jimwhite Posted November 22, 2009 Share Posted November 22, 2009 I mapped two drives on my Windows 7 machine to the cache drive, one via SAMBA and one via NFS. Interestingly enough, copies to and from the cache drive via SAMBA were aboyt 80 mB/sec while via NFS they were about 33 MB/sec. I am running beta 11, NCQ enabled and the extra config posted by Limetech. Quote Link to comment
BLKMGK Posted November 22, 2009 Share Posted November 22, 2009 Argh! Saw this behaviour for th efirst time ever last night while watching a recorded TV show. Lag was so bad the stream timed out and I had to restart it from where it dropped. I have only JUST switched to a user share for those shows but I don't think this was the issue. I had my laptop out and spun up a drive trying to access something - was shocked when the stream died. It never dies! Possibly this is the first time I've ever caused a drive to spin up but at least having read this I knew what was going on. This box has a couple of good SATA drives in it and a pile of older IDE I need to kill off. Below is my hardware dump. Sadly I don't recall which board and CPU is in this beast but if it turns out to be important I can pull out a screwdriver, I just hate getintg into it because I always bump an IDE cable out of kilter - grr! I think I will bump cache way up on XBMC for starters, we'll see if I can solve this with that. Now here's where I *DO* see something strange. Previously copying to this as a disk share (on a SATA) on beta 8 I saw FAST write speeds - awesome! Moved to beta 11 and still good speeds. But... Occasionally I see SLOW speeds. Fast can be as high as 30megs, but when it is slow it's as low as 3megs. I THINK I never saw this before because I was using a disk share mapped to a SATA drive. Now with a user share I'm pretty sure an IDE has gotten thrown in there - yuck! But 3megs on a write still seems awful low! Are others observing this? That machine has a little bit of memory but I've not yet tweaked the settings. I run unMENU but little more in the way of tweaks right now. I'm tempted to bump memory and tweak things. so far my second box full of just SATA seems to have no issues with write speeds and I've not yet checked it for spinup issues. parity device: pci-0000:00:1f.2-scsi-0:0:0:0 (sda) ata-ST31500341AS_9VS0SMTZ disk1 device: pci-0000:00:1f.2-scsi-1:0:0:0 (sdc) ata-ST31500341AS_9VS0KRYR disk2 device: pci-0000:00:1f.2-scsi-0:0:1:0 (sdb) ata-ST31500341AS_9VS0T46P disk3 device: pci-0000:00:1f.2-scsi-1:0:1:0 (sdd) ata-ST31500341AS_9VS0X1T5 disk4 device: pci-0000:03:03.0-ide-0:0 (hdi) ata-ST3300831A_4NF08RJX disk5 device: pci-0000:03:03.0-ide-0:1 (hdj) ata-ST3300831A_3NF0WH07 disk6 device: pci-0000:03:03.0-ide-1:0 (hdk) ata-ST3500641A_3PM0JFCG disk7 device: pci-0000:03:03.0-ide-1:1 (hdl) ata-ST3500641A_3PM00HC4 disk8 device: pci-0000:03:02.0-ide-0:0 (hde) ata-ST3400620A_5QH030AD disk9 device: pci-0000:03:02.0-ide-0:1 (hdf) ata-WDC_WD3000JB-00KFA0_WD-WCAMR1641443 disk10 device: pci-0000:03:02.0-ide-1:0 (hdg) ata-WDC_WD3000JB-00KFA0_WD-WCAMR1587689 disk11 device: pci-0000:03:02.0-ide-1:1 (hdh) ata-ST3500641A_3PM11V97 P.S. Looking at tuning some of the new variables through the WEB interface. Previously if these changes were made in the config file you had to start\stop the array. However these cuold be issued via Telnet and would apply immediately. With these being moved to the WEB interface from the config file do we still need to start\stop the array? Umm has anyone considered making some sort of calculator app so that we could tune these? Maybe we could plug in system memory, number of disks, and ballpark some of the other variables? Maybe unRAID could look at max system emmory and modify these on the startup? Dynamic changes on the fly might also be interesting - what happens to a running system if you reduce these numbers from the commandline via Telnet - anything bad? If not then dynamic seems potentially applicable if there's CPU to spare... Quote Link to comment
