[Logo] Jaikoz and SongKong Forums
  [Search] Search   [Recent Topics] Recent Topics   [Members]  Member Listing   [Groups] Back to home page 
[Register] Register / 
[Login] Login 
Messages posted by: dbaps  XML
Profile for dbaps -> Messages posted by dbaps [39] Go to Page: 1, 2 Next 
Author Message
Paul,

Do you have any documentation on what fields or columns are supported in Discogs versus Musicbrainz? It would be interesting to have a list of Discog fields or columns that are unique and not in Musicbrainz.

Thanks!
First I owe you a huge apology. I guess it happens when I'm doing five things at once. This error was in Jaikoz and not in Songkong . I will send you the log so you can take a proper look. Please either move this to Jaikoz or delete it. I believe the error occurred when running, "autocorrect data from Musicbrainz."

Thank you for clarifying how duplicates work. It's good to know that you throw the song out if one of the fields to identify duplicates is blank.
I'm getting the following error for a large percentage of my songs:

WARNING: Unable to update song 244 with Id ff1d229b-95c9-4ac9-9028-bff35dc4204b because the song has no Release Id

Of course the sequential number (244) and the ID change with each warning.

I'm also concerned about having a blank Release ID when it is one of the parameters I use to find duplicates.

I saw the java error with Discogs and Release ID. Is this something different?

Thanks...
My guess is that you can tell if a song has been matched before? If a song has been matched before and I don't want it processed again, can there be an option to move the song to the matched folder? This way it doesn't split my matched files into two separate groups.
I still code the info.plist value:

<string>-Xmx2000m</string>

I usually set it at 8000.

This is what shows up in activity monitor:

Real Memory - 3.5GB
Virtual Memory Size - 13GB
Shared Memory - 28MB
Private Memory - 3.25GB

That's about 20GB on a 32GB Mac, OSX 10.9 Mountaineer

I remember one time you told me it doesn't make as big a difference manually adjusting the memory in OSX like it does in Windows.

I don't know a lot about memory allocation on a Mac. I have received warnings that my Startup disc was full due to the Virtual Memory allocation. I'm guessing it must be like a pagefile in Windows? So having an SSD for a startup OSX disc does make a difference with Songkong when storing the input file there.That I know from personal experience.

Do you think implementing a percentage heap file value in OSX, similar to what was done in Windows would be helpful? Also would increasing the real memory and/or private memory improve performance and reduce the virtual memory size requirement? If it's like a RAMDISK put into memory, probably would not make a difference, I'm guessing. if it's paging to disk then it would probably help?

Finally with Java 8, are there any new tuning parameters coming that may help Songkong performance?

Apologize for all of the questions but they are somewhat related

As always, thank you for all of your hard work.

Paul
Hi Paul,

I may not be up to speed on enhancements. I saw one thread about a possible future enhancement to Songkong to lift the throughput limitation for those of us running our own Musicbrainz Server. Will this change be made soon? I don't know the difficulties on the programming side.

That was an interesting comment about how a Musicbrainz Server will not really improve performance for Songkong due to this limitation. What is it, one song per second, or something like that?

Thanks!
Paul
Not working for me. See attachments.

1). When you click on the header of your search results, it sorts the data. For example, "last message" would be sorted descending, click a second time and it sorts ascending. views, same thing, etc.

2). Create an "advanced search" like many forums have where you can do things like search only on the subject line and not the body. Just one example.

My guess is this involves an upgrade of the forum software you use?

Thank you for your consideration,
Paul S.
Paul,

Maybe I've missed it but what user id and password are used for the optional local database and search server for Songkong?

As you know, in Jaikoz a user can specify the user and password.

Thanks for your help,
Paul S.
Paul,

Maybe a silly question...When processing, what does the word "completed" actually mean?

The reason I'm asking is for this latest run I'm only saving at a rate of about 10% of completed. Does completed include songs that it finds in the database on my computer and does not need to go out and retrieve metadata from Musicbrainz? Does it not save a song under certain non-error conditions? I'm looking for any logical reason besides errors why this ratio is so low.

Thank you!
Sharing a database...An interesting idea though

Just turn off the write locks and watch the corruption reign in! That or deadlocks and database write waits...with timeouts

Thank you very much. I will let you know how it goes...

Paul
Make the report optional. When I get an error sometimes I don't want to wait for the report. Just ask at the end, do you want the report.

Lots of corruption database errors. I have multiple machines, perhaps that has something to do with it. I don't know. I know this is open-ended and without my files you can't really help. I will start send them to you.

Put a bandwidth meter in the software so that if I use multiple machines, the grand total stays below the through-put limit rather than blowing up.

