File Identification Differences using haneWin NFS and Win7 SMB

[bbullins]

All I finally got a chance to setup haneWin NFS server and did notice a substantial speed up over Win7 SMB. I had done many many tweaks to improve SMB and over all it was good, around 40mb speeds but now with NFS bumping 80mb. One difference that has surprised me is that even though the file scanning and ID process is now much faster it does not seem to use the NFO files unless it cant ID the file on its own.

I had corrected file naming structures, created NFO's etc and pretty much had the boxee id'ing perfectly when I had to re-add my sources. My experience with SMB is it would always use my NFO upon system scan. With NFS it seems to try and indentify first and then if it cant uses the NFO. The problem is it erroneously ids many things and even when they are right I had found a few things to have some pretty bad cover art and fixed with NFO files. bottom line is SMB ids correctly but seems much slower, NFS and Hanewin much faster but ID's differently.

The question - is there a way to force it to use NFO first like it seems to do with SMB, could this be a setting on the NFS server? Thanks

[war59312]

You simply need to remove the "id" field from all .nfo files.

Easy enough to do using grepWin @ https://code.google.com/p/grepwin/

I used it to remove the id field from over 10,000 files and took less than 1 minute for it to do so.

Note: I use a very fast RAID setup in hardware so expect it to take a bit longer on a "normal" setup. It still is quick and does all the work for you.

Search For:

\s\s<id>.+</id>\s\s

And leave "Replace With" blank.

[bbullins]

Thanks, Cool tool! I have installed and got rid of all id labels. Is there a way to tell ember to not put them in there for the future? Also any clue why it causes issues with NFS but not SMB?

You simply need to remove the "id" field from all .nfo files.

Easy enough to do using grepWin @ https://code.google.com/p/grepwin/

I used it to remove the id field from over 10,000 files and took less than 1 minute for it to do so.

Note: I use a very fast RAID setup in hardware so expect it to take a bit longer on a "normal" setup. It still is quick and does all the work for you.

Search For:



And leave "Replace With" blank.

[war59312]

Sadly you can NOT tell Ember not to add IDs.

And no clue why NFS was only effected for you. Both SMB and NFS behavior this way for me.

PS: Please don't quote entire posts when replying. It makes it harder to read your response and is a waste of space.

Quoting is meant to be used when replying to something specific within a post and normally open if replying on a different page.

Thanks!

[bbullins]

Thanks, ill try to remember that

a waste of space.

[war59312]

No problem. Thanks!

So did you clean out your database in settings and do a rescan and confirmed its working as expected?

[bbullins]

yes, thanks

[war59312]

Glad to hear it.

[Nicias]

Here is a linux one-liner that removes the <id> fields from all of the .nfos in a folder (including subfolders)

find -name *.nfo -exec sed -i -e '/<id>/d' {} \;

[aaronwt]

Thanks, Cool tool! I have installed and got rid of all id labels. Is there a way to tell ember to not put them in there for the future? Also any clue why it causes issues with NFS but not SMB?

I've been using Ember for a while with my content. Whether the content is over NFS or SMB my Boxee Boxes have always pulled the info from the NFO without issue.

The only reason I even switched some of my shares to NFS was because i ran into a couple of ISOs that had issues over SMB and they were fine when I played it over NFS. So I figured I would use NFS to avoid the chance of an issue in the future. Otherwise my ISOs played fine over SMB. Scanning has been just as quick or slow(depending on how you look at it)