NGRP problems

Hi,

We have found the same problem two times related with name groups (ngrp).

The problem is that we define one ngrp, named ‘pc’. Then one day, this ngrp stops working and we can only solve the problem by copying this ngrp with the same qualities but of course diferent name.

NGRP: pc → it doesn’t works

${SourceStreamName}_pc

750

750

false

false

500

500

false

false

300

300

false

false

NGRP: aa (copy&paste from ‘pc’) → it works

${SourceStreamName}_aa

750

750

false

false

500

500

false

false

300

300

false

false

This is the response we have by accessing to ‘pc’ NGRP:

curl http://sstream03.dtvc.local:1935/int/ngrp:tv3_pc/playlist.m3u8?DVR

#EXTM3U

#EXT-X-VERSION:3

#EXT-X-STREAM-INF:BANDWIDTH=817152,CODECS=“avc1.77.31,mp4a.40.2”,RESOLUTION=640x360

chunklist_b817152_DVR.m3u8

#EXT-X-STREAM-INF:BANDWIDTH=549152,CODECS=“avc1.77.30,mp4a.40.2”,RESOLUTION=640x360

chunklist_b549152_DVR.m3u8

#EXT-X-STREAM-INF:BANDWIDTH=349152,CODECS=“avc1.77.21,mp4a.40.2”,RESOLUTION=320x180

chunklist_b349152_DVR.m3u8

curl -v http://sstream03.dtvc.local:1935/int/ngrp:tv3_pc/chunklist_b817152_DVR.m3u8

  • About to connect() to sstream03.dtvc.local port 1935 (#0)

  • Trying 10.111.0.73… connected

  • Connected to sstream03.dtvc.local (10.111.0.73) port 1935 (#0)

GET /int/ngrp:tv3_pc/chunklist_b817152_DVR.m3u8 HTTP/1.1

User-Agent: curl/7.19.7 (x86_64-redhat-linux-gnu) libcurl/7.19.7 NSS/3.15.3 zlib/1.2.3 libidn/1.18 libssh2/1.4.2

Host: sstream03.dtvc.local:1935

Accept: /

< HTTP/1.1 404 Not Found

< Accept-Ranges: bytes

< Server: WowzaStreamingEngine/4.0.6

< Content-Length: 0

<

  • Connection #0 to host sstream03.dtvc.local left intact

  • Closing connection #0

Thanks.

Toni

Hi,

When we restarts wowza service, we have detected that if there is any client trying to access to any NGRP defined, this NGRP will not start correctly.

In our system, we have a load balancer that is permanently calling the NGRP which is in fault.

If we stop the load balancer and then we restarts the wowza service, this NGRP works well

Regards,

Toni

Hi,

We have found the same problem two times related with name groups (ngrp).

The problem is that we define one ngrp, named ‘pc’. Then one day, this ngrp stops working and we can only solve the problem by copying this ngrp with the same qualities but of course diferent name.

NGRP: pc → it doesn’t works

${SourceStreamName}_pc

750

750

false

false

500

500

false

false

300

300

false

false

NGRP: aa (copy&paste from ‘pc’) → it works

${SourceStreamName}_aa

750

750

false

false

500

500

false

false

300

300

false

false

This is the response we have by accessing to ‘pc’ NGRP:

curl http://sstream03.dtvc.local:1935/int/ngrp:tv3_pc/playlist.m3u8?DVR

#EXTM3U

#EXT-X-VERSION:3

#EXT-X-STREAM-INF:BANDWIDTH=817152,CODECS=“avc1.77.31,mp4a.40.2”,RESOLUTION=640x360

chunklist_b817152_DVR.m3u8

#EXT-X-STREAM-INF:BANDWIDTH=549152,CODECS=“avc1.77.30,mp4a.40.2”,RESOLUTION=640x360

chunklist_b549152_DVR.m3u8

#EXT-X-STREAM-INF:BANDWIDTH=349152,CODECS=“avc1.77.21,mp4a.40.2”,RESOLUTION=320x180

chunklist_b349152_DVR.m3u8

curl -v http://sstream03.dtvc.local:1935/int/ngrp:tv3_pc/chunklist_b817152_DVR.m3u8

