2016-08-25, 09:35 PM
Here is the log for the last 15 minutes, it hasn't any messages since soon after I started streaming though. It's still playing now at 9:32, last log messages where 9:19.
2016-08-25, 09:35 PM
Here is the log for the last 15 minutes, it hasn't any messages since soon after I started streaming though. It's still playing now at 9:32, last log messages where 9:19.
2016-08-25, 09:35 PM
CPU usage is sitting around 22%
2016-08-25, 09:43 PM
That's odd probably debug logging is needed I just never noticed. If you add <loglevel>10</loglevel> to advancedsettings.xml you get debug logging without the verhead of on screen debug.
I see that you have use 360 for me that does keep the CPU around 40-50% Martin
2016-08-25, 10:15 PM
I didn't really pick 360p for any reason. I was just trying out your settings screen, and must of left it on that.
I'm on 720p now, and it doesn't really seem to have made a lot of difference to the CPU.
2016-08-25, 10:25 PM
My backend just isnt up to HD transcoding and play. it is as old as Vista and just a Core2Duo so I am not surprised.
Thanks for checking. From your logs I am not seeing the timeout on a lease but I do see the the .status problem where it takes more than a second to return a GET so I abort. For me like you 75% is the first one I don't fail on Code: 9:58:05 T:12816 DEBUG: http://127.0.0.1:8866/services/services?method=channel.transcode.initiate&profile=720p-1024kbps&channel=1&sid=3f461180c7ec47b2a2b8421f42436625 Martin
2016-08-25, 10:38 PM
Yeah, I've got a pretty decent CPU in this machine. The CPU is a 4 year old i7. It was a lower end i7, but all the i7's are pretty damn fast.
2016-08-26, 07:14 PM
(This post was last modified: 2016-08-26, 07:20 PM by mvallevand.)
I think I have found the problem with .status I see that the browser fails often too before streaming starts. I removed a lot of waits and I think the call is failing until shortly after ffmpeg logs this
2016-08-26 15:03:30.303 [DEBUG][18] ErrorData: Output #0, hls, to 'C:\Users\Public\NPVR\web\temp\CJOH-3d7c93c43ae34c979abffd12e5466f74.m3u8': I think the file open for the buffer status may be returning an unhandled error and so there is no payload until it has data. Martin
2016-08-26, 07:43 PM
Yeah, I'd thought it'd be around this. I'll try find a fix today.
|
|