I am having some trouble with intermittent drop outs. The problem seems to be more sever when the drives first spin up but it is an issue when all drives are spinning as well. I have loaded the latest beta 4.3.6 and I am still having the same problem. I am running Vista ultimate Media Center with the My Movies plug-in that loads my ISO from the unraid server to the Vista box with Daemon tool. When I fist load the movie it starts fine and a couple of minutes in it will freeze and Daemon Toll will un- mount the image. I have traced it back to the Unraid box dropping the connection, sometimes it will just freeze for a couple of seconds and continue to play and other times it will freeze for too long and Daemon tools will drop the ISO file. I have monitored this in Task manager under the network tab. I have this same issue on the 3 Vista boxes in my house.

 

I have no problems writing files to the Unraid box or copying file from it, my problem is just when I stream.

 

I am using an AMD setup with Asus M2NPV-VM mother board; I am using the integrated network 1GB I have 8 serial drives and 3 IDE, Silencer 610 W power supply

 

I have attached my syslog. If you need any more info let my know.

 

Thanks,

 

Normally, streaming should be less of a problem than copying a file, since streaming even an HD video is a fairly low bandwidth operation.  I don't recall the exact numbers, but reading a file from your server could be about 25MB/s, streaming HD video might be about 8MB/s, streaming an ordinary video might be 3 MB/s, and streaming an audio file might be about 1MB/s (all estimated numbers to give a rough relation between them).

 

You are having drive problems, which may or may not explain the issues you are seeing.  They involve the 4 Maxtor 500GB drives (Maxtor_7H500F0), Disk 1 (sdd), Disk 2 (sda), Disk 3 (sdb), and especially Disk 7 (sdg).  The errors are triggered in every case by a direct command to spin down, which you appear to have repeatedly directed, probably while testing.  I have no idea why these drives don't like being told to go to sleep, but they don't.  There is no evidence of problems with spinning up, or their own built-in spin down clocks, or from any other cause.  The problems were usually exception Emask (timeout) errors, but Disk 7 also had exception Emask (host bus error)'s, and more errors than the other 3.  One possibility, other than a general dislike of being told to sleep, is that these drives have inferior SATA cables.  You might substitute different cables, preferably higher quality SATA II rated cables, and see if that makes a difference.  The obvious cure is never click the Spin Down button, but let them sleep when their own timers tell them to.  I suspect that delays spinning the drives up and/or initiating transfers or streaming may have looked like the fault of the Spin Up button, but were in fact caused by the previous Spin Down action that still hadn't finished because of errors.  Those errors took awhile to resolve.

 

Another and different issue, Disk 8 (hda) had difficulty just after startup, but after slowing itself down to UDMA/66 (was initiated at UDMA/100), never had any more trouble.  That may mean it is an old UDMA/66 drive, or it may mean that it has a bad cable.  See this Wiki section about IDE cables: http://lime-technology.com/wiki/index.php?title=Troubleshooting#Hard_drive_failures.

 

I know this doesn't completely explain the difficulties with streaming you are seeing, but there weren't any other clues in your syslog to help me diagnose it.

 

Thank for taking the time to look it over. I am using high quality locking sata cables, all sata cables are the same. I think the drop outs are related to Vista, I have been streaming to an XP box and it is a smooth. the Vista box streams big chunk and then goes ideal for a couple of seconds and then takes another chunk. I am in the proses of loading XP Media center, I will let you know if this solves my problems.

 

Thanks,

Same problem here. Interestingly I also have Vista machines around the house. I was playing a divx file yesterday, and it stopped 5 times. It started again after 5-10 seconds freeze. I am running 4.3 beta 1 currently.

BTW, this is my syslog that does not seem right...

 

Tower login: root

Linux 2.6.24.3-unRAID.

root@Tower:~# cat /var/log/syslog

May  4 04:40:01 Tower syslogd 1.4.1: restart.

root@Tower:~# cat /var/log/syslog

May  4 04:40:01 Tower syslogd 1.4.1: restart.

root@Tower:~#

How long has the machine been running?

logrotate truncates the file when it is > 1M.

Mine not more than two weeks. I had to reboot as I could not browse any of the data, whilst the server management page was not reporting any problems.