Kaygee Posted November 22, 2009 Share Posted November 22, 2009 Download the smb-extra.conf file from page six I think. That should restore your upload speeds nicely, lot more stable too. It looks like all your sata devices are in IDE mode (paired). Worth checking in the bios to see if the controller supports AHCI. Parity + disk2, disk1 + disk3. Worth trying force ncq disabled = yes. Also I only notice this issue when cache is emptied so adding more memory to the server if possible is a good idea. parity device: pci-0000:00:1f.2-scsi-0:0:[glow=red,2,300]0[/glow]:0 (sda) ata-ST31500341AS_9VS0SMTZ disk2 device: pci-0000:00:1f.2-scsi-0:0:[glow=red,2,300]1[/glow]:0 (sdb) ata-ST31500341AS_9VS0T46P disk1 device: pci-0000:00:1f.2-scsi-1:0:[glow=red,2,300]0[/glow]:0 (sdc) ata-ST31500341AS_9VS0KRYR disk3 device: pci-0000:00:1f.2-scsi-1:0:[glow=red,2,300]1[/glow]:0 (sdd) ata-ST31500341AS_9VS0X1T5 Quote Link to comment
BLKMGK Posted November 22, 2009 Share Posted November 22, 2009 Download the smb-extra.conf file from page six I think. That should restore your upload speeds nicely, lot more stable too. It looks like all your sata devices are in IDE mode (paired). Worth checking in the bios to see if the controller supports AHCI. Parity + disk2, disk1 + disk3. Worth trying force ncq disabled = yes. Also I only notice this issue when cache is emptied so adding more memory to the server if possible is a good idea. parity device: pci-0000:00:1f.2-scsi-0:0:[glow=red,2,300]0[/glow]:0 (sda) ata-ST31500341AS_9VS0SMTZ disk2 device: pci-0000:00:1f.2-scsi-0:0:[glow=red,2,300]1[/glow]:0 (sdb) ata-ST31500341AS_9VS0T46P disk1 device: pci-0000:00:1f.2-scsi-1:0:[glow=red,2,300]0[/glow]:0 (sdc) ata-ST31500341AS_9VS0KRYR disk3 device: pci-0000:00:1f.2-scsi-1:0:[glow=red,2,300]1[/glow]:0 (sdd) ata-ST31500341AS_9VS0X1T5 Okay go slow - trying to comprehend I have NCQ disabled = yes set in the WEB gui. This machine has at least a gig of memory but bumping it is likely doable. I popped the side off, this is a system I upgraded most recently! Supermicro board with DDR3 memory Stupidly it seems I pitched the Mobo box in a cleaning fit. Looks like it was a C2SEE-O motherboard and it has 2gig memory in it. Looks like I have room to tune memory parameters! I have two Promise IDA cards in there. I have, over time, tried to push all of the data OFF the IDE drives. I need to finish this I thihnk and yank them. I have a spare 1.5TB SATA I can put in there to bump storage. I will work today to remove drives. Looks like I might need to purchase a SATA card soon - need to research that purchase. Fun stuff! I am ASSuming that removing the IDE drives will help this situation. I will also look into AHCI, I'm almost certain this board supports it. http://www.newegg.com/Product/Product.aspx?Item=N82E16813182162 Quote Link to comment
limetech Posted November 22, 2009 Author Share Posted November 22, 2009 Working on a solution to this (spinup groups), it's a pretty big change. Also, I'm not sure inspection of the PCI id string on the Devices page is good enough to tell if disks share a controller. Another place to check is to do a listing of /sys/block, as in v /sys/block Quote Link to comment