Extreme slow processing- very inconsistent on both Windows and Macs.

Even though this software is trying to make things as simple as possible, is there a way to minimize the time that the bandwidth restriction is active? For example, can we use our own MB database and search server? Can any of the processing that is tied up in the bandwidth limitation routine be pulled out and put somewhere else in the processing because it does not apply?

I started watching what this software is actually doing on the network. Sometimes it's accessing or trying to access over 100 ip addresses according to "Little Snitch."

What am I losing with a corrupt database? What I mean by that is the following:

Is processing more efficient when there is data in the database?
Is the quality of the data better because it has part of my collection in this database?

I am going back to just using one machine. I love the software. I hate the errors (lol). I know you told me once to run Songkong first, then run Jaikoz afterward.

I apologize if this is showing my level of frustration. I probably should write this later but will forget what I want to say.

Thanks for all that you do!
Paul,

You made a comment somewhere that when you finished the last update to Songkong, you were now going to work on Jaikoz so that both products are equal in efficiency (efficiency may not be the correct word for it). Anyway is this now where the two products are in their releases?

By the way, so far all of my earlier problems seem to be fixed.

Thanks!
My image folder is up to 27,000 files. My log file is also growing rapidly. I'm assuming if I have a corrupt database and wipe out that folder on Windows, that I should do the same with the logs folder.

Does the growth of either of these folders slow down processing on a Mac or Windows? Is there any scenario where the logs and images folders' contents should be deleted?

Same question for the size of the database. Is there a point where it should be deleted? On my Mac it's at 5GB.

Thanks again.
I got a message today with 1.4 on Windows. There was a pop-up saying that I was already accessing Musicbrainz from this IP address and therefore I was blocked.

Nw I know you said that was not allowed with Songkong. I'm just pointing out it's the first time I've ever seen a pop-up message for it. I'm assuming it came from the SongKong code?

Thanks!
Thanks!
Not holding my breath on IPv6. Verizon wasn't involved in IPv6 day last year which probably means it's not on their radar. I know Comcast was part of that effort.

The only other solution would be to buy software that allows you to have an ip from some other country or some other part of the US.

I will probably just run the software on one computer at a time.

Thank you.
Paul
Absolutely. I've been trying to run it on my Mac and my Windows computer at the same time.

I guess I could run one through a VPN tunnel?

Does this same rule apply to Jaikoz?
You don't need it if there's a way to make Musicbrainz queries work all the time

I'm not up to date on the R&D for that product but I really do hope it is improved in the near future.

My suggestion is an improvement in their database design, but I know that's a lot of work.

At some point you have to ask yourself is there another lookup product that works better or how to improve what they have in Musicbrainz.

I have very little to no knowledge in this area. Just wondering if other databases exist of music tags based on Amazon ID, Google ID, AMG, etc. Something that can be coded for, even optional in Jaikoz and Songkong, to improve the chances of getting the meta-data to come back for music tags and not blow up, plus be accurate.

Thank you again for all of your hard work. I know you put a lot of time and effort into making these two products the best they can be.
Maybe I missed it but can't find anything on pointing to your own Musicbrainz server like you do in Jaikoz. I tried searching the forum and using the issue navigator. I didn't see anything about this in either place. Did I miss it?