  • About to connect() to sstream03.dtvc.local port 1935 (#0)

  • Trying 10.111.0.73… connected

  • Connected to sstream03.dtvc.local (10.111.0.73) port 1935 (#0)

GET /int/ngrp:tv3_pc/chunklist_b817152_DVR.m3u8 HTTP/1.1

User-Agent: curl/7.19.7 (x86_64-redhat-linux-gnu) libcurl/7.19.7 NSS/3.15.3 zlib/1.2.3 libidn/1.18 libssh2/1.4.2

Host: sstream03.dtvc.local:1935

Accept: /

< HTTP/1.1 404 Not Found

< Accept-Ranges: bytes

< Server: WowzaStreamingEngine/4.0.6

< Content-Length: 0

<

  • Connection #0 to host sstream03.dtvc.local left intact

  • Closing connection #0

Thanks.

Toni

Hi,

We’ll need to look into this issue more closely.

Please send an email to support@wowza.com making a reference to this thread’s URL

and include a ZIP file containing your conf/, logs/, transcoder/ and manager/logs folders from all servers involved.

If you are not sure how to get this information please see this tutorial. Please do make sure you provide logs which show Wowza server starting.

Outline, in the same email, the type of encoder in use along with the encoding settings for the stream. If you are using a .stream file to start restreaming please provide this.

Daren

Hi,

As Daren recommended I think you should open a ticket so that we can look at this in more detail.

If you have a ticket open already please update this thread with the ticket number.

If you have not created a ticket yet, please do so my sending an email to support@wowza.com

Include a detailed description of the problem, a link to this thread for reference and a ZIP file which contains the following directories for analysis:

[Wowza-Install]/conf/

[Wowza-Install]/logs/

[Wowza-Install]/transcoder/

[Wowza-Install]/manager/logs/

Please make sure you provide logs which show Wowza server starting.

If you are not sure how to get this information please see the following tutorial.

How to create a compressed zip file in Windows, OS X, and Linux

  • If you have a live stream please outline the type of encoder in use along with the encoding settings for the stream.

  • If you are using a .stream file to start re-streaming please provide this.

  • If you are using multiple bitrates (ABR) then please provide the SMIL file associated with your configuration.

Thanks

Jason

Hi,

Thanks for the update.

This will now be handled in the created ticket (106365).

Regards,

Jason

I’m having similar issues. Driving me crazy. Once I restart server or create a new one (Amazon), I CANNOT stream using the NGRP m3u8 url that worked 10 minutes before!

It will stream using the specific stream names (eg. stream_360p) but will NOT stream using ngrp:stream_all

Please help me figure this out. Thanks

PS If I use a new stream name ngrp works fine. Just can’t use old stream names before restart or recreate from image

Hi,

This issue is being handled in ticket #135861 which you have opened with the support team.

Make sure to send in your Wowza configuration and log files so we can have a closer look at your setup.

We need a ZIP file (or other archive) which contains the following directories for analysis:

[Wowza-Install]/conf/

[Wowza-Install]/logs/

[Wowza-Install]/transcoder/

[Wowza-Install]/manager/logs/

Please make sure you provide logs which show Wowza server starting.

Zoran

Hi again,

Let me add some more information to this issue that’s driving us crazy. It seems that

  • We start the Wowza service having the encoder stopped

  • The load balancer is checking the URL of one of the chunklists, receiving a 404 from wowza. You can emulate the load balancer behaviour by doing

while true; do curl http://wowza:1935/int/ngrp:theStream/chunklist_b349152_DVR.m3u8; sleep 1; done

  • Start the encoder

  • Requests to that chunklist begin to take around 8-9 sec to respond, with the same 404

  • Logs are showing the following warning

WARN server comment 2014-10-16 00:19:25 - - - - - 1129.555 - - - - - - - - HTTPStreamerAdapterCupertinoStreamer.service: Request timeout: 8000

  • Shutting the load balancer process down for a while (say 5 minutes) makes the stream recover.

Any idea?

Thanks

Thanks,

We already have opened a ticket, so we will update it.

Toni