BLKMGK Posted November 22, 2009 Share Posted November 22, 2009 Working on a solution to this (spinup groups), it's a pretty big change. Also, I'm not sure inspection of the PCI id string on the Devices page is good enough to tell if disks share a controller. Another place to check is to do a listing of /sys/block, as in v /sys/block Well, I think dumping the IDE nees to be done on my box anyway and truthfully I'd never seen this issue until last night. I wouldn't have even recognized it had I not seen this discussion! I cannot recall ever having had a stream stop like that but I was able to connect my action with what occured and recognized it. No biggie really, I'd have probably written it off to XBMC otherwise Here's the output of that command on my box - note that I've added another SATA since my previous posting and that it's clearing now. As soon as it's done I will clear off the IDE and remove them from th ebox... root@BEAST:~# v /sys/block total 0 lrwxrwxrwx 1 root root 0 Nov 22 15:19 hde -> ../devices/pci0000:00/0000:00:1e.0/0000:03:02.0/ide2/2.0/block/hde/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdf -> ../devices/pci0000:00/0000:00:1e.0/0000:03:02.0/ide2/2.1/block/hdf/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdg -> ../devices/pci0000:00/0000:00:1e.0/0000:03:02.0/ide3/3.0/block/hdg/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdh -> ../devices/pci0000:00/0000:00:1e.0/0000:03:02.0/ide3/3.1/block/hdh/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdi -> ../devices/pci0000:00/0000:00:1e.0/0000:03:03.0/ide4/4.0/block/hdi/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdj -> ../devices/pci0000:00/0000:00:1e.0/0000:03:03.0/ide4/4.1/block/hdj/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdk -> ../devices/pci0000:00/0000:00:1e.0/0000:03:03.0/ide5/5.0/block/hdk/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdl -> ../devices/pci0000:00/0000:00:1e.0/0000:03:03.0/ide5/5.1/block/hdl/ lrwxrwxrwx 1 root root 0 Nov 22 18:14 loop0 -> ../devices/virtual/block/loop0/ lrwxrwxrwx 1 root root 0 Nov 22 18:14 loop1 -> ../devices/virtual/block/loop1/ lrwxrwxrwx 1 root root 0 Nov 22 18:14 loop2 -> ../devices/virtual/block/loop2/ lrwxrwxrwx 1 root root 0 Nov 22 18:14 loop3 -> ../devices/virtual/block/loop3/ lrwxrwxrwx 1 root root 0 Nov 22 18:14 loop4 -> ../devices/virtual/block/loop4/ lrwxrwxrwx 1 root root 0 Nov 22 18:14 loop5 -> ../devices/virtual/block/loop5/ lrwxrwxrwx 1 root root 0 Nov 22 18:14 loop6 -> ../devices/virtual/block/loop6/ lrwxrwxrwx 1 root root 0 Nov 22 18:14 loop7 -> ../devices/virtual/block/loop7/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 sda -> ../devices/pci0000:00/0000:00:1f.2/host1/target1:0:0/1:0:0:0/block/sda/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 sdb -> ../devices/pci0000:00/0000:00:1f.2/host1/target1:0:1/1:0:1:0/block/sdb/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 sdc -> ../devices/pci0000:00/0000:00:1f.2/host2/target2:0:0/2:0:0:0/block/sdc/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 sdd -> ../devices/pci0000:00/0000:00:1f.2/host2/target2:0:1/2:0:1:0/block/sdd/ lrwxrwxrwx 1 root root 0 Nov 22 15:22 sde -> ../devices/pci0000:00/0000:00:1f.5/host3/target3:0:0/3:0:0:0/block/sde/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 sdf -> ../devices/pci0000:00/0000:00:1d.7/usb2/2-4/2-4:1.0/host0/target0:0:0/0:0:0:0/block/sdf/ P.S. And no I don't know Linux nearly well enough to make heads or tails of that! Quote Link to comment
