Streaming HD Video from NAS to WDTV Live Streaming

The Lacie is currently only operating as a Media Server.  I am not using it for any other functions.  Any downloading is done using my PC’s resources, and then transferred over to the NAS at a  later time.  I’ve tried to isolate the NAS from as many operations as possible.

In terms of checking the system resources usage on the Lacie, I haven’t found a way to do that, though you would think there must be a way.  Believe me, over the last couple of weeks I’ve read the user manual back to front and can’t find anything relating to tracking system resources.

The only interesting thing I found was one page describing the blue LED lights behaviour on the back of the NAS.  Unless the NAS is booting up or being backed up, the LED’s should be solid.  But I’ve started to notice that every 3-5 seconds the LED’s will blink, though not in unison.  I have no Backup tasks scheduled for the NAS, I do all the backing up manually to other drives when needed.  So what is the Lacie doing?  Maybe it is a faulty power supply causing the lights to blink?

Another thing I tried to do today was have the Lacie Reindex the media files.  This task failed and this is the first time it has failed.  I’ve run this task a number of times in the past with no issues.  I’ve rebooted the NAS a couple of times and tried to Reindex again and it failed every time.  Maybe this is a clue to the overall issue as well?

Also, coming back to the WDTV for a second, is it better to simply turn it off from the remote, or also turn off the power to it on the powerboard so it doesn’t have power to it when not in use?  Or would I be straining it more by having it reboot every time I want to use it?

I think the Lacie is just “checkin’ stuff” when it’s LED flickers every so often.  I have external drives that do this sometimes.  Not a problem for me.

As for turning it off and on, well, I prefer drives on my WD that have a sleep function; they shut down after 5 mins of non-use, but then they ramp back up when accessed.  We either watch a movie, or listen to music, so they run during that time, but when we are done, it is for quite a while, so then the drives idle.  The ones for the WD don’t get accessed off and on a lot all day long.  So, I never turn off the WD or it’s drives, and the only start up time is the drives waking up…  Maybe the Lacie has a function like sleep/idle as well.

Ok I think the issue is pretty clear now.  One of the disks must be failing.  I’m hoping the Raid Management determines that soon so I can replace it.

I plugged in a fairly new laptop to the network and streamed a file from both the Lacie and a HDD plugged via USB to my PC on the network.  When streaming from the Lacie the video would again intermittently freeze, then pixalation would occur for a second or two, and then the video would resume.  Similar symptoms to streaming through the WDTV.  When streaming from the HDD attached to my PC the same file played with no issues,

I was also sitting next to the Lacie while streaming and I heard the same distinct clicking sounds I heard before one of the disks failed last year.

The frustrating thing is the Raid is not picking up the issue yet which means I can’t replace the disk until it does.  But it’s pretty clear to me that one of the disks is the issue.

well that’s a bummer, I hope it’s not data corruption

RAID does not solve or detect data corruption

http://www.tldp.org/HOWTO/Software-RAID-HOWTO-6.html

probably first thing you should do is make sure you have backups of everything

This is probably a really dumb question but is data corruption a symptom of an impending mechanical failure of a disk?  I’ve basically only got some practical experience from working for Telco’s over the last 10 years, and with playing around with my home network, but with no IT training I reach a point where fundamental knowledge is needed.  Practical experience is only getting me so far.

I"ve been slowly pulling things off the Lacie over the last week.  Once I’ve backed everything up I’ll do a format and rebuild the Raid.  Hopefully through that process it will find a fault with one of the disks.  It was only a matter of time I suppose.  Considering I still have 4 out of the 5 original disks from when I first purchased the Lacie almost 4 years ago, with the 5th failing and being replaced about 12 months ago, the remaining 4 were due to fail eventually.

Moving forward I’m not sure if I should be always leaving the Lacie on or powering it down when not in use.  There seems to be 2 schools of thought on this.  A lot of people in the IT department at my work say they always power down their NAS when not in use, while others say repeated reboots strain the disks even more.  The other downside to rebooting the Lacie, which is what stopped me from doing it in the past, is that it takes a good 15-30 minutes for the media library to become fully visible on the WDTV after a reboot of the Lacie.

  This is probably a really dumb question but is data corruption a symptom of an impending mechanical failure of a disk?

Well, it could have both mechanical or electrical failure.  Worst case kind of mechanical is the arm with the pickup on it could actually break, or take a dive and gouge data.  See if you can find out which one is failing  (look on the drives to see which videos have been bad – could all be on the same disk.)  If you can tell, back that one up first.  Damage could have already occured, but nothing can be done about that  – just hope most of the data can be saved.  I would actually turn the thing off if not hunting down data.  If a drive completely fails you will be hard pressed to recover data; I would not wait for that to happen.  I have a friend who’s drive lost one of the arm pickups.  He opened it up, attached a new arm and was able to recover the data.  Kind of a tricky manuver, of course.  You don’t want to go there.

On or off?  Well, since it takes 10-15 mins before it is all useable, that suggests leaving it on, but spinning 24/7 really adds wear and tear from friction, whereas turning it off lowers that problem.  I leave components with no moving parts, on, and turn off those with moving parts (or at least they can go into sleep mode part of their lifetime.)  Works for me…

Hmmm, not seeing an obvious way to identify which file is on which disk.

Also, since I’m on Raid 5, should the disk fail, wont the data be reconstructed once I replace it with a new disk?  The Lacie supports hot swap.

