I'm having some issues where if I'm accessing a file on one share (eg streaming via XBMC box, or torrenting a file), and I access another share (even to browse), my original session will stutter. If streaming a movie, it will pause to buffer, or if I'm writing or seeding a file, the torrent will halt claiming the network share was no longer accessible. Granted it's not all of the time, but maybe a 1/4 of the time

 

I'm just wondering if this is normal behavior due to spin-downs and the such. It doesn't seem like it should be, and I'm hoping I just screwed up somewhere, but outside of my virtual setup, my UnRAID install is very simple (Lots of drives, no plugins, minimal users, etc). Any guidance is much appreciated

Is the same client performing both accesses? Or 2 different client?

Is the same client performing both accesses? Or 2 different client?

 

Two different clients. For instance, this just happened right now (which caused me to remember my post!)

 

Machine 1: OpenELEC/XBMC machine streaming from "HD Movies" share

Machine 2: Windows 8 browses to "TV" share (Browsing to share, no file access yet), unRAID spins up drive, Machine 1's stream pauses until drive spin up is finished (and Windows loads/sees directory at practically same time)

Is the same client performing both accesses? Or 2 different client?

 

Two different clients. For instance, this just happened right now (which caused me to remember my post!)

 

Machine 1: OpenELEC/XBMC machine streaming from "HD Movies" share

Machine 2: Windows 8 browses to "TV" share (Browsing to share, no file access yet), unRAID spins up drive, Machine 1's stream pauses until drive spin up is finished (and Windows loads/sees directory at practically same time)

It is for this exact reason spinup groups were created.

It is for this exact reason spinup groups were created.

 

I had to Google that, as I had no idea I needed to set it. This is what I found: http://lime-technology.com/wiki/index.php/Configuration_Tutorial

 

Not much information there, and the thread it links to is 4 years old. Is all that still up to date?

 

Also, what is the reason this is required? Is it a software limitation? I came from a Windows Server based solution, and I had drives spinning down and waking on the same controller without affecting current access

 

Also, what is the reason this is required? Is it a software limitation?

 

Bump.

 

Still looking for answers, as this plagues me multiple times a day. Hopefully this is something that will be addressed in the newer releases

It is a driver, hardware, or VM issue. It's not an unRAID issue; although, if it's caused by the driver then a future version of unRAID with a future driver may solve the problem; however, it is unlikely that the driver is the cause. Use spin-up groups if this is an issue.

Is there anyway we can get native driver support? Because this is almost crippling. I've got files that are in constant use on one (or more) terminal(s), and whenever I try to access other files, or create new ones from another terminal, it just stalls activity on both terminals for 60 seconds or so. I even turned off spindown a few months ago, so I can't see what else the issue would be anymore besides the open source implementation of the LSI driver

This should not be happening... moving this to Defect List board...

This should not be happening... moving this to Defect List board...

 

Any word? I'm currently running natively instead of virtualized, and I'm getting the same issues

This should not be happening... moving this to Defect List board...

 

Any word? I'm currently running natively instead of virtualized, and I'm getting the same issues

Please re-test, numerous changes have been made which should help this situation of multiple streams.

Please re-test, numerous changes have been made which should help this situation of multiple streams.

 

Noticed the issue again today. Working on my Windows laptop, editing NFO files (for XBMC/Kodi) in Notepad directly on the share, while on my HTPC I was running Kodi/OpenELEC browsing and watching files (read only user) from the same share. I could open the NFO files just fine, but it seemed when I changed something in them, and it was on a drive that hasn't been used in a while, it would lock up the system, and I wouldn't be able to access anything either on my laptop or my HTPC. Kodi would spit out a "File not available, would you like to remove from library?" error until about 30 seconds later, when unRAID would become responsive again, and I could access it from both systems

 

Just to note: I do not have a spin down delay (Set to never), and the share/root folder I was working within is set for a split level of 1/Automatically split the top level directory as required (In case there was a thought unRAID might be trying to relocate the NFO file when saving)