BRiT Posted November 22, 2009 Share Posted November 22, 2009 lrwxrwxrwx 1 root root 0 Nov 22 15:19 hde -> ../devices/pci0000:00/0000:00:1e.0/0000:03:02.0/ide2/2.0/block/hde/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdf -> ../devices/pci0000:00/0000:00:1e.0/0000:03:02.0/ide2/2.1/block/hdf/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdg -> ../devices/pci0000:00/0000:00:1e.0/0000:03:02.0/ide3/3.0/block/hdg/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdh -> ../devices/pci0000:00/0000:00:1e.0/0000:03:02.0/ide3/3.1/block/hdh/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdi -> ../devices/pci0000:00/0000:00:1e.0/0000:03:03.0/ide4/4.0/block/hdi/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdj -> ../devices/pci0000:00/0000:00:1e.0/0000:03:03.0/ide4/4.1/block/hdj/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdk -> ../devices/pci0000:00/0000:00:1e.0/0000:03:03.0/ide5/5.0/block/hdk/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdl -> ../devices/pci0000:00/0000:00:1e.0/0000:03:03.0/ide5/5.1/block/hdl/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 sda -> ../devices/pci0000:00/0000:00:1f.2/host1/target1:0:0/1:0:0:0/block/sda/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 sdb -> ../devices/pci0000:00/0000:00:1f.2/host1/target1:0:1/1:0:1:0/block/sdb/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 sdc -> ../devices/pci0000:00/0000:00:1f.2/host2/target2:0:0/2:0:0:0/block/sdc/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 sdd -> ../devices/pci0000:00/0000:00:1f.2/host2/target2:0:1/2:0:1:0/block/sdd/ If I'm looking at it correctly, you have a lot of drives sharing connections, in the following sets: (hde, hdf) (hdg, hdh) (hdi, hdj) (hdk, hdl) (sda, sdb) (sdc, sdd) In your case, the only drive not sharing a connection is SDE. Quote Link to comment
prostuff1 Posted November 22, 2009 Share Posted November 22, 2009 Since Tom is asking for all of this information here is the output from my setup. I have not done extensive testing, as I have not had a lot of time to do it, but if/when I get a chance I will do some testing. Anyway, here is the output from v /sys/block on my system: root@Tower:~# v /sys/block total 0 lrwxrwxrwx 1 root root 0 Nov 22 17:34 fd0 -> ../devices/platform/floppy.0/block/fd0/ lrwxrwxrwx 1 root root 0 Nov 22 17:34 loop0 -> ../devices/virtual/block/loop0/ lrwxrwxrwx 1 root root 0 Nov 22 17:34 loop1 -> ../devices/virtual/block/loop1/ lrwxrwxrwx 1 root root 0 Nov 22 17:34 loop2 -> ../devices/virtual/block/loop2/ lrwxrwxrwx 1 root root 0 Nov 22 17:34 loop3 -> ../devices/virtual/block/loop3/ lrwxrwxrwx 1 root root 0 Nov 22 17:34 loop4 -> ../devices/virtual/block/loop4/ lrwxrwxrwx 1 root root 0 Nov 22 17:34 loop5 -> ../devices/virtual/block/loop5/ lrwxrwxrwx 1 root root 0 Nov 22 17:34 loop6 -> ../devices/virtual/block/loop6/ lrwxrwxrwx 1 root root 0 Nov 22 17:34 loop7 -> ../devices/virtual/block/loop7/ lrwxrwxrwx 1 root root 0 Nov 22 17:34 md1 -> ../devices/virtual/block/md1/ lrwxrwxrwx 1 root root 0 Nov 22 17:34 md2 -> ../devices/virtual/block/md2/ lrwxrwxrwx 1 root root 0 Nov 22 17:34 md3 -> ../devices/virtual/block/md3/ lrwxrwxrwx 1 root root 0 Nov 22 17:34 md4 -> ../devices/virtual/block/md4/ lrwxrwxrwx 1 root root 0 Nov 22 17:34 md5 -> ../devices/virtual/block/md5/ lrwxrwxrwx 1 root root 0 Nov 22 17:34 md6 -> ../devices/virtual/block/md6/ lrwxrwxrwx 1 root root 0 Nov 22 17:29 sda -> ../devices/pci0000:00/0000:00:1d.7/usb2/2-1/2-1:1.0/host0/target0:0:0/0:0:0:0/block/sda/ lrwxrwxrwx 1 root root 0 Nov 22 17:13 sdb -> ../devices/pci0000:00/0000:00:1f.2/host3/target3:0:0/3:0:0:0/block/sdb/ lrwxrwxrwx 1 root root 0 Nov 22 17:17 sdc -> ../devices/pci0000:00/0000:00:1f.2/host4/target4:0:0/4:0:0:0/block/sdc/ lrwxrwxrwx 1 root root 0 Nov 22 17:32 sdd -> ../devices/pci0000:00/0000:00:1f.2/host5/target5:0:0/5:0:0:0/block/sdd/ lrwxrwxrwx 1 root root 0 Nov 22 17:12 sde -> ../devices/pci0000:00/0000:00:1f.2/host6/target6:0:0/6:0:0:0/block/sde/ lrwxrwxrwx 1 root root 0 Nov 22 17:20 sdf -> ../devices/pci0000:00/0000:00:1f.2/host7/target7:0:0/7:0:0:0/block/sdf/ lrwxrwxrwx 1 root root 0 Nov 22 17:35 sdg -> ../devices/pci0000:00/0000:00:1f.2/host8/target8:0:0/8:0:0:0/block/sdg/ lrwxrwxrwx 1 root root 0 Nov 22 17:24 sdh -> ../devices/pci0000:00/0000:00:1c.5/0000:05:00.0/host9/target9:0:0/9:0:0:0/block/sdh/ lrwxrwxrwx 1 root root 0 Nov 22 17:26 sdi -> ../devices/pci0000:00/0000:00:1c.5/0000:05:00.0/host10/target10:0:0/10:0:0:0/block/sdi/ This is again on an Abit AB9 Pro with a Celeron Dual Core 1.6Ghz and 2GB of Ram. Quote Link to comment
