Film-Tech Forum ARCHIVE: Dolby DSS220 There are picture pauses during multiple screenings

This is topic Dolby DSS220 There are picture pauses during multiple screenings in forum Digital Cinema Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=16;t=003362

Posted by Tacky Ting (Member # 8686) on 06-30-2018, 01:58 AM:

There are picture pauses during multiple screenings
DSS220 Version is 4.9.0.96
Barco DP-4K 23B Version 2.5.57
2018-06-30 09:45:41,122 ERROR ImbProxy.UslVideoStream - Session TID (0x74effb90) Q, PIX Callback Performance Test, Measured system latency (53704) ms
Line 241384: 2018-06-30 09:51:06,399 INFO [USLControlPoint] tiprojector.TIProjectorStateMachine - Received time update greater than 10 seconds from last update. Status code = IN_PROGRESS
Hard Disk SMART is healthy
Have you ever had this problem?

Posted by Steve Guttag (Member # 268) on 06-30-2018, 06:35 AM:

Nope...but when I hear of freeze frames I look for buffer underflows. Underflows are generally when the server (RAID system) can't keep up with the demands of the mediablock. When the mediablock runs out of bits, the image freezes.

Look closer at the RAID logs. Do ANY of the drives have reallocated sectors? Check for ATA errors (note, both of these are shown in the Dolby Log Analyzer. Another thing to look at are the read speeds in the SMART logs to see if one drive is dramatically different than the rest.

9 times out of 10 (probably greater)...when show freezes occur, there is a drive or drives responsible.

Posted by Tacky Ting (Member # 8686) on 06-30-2018, 08:28 AM:

http://loganalyzer.dolbycustomer.com/index.php/reports/analyze/97606/bd1b3bcff9f93b08abf634bf40b918a4
Raid & Hard disk is healthy

Posted by Stephan Shelley (Member # 8587) on 06-30-2018, 01:26 PM:

You defiantly have video underflows. I also see media block disconnects but not at the time the video underflows.

When the playback pauses look at the disk activity lights. Is one staying on for an unusually long time? If so that is your problem. You might want to replace the media block data cable.

As you are using cat 745 have you exchanged it for a battery upgrade? The original coin batteries die after about 3 years.

Posted by Alexandre Pereira (Member # 9295) on 06-30-2018, 01:29 PM:

I had exactly the same thing happen with my DSS200 - in my case two drives were starting to fail or at least had too many bad sectors causing pauses in the show. The drives were replaced and the raid rebuilt. It can even be done without shutting down the server if you know which drive is bad.
Try running a show from the CRU - if it runs fine it is an issue with the raid drives.

Posted by Steve Guttag (Member # 268) on 06-30-2018, 02:18 PM:

The DSS220, with the software RAID, doesn't give as much information as the 3WARE hardware RAID. Yes, no ATA errors or Reallocated sectors are showing, at this time. I did see some of the Temperatures of the drives seeming getting hot (over 40C).

The buffer underflows are reported. When you have buffer underflows, 9 times out of 10, it is the hard drives/RAID system that is the cause of it. The other 10% is the Ethernet cable/connectors that link the server to the IMB.

The Cable is the cheap thing to try and cleaning the server plus having better cooling would be good for the drives. Failing that, I'd start a plan to change out the Drives...either 1 at a time to retain the content or wholesale BUT you MUST have a properly formatted/loaded Install THUMB DRIVE before doing a complete RAID change. The advantage of running the "install" is that it will also wipe anything bad that has crept into the system over the years.

Posted by Tacky Ting (Member # 8686) on 06-30-2018, 09:09 PM:

40C is not at work late at night. Normal work it is at 34C
I can reinstall System, but can't find the reason. It has an opportunity to happen.
IMB Cable & battery Has been replaced

What does this log mean? When is it normal?
USLControlPoint] tiprojector.TIProjectorStateMachine - Received time update greater than 10 seconds from last update. Status code = IN_PROGRESS