I do not have a RAID unit, but from what I know I suppose it will reconstruct, but you know better than I do on this matter.  Too bad the unit apparently can’t diagnose its own demise or tell you where problems could be.  Meanwhile, you can’t use the thing now in a quality manner because your videos work poorly.

Is there some way to see the logs on the NAS unit?  When a video freezes/stutters, it is likely that you are getting a read error on one of the drives.  The fact that it eventually continues would suggest to me that the unit then reads the block it needs from another drive.  There should be indications of problems in the unit’s logs.  What I don’t understand is why a read error doesn’t  cause some kind of alert for you.  I would suggest trying to look at the logs; a quick google search came up with an option to download the log file from one model.  Hopefully yours has such an option.

Frankly, I have yet to encounter a failure in one of my (Linux) RAID drives, so I don’t have any detailed knowledge of how standard Linux RAID handles disk read errors (and we don’t know how your unit is modified).  The Linux md man page discusses it, though.  I did just have one of my external video drives fail a few weeks ago.  Failure was clear in the logs as read errors (kernel attempted to read a certain file block from drive and failed).  As you indicated, clicking noises as drive attempted to read requested block.  Read error would result in Samba returning an error, which caused the SMP to quit playing and return to menu.

@ncarver:

I’m curious… So, was the read error you experienced recently repairable via a re-format, or did the drive physically go kaput and need replacing?

I had the same thing happen to me recently with a non-RAID disk

I had to replace the drive

mike27oct wrote:

@ncarver:

 

I’m curious… So, was the read error you experienced recently repairable via a re-format, or did the drive physically go kaput and need replacing?

I did not attempt to “save” the drive, since my past experience has been that such drives are not trustworthy again even if you can temporarily get them functioning.  This drive developed read errors when accessing several of the video files, so it was not just a minor problem (i.e., one file).  After trying to backup the drive to a new one and being unable to copy several files, I retrieved the backup drive (stored at work) and immediately copied it to a new drive.  The failed drive was of course just out of its warranty period, so $120 down the drain!  (A Seagate drive with one year warranty, BTW.)

Well, too bad about the drive; at least you had a backup!  It could have been any brand that went; I have WD, Hitachi, Seagate, Samsung, and Toshiba either in my PCs and/or external collection.  So, far the only drive that failed was a Seagate in the ol’ XP laptop.  Could not read a certain Windows boot file anymore.  Fortunately, the PC  could be booted via the CD player using some sort of Puppy Linux on CD,  then was able to access the internal HD, hook up an external HD and copy ALL my personal data off the drive!  Considered myself very fortunate.  Bought a new Win 7 and later ordered a new drive for the XP, set it up, and XP is still is working fine today (and currently OFF the internet anymore since April 8, but still on the home network.)  Not all stories end well this way; I was lucky.

Finaly, one of the disks has failed.  Now I can just replace it, let the Raid reconcstruct the data, and hope it goes back to working as normal.  I can hopefully pick up a new disk tomorrow and let you guys know how it goes.

As for the logs, I have downloaded all the logs from the Lacie and there are at least 50 different logs.  I’ve started skimming through a few but not yet finding anything that would’ indicate an issue with the disk.

Anyway, the Raid found the broken disk over night as I was transferring files to an external disk.  The clicking sounds were getting louder and more frequent since my last post.  So at the end of the day, it’s basically taken the disk 3 weeks from first degraded performance to full failure.

Yep, troubleshooting tips from drive manufactures say that clicking disks is a sign of impending failure, and looks like it was true in this case.  Good luck getting it all going again.

Well I’m happy to report that the new disk is in, the RAID has successfully reconstructed the data and everything is playing again perfectly.

It’s a bit disappointing that the disk clearly showed symptoms for a number of weeks before the RAID decided the disk was faulty, but at least everything is now back up and running.

Thanks again for everyone’s help and advice!

I have the same problem, the pausing and then FF to where it catches up.  My disks are fine though.  This happens whether I stream from network shares on my pc or from my nas.

I wonder if my drive’s sleep settings on the pc & NAS could be affecting.  They are set to spin-down at 30 minutes.

Or, could it be my network?  I have attached a diagram of the network.  It was built about 6 months ago using the same brand gigabit switches and cat6 cables.  Excuse the crudity of the sketch.

BTW, the NAS OS is Windows8 and has a span of 5TB (pool, no redundancy) with shares.  The pc is Win7, with the same duplicate HDD setup and shares.  Like I mentioned above, both sources exibit this pausing/FF problem.

Thanks in advance.

newtrk.jpg

Well, four screens full of posts later, and you are back in business.  Congrats.

blanker wrote:

I have the same problem, the pausing and then FF to where it catches up.  My disks are fine though.  This happens whether I stream from network shares on my pc or from my nas.

 

I wonder if my drive’s sleep settings on the pc & NAS could be affecting.  They are set to spin-down at 30 minutes

Wow.   If I didn’t know better, I’d say you were plagiarizing my network diagram.  :)

Mine is set up almost identically;  I use only a NAS though, no PC involved.   But still three switches in the path between my NAS and SMP.   I’ve not had any issues at all.

Pausing in the middle of the movie with a 30-minute “Sleep” setting seems very unlikely;  the server is going to be sending data constantly.   There’s no way the WD could buffer 30-minutes worth, thus no chance your drives should spin down in the middle of a movie … UNLESS the movie spans across disks, and it’s reading from “Disk A” for X amount of time, and then moving onto more data that’s on Disk B which might be spun down.

Should be simple to test – just turn off the drive spin-down and see if the problem goes away.