BLKMGK Posted November 22, 2009 Share Posted November 22, 2009 lrwxrwxrwx 1 root root 0 Nov 22 15:19 hde -> ../devices/pci0000:00/0000:00:1e.0/0000:03:02.0/ide2/2.0/block/hde/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdf -> ../devices/pci0000:00/0000:00:1e.0/0000:03:02.0/ide2/2.1/block/hdf/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdg -> ../devices/pci0000:00/0000:00:1e.0/0000:03:02.0/ide3/3.0/block/hdg/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdh -> ../devices/pci0000:00/0000:00:1e.0/0000:03:02.0/ide3/3.1/block/hdh/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdi -> ../devices/pci0000:00/0000:00:1e.0/0000:03:03.0/ide4/4.0/block/hdi/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdj -> ../devices/pci0000:00/0000:00:1e.0/0000:03:03.0/ide4/4.1/block/hdj/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdk -> ../devices/pci0000:00/0000:00:1e.0/0000:03:03.0/ide5/5.0/block/hdk/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 hdl -> ../devices/pci0000:00/0000:00:1e.0/0000:03:03.0/ide5/5.1/block/hdl/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 sda -> ../devices/pci0000:00/0000:00:1f.2/host1/target1:0:0/1:0:0:0/block/sda/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 sdb -> ../devices/pci0000:00/0000:00:1f.2/host1/target1:0:1/1:0:1:0/block/sdb/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 sdc -> ../devices/pci0000:00/0000:00:1f.2/host2/target2:0:0/2:0:0:0/block/sdc/ lrwxrwxrwx 1 root root 0 Nov 22 15:19 sdd -> ../devices/pci0000:00/0000:00:1f.2/host2/target2:0:1/2:0:1:0/block/sdd/ If I'm looking at it correctly, you have a lot of drives sharing connections, in the following sets: (hde, hdf) (hdg, hdh) (hdi, hdj) (hdk, hdl) (sda, sdb) (sdc, sdd) In your case, the only drive not sharing a connection is SDE. This is quite posisble. I am using a pair of Promise IDE cards on the PCI bus and nearly all of the connectors for SATA on the motherboard. The IDE I would expect to be sharing connections for sure. the newest added SATA is using a secondary controller on this motherboard I *think*. As soon as this drive clears I will move the data from the IDE, remove them, and make sure that AHCI is set in the BIOS. Considering how slug slow this clearing process is going it's going to be a late night Quote Link to comment
purko Posted November 23, 2009 Share Posted November 23, 2009 Working on a solution to this (spinup groups), it's a pretty big change. I'd rather we get a stable version out the door, and leave the "big changes" (and new bugs) to the version 5 betas. Personally, that "spinup-stutter" thing doesn't bother me at all. I don't let unRAID spin down my disks anyway. I take care of spinning down my unRAID disks myself (via script) after all my HTPCs have gone to sleep. Purko Quote Link to comment
