Timer recording causes often enima2 restart with Apollo 37

[BoxFreak]

Hello,

Apollo 37 seems to unstabile with Venton HD3 box. Often when timer recording starts the Enigma2 restarts, no core files found. Also after the box has been up couple of days there is no possibility to login into it via ssh, could it be some kind of memory leak? With Apollo 35 no problems at all, so had to revert back to it.

[BoxFreak]

Same thing happened now also on 35 Apollo more occasional than in 37 though. It is somehow related to an situation where box should change bouquet/channel on the background before recording starts. Then Enigma2 restarts.

[judge]

An E2 restart when a recording timer kicks in often indicates that your flash is full.
Have you any crash logs in /home/root/logs/?

[Larry-G]

As judge suggests that sounds very much like the flash is full. Check the recording paths and make sure they have not defaulted to the flash.


Sent from my iPad Air using Tapatalk HD

[BoxFreak]

No crash logs found. Recording paths are correct (internal hdd) where also logs and cores should be copied. Memory information tells me that 368,4M is free out of 440,7M total flash. After checking now flashed to version 41. Really odd...

[BoxFreak]

An update on this issue. Now it is clear that the problem is caused by box memory running out and that seems to be caused by oscam (1.20 loaded from OpenViX feeds) leaking into memory. When the box runs out of memory I cant login into it anymore, but can see from GUI that memory is running low and if I the restart oscam memory is freed. Ssh or ftp connections cannot be made to box after the situation anyway until box reboot.

[BoxFreak]

Now tried with Apollo 47 and oscam (9909 build). Seems to be a lot more stabile. But still couple of crashes seen attached crashlogs and logs for your analysis.

[BoxFreak]

Similar crashes as reported above happened also with Apollo build 54... Now on upgraded to build 62. Is this an skin issue?