Thank you.
Memory error: out of heap space

06/03/2013 07.03.45:com.jthink.songkong.ui.startdialog.StartFixSongsDialog$Fix:doInBackground:SEVERE: Java heap space
java.lang.OutOfMemoryError: Java heap space
at java.awt.image.DataBufferByte.<init>(Unknown Source)
at java.awt.image.ComponentSampleModel.createDataBuffer(Unknown Source)
at java.awt.image.Raster.createWritableRaster(Unknown Source)
at javax.imageio.ImageTypeSpecifier.createBufferedImage(Unknown Source)
at javax.imageio.ImageReader.getDestination(Unknown Source)
at com.sun.imageio.plugins.jpeg.JPEGImageReader.readInternal(Unknown Source)
at com.sun.imageio.plugins.jpeg.JPEGImageReader.read(Unknown Source)
at javax.imageio.ImageIO.read(Unknown Source)
at javax.imageio.ImageIO.read(Unknown Source)
at com.jthink.songkong.analyse.analyser.SongChangesReport.createImageUrl(SongChangesReport.java:1270)
at com.jthink.songkong.analyse.analyser.SongChangesReport.outputSongChangedDetails(SongChangesReport.java:1361)
at com.jthink.songkong.analyse.toplevelanalyzer.FixSongsReportCreator.createReport(FixSongsReportCreator.java:53)
at com.jthink.songkong.ui.startdialog.StartFixSongsDialog$Fix.doInBackground(StartFixSongsDialog.java:1384)
at com.jthink.songkong.ui.startdialog.StartFixSongsDialog$Fix.doInBackground(StartFixSongsDialog.java:1374)
at javax.swing.SwingWorker$1.call(Unknown Source)
at java.util.concurrent.FutureTask$Sync.innerRun(Unknown Source)
at java.util.concurrent.FutureTask.run(Unknown Source)
at javax.swing.SwingWorker.run(Unknown Source)
at java.util.concurrent.ThreadPoolExecutor.runWorker(Unknown Source)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(Unknown Source)
at java.lang.Thread.run(Unknown Source)
I've seen the errors in the forum for Jaikoz. Any suggestions on what to do about this for SongKong?. Windows seems to be especially bad. Sometimes it will load all the songs and the fingerprints and just stop with 503 error.
Maybe it's just me but I'm getting a much higher percentages of matches on Discogs with SongKong than with Jaikoz. For whatever reason it's nice to see.
I created a smart playlist with all of the songs/files that needed to have the tags cleaned up in Itunes. I selected all of the files and dragged them over to Jaikoz. I then cleaned them up and did a save command. The files were put into the move folder defined in Jaikoz preferences. However the original files defined in Itunes, the smart playlist,, and copied to Jaikoz are all still in the original location. So it appears that the files were copied rather than moved to the move folder defined in Jaikoz preferences.

Is there any way to do a move instead of the save doing a copy? The advantage of dragging the songs from the smart playlist in Itunes is that I don't need to look up the file location for each song.

Thanks everyone...
As Paul knows, I am one of the people that has reported this difference. To be clear, I noticed this when running Musicbrainz Picard. For my Windows 7 64-bit computer it was an unbelievabel difference in performance. So much so that it made me wonder if I have an issue that's uniqie to my own computer. However recently I upgraded my "guts" of my computer from an AMD 3 year old environment to an Intel Ivy Bridge setup and noticed the same improvement.

It seems to be a difference in performance of the database lookups. I will switch back to using musicbrainz.org and do some more testing.

Paul, I apologize for not sending the data your way. I forgot but will start on it now.

Paul S.
Hi everyone,

I get an error 400 sometimes when I submit a query to Musicbrainz. I know the error is well documented. However if a person is running their own Musicbrainz database server with a search index, defining both in the Jaikoz options, what exactly is occuring when a query is submitted? I only get a few of these out of several hundred. However I thought this error would go away when using my own server and search index. So is the error 400 a query failure because the data is either not in the Musicbrainz database or the search index? This is not an issue of too many queries being submitted. Also my database is not up to date. I haven't run replication in several weeks.

In a related question, if I submit a pairing, does the Jaikoz software know enough to send the updates to the normal musicbrainz.org site, even when I don't have it defined at all in my preferences. Or does the update take place in my database and sent over another way?

Thanks, Paul
35Mb up and down. Two machines going...
Any chance I can get more of you to post this database file? I promise to keep it up for awhile since I have fios. Just taking forever to download...need more participation please
So I get one of these timeout issues. I save what's been processed. Then I want to restart autocorrect metadata from MB. I can easily see in the Finder or Windows Explorer what songs (files) have been updated and reload what was not processed. Is there another way to get it to restart from the point of the timeout, maintaining currency for what's been processed? The filters don't seem to help. Maybe this is a newbie question but I couldn't find an answer for it in the forum.

Thanks, Paul S.
So after my 20th timeout of Musicbrainz, even though I've never tried Linux, I'm thinking about building the server at home.

I see Paul has the manual instruction. There is a virtual machine you can download if you don't want to use Amazon. Amazon you must manually build the database server.

http://musicbrainz.org/doc/MusicBrainz_Server/Setup

https://github.com/metabrainz/musicbrainz-server/blob/master/INSTALL.md

So I'm interested in getting some ideas from anyone who has done this before. I use Vmware at work so I'm very familiar with that software.

I have both Fusion and Workstation products at home. We use ESX at work and VSphere 4.1. Haven't tried Esxi yet.

Putting it on my Qnap NAS would be the best place for it. However if I must put it on a Mac or Windows machine and make this virtual I will do it.

Can anyone offer some suggestions or tips before I start? Has anyone tried either the manual setup or the Linux virtual machine for an MB database?

Thanks! Paul S.
 
Profile for dbaps -> Messages posted by dbaps [39] Go to Page: 1, 2 Next 
Go to:   
Powered by JForum 2.1.6 © JForum Team