my main box. Mainly ping, rec and streaming number, status, webif status, ports status.
7. Load is mainly 0.05-0.01. So it's not much 
8. Mem used is usualy around 200M/708M.
9. swap is turned off.
My setup is:
Vu+ Uno4K SE
OpenPLi 8.1-release (2022-01-20-release-8.1)
All 8 tuners connected to one converter via unicable (only one socket is plugged)
SSD disk connected for recording and logs purpose.
Few SMB mounting points (for playing some of my private videos and photos).
There are 5-6 different clients:
- zgemma h9s
- zgemma h7s
- Vu+ zero
They are all connected through wireless bridges and routers (almost all are behind NATs).
All boxes are in state of no sat dish connected and all have configured a reserve remote box to connect to.
All clients are openpli 8.1.
What i already did:
1. i have a script on main box that checks if the oscam-emu is up. In case it's not - it runs it (/usr/bin/oscam-emu -
2. Regular reboot every 24h.
3. I'm working on a script that will do curl localhost - in case of 5x problem, it will dump some info and reboot the box.
4. I have a smart 220V socker before the box - in case of problems, i turn the power off for few seconds and back on. It's my way for hard restart when i'm not around.
5. I changed the image version (8.0 to 8.1). I see there is 8.2 - i will upgrade next week probably.
There is no crash dump during the "problem". Lately i discovered a crash dump during the reboot (cron does that at 6:00:00 and crash was 6:00:02). I attached it to the post.
I have no idea what to look for. Can someone point me the right way to diagnose the problem?
