Hi Maxim, you are right, to add this would help, if it is a common problem.
This is a really old file. It seems that it was already a pdf-file. Original content of the folder was HASH.pdf, HASH.swf, library.xml and for each page of the pdf a _thumb_page-XXXX.jpg. Greetings Peter Am 31.05.19 um 18:39 schrieb Maxim Solodovnik: > Hello Peter, > > I guess you propose to add "profile" param to OM code :) > I'm OK with it, could you check if your original document is being > converted by OM UI as expected or not? > > On Fri, 31 May 2019 at 20:03, Peter Dähn <[email protected]> wrote: >> Hi Maxim, >> >> I needed to convert some room files afterwards. I user command out of >> the logs and got for some files an error. >> >> /usr/bin/convert -density 125 >> /SERVER/webapps/openmeetings/data/upload/files/HASH/HASH.pdf -quality >> 100 /SERVER/webapps/openmeetings/data/upload/files/HASH/page-%04d.png >> >> convert: profile 'icc': 'RGB ': RGB color space not permitted on >> grayscale PNG >> `/SERVER/webapps/openmeetings/data/upload/files/HASH/page-%04d.png' @ >> warning/png.c/MagickPNGWarningHandler/1654. >> >> I needed to add parameter +profile '*' . Now convert is working without >> error. >> >> /usr/bin/convert -density 125 >> /SERVER/webapps/openmeetings/data/upload/files/HASH/HASH.pdf +profile >> '*' -quality 100 >> /usr/lib/red5/webapps/openmeetings/data/upload/files/HASH/page-%04d.png >> >> Greetings Peter >> >> Am 15.05.19 um 13:15 schrieb Peter Dähn: >>> Hi, >>> >>> I had a little problem during the last update (from 3.0.7 to 4.0.8). I >>> didn't realise, that ffmpeg didn't work after server update. The >>> following om update went throug, despite the fact, that >>> screen-recordings were not converted. >>> >>> That why I needed to figure out (with a little bit help of Maxim) how to >>> manage that afterwards. I would like to share my steps, in case someone >>> has a similar problem. >>> >>> 1. Get needed information from database (in my case it is postgres) >>> >>> select id,hash from om_file where deleted=false and type='Recording' >>> and dtype='Recording' order by id; >>> >>> 2. save the result in a text file e.g. convert_recordings.txt >>> >>> 3. prepare a little script (linux/bash) >>> >>> #!/bin/bash >>> i=0 >>> while read line >>> do >>> #echo "$line" >>> id=`echo $line | awk '{split($0, a, "|"); print a[1]}'` >>> #echo $id >>> hash=`echo $line | awk '{split($0, a, "|"); print a[2]}'` >>> #echo $hash >>> # echo "bearbeite $id" >>> command_mp4="/usr/local/bin/ffmpeg -y -nostdin -i >>> /OM_OLD_DIR/webapps/openmeetings/data/streams/hibernate/flvRecording_$id.avi >>> -c:v h264 -crf 24 -pix_fmt yuv420p -preset medium -profile:v baseline >>> -level 3.0 -movflags faststart -c:a aac -ar 22050 -b:a 32k >>> /OM_DIR/webapps/openmeetings/data/streams/hibernate/$hash.mp4" >>> # echo $command_mp4 >>> command_png="/usr/local/bin/ffmpeg -y -nostdin -i >>> /OM_DIR/webapps/openmeetings/data/streams/hibernate/$hash.mp4 -vf >>> thumbnail,scale=640:-1 -frames:v 1 >>> /OM_DIR/webapps/openmeetings/data/streams/hibernate/$hash.png" >>> # echo $command_png >>> $command_mp4 >>> $command_png >>> done < convert_recordings.txt >>> >>> 4. wait a looooong time (I had over 500 videos) >>> >>> >>> Maybe ffmpeg command is a bit different for other versions. >>> >>> Greetings Peter >>> >>> >>> >>> >>> >>> >> >