Joe L. Posted November 23, 2009 Share Posted November 23, 2009 Working on a solution to this (spinup groups), it's a pretty big change. I'd rather we get a stable version out the door, and leave the "big changes" (and new bugs) to the version 5 betas. Personally, that "spinup-stutter" thing doesn't bother me at all. I don't let unRAID spin down my disks anyway. I take care of spinning down my unRAID disks myself (via script) after all my HTPCs have gone to sleep. Purko With all due respect, the 5.0 series is intended for a user-interface API overhaul... This data freeze is a operational performance issue, and the current beta series is a good as any place to address it. I think Tom is doing what he needs to do to address an issue that more and more customers are experiencing and complaining about. (I know my wife has complained about the freezing of movies...) I'm very happy investigation of the cause is under way. The "pairing" changes involve the spin-up and spin-down logic, not that used in maintaining parity to protect our data. From what I've seen in the past, these changes are small compared to those needed too support SATA drives a few years ago. Joe L. Quote Link to comment
purko Posted November 23, 2009 Share Posted November 23, 2009 Working on a solution to this (spinup groups), it's a pretty big change. I'd rather we get a stable version out the door, and leave the "big changes" (and new bugs) to the version 5 betas. Personally, that "spinup-stutter" thing doesn't bother me at all. I don't let unRAID spin down my disks anyway. I take care of spinning down my unRAID disks myself (via script) after all my HTPCs have gone to sleep. Purko With all due respect, the 5.0 series is intended for a user-interface API overhaul... This data freeze is a operational performance issue, and the current beta series is a good as any place to address it. Well, Joe, when you're right, you're right. I'm just afraid of "big changes" now that 4.5.11 is so close to "perfect". Purko Quote Link to comment
Joe L. Posted November 23, 2009 Share Posted November 23, 2009 Personally, that "spinup-stutter" thing doesn't bother me at all. I don't let unRAID spin down my disks anyway. I take care of spinning down my unRAID disks myself (via script) after all my HTPCs have gone to sleep. We've got my wife's sister and brother-in-law as house guests this week... and lots of movies being watched from the server... and spin-down set to "Never" (Don't want to have anything to spoil the movie experience while guests are visiting this week) Joe L. Quote Link to comment
Stokkes Posted November 23, 2009 Share Posted November 23, 2009 I just want to add this (probably won't help, but I have to say it). I've only been using unRAID 4.5b## series (haven't used any versions prior). The same hardware prior to using unRAID was being used in a ubuntu Software RAID5 solution and I never experienced any streaming glitches. Since moving to unRAID back in August, I've experienced numerous occasions where my stream (from Plex media center) will simply flake out and stop. Stuttering (for me anyway) is less of an issue than the stream being completely stopped. Again, same hardware, same media center, the only change was going from software raid to unRAID. It happens quite often that I'm watching a video and I'll get kicked out of playback due some "glitch" (which I still think is the spin-up issue). Probably doesn't make sense, but I thought I'd send it out there. Quote Link to comment
prostuff1 Posted November 23, 2009 Share Posted November 23, 2009 I just want to add this (probably won't help, but I have to say it). I've only been using unRAID 4.5b## series (haven't used any versions prior). The same hardware prior to using unRAID was being used in a ubuntu Software RAID5 solution and I never experienced any streaming glitches. Since moving to unRAID back in August, I've experienced numerous occasions where my stream (from Plex media center) will simply flake out and stop. Stuttering (for me anyway) is less of an issue than the stream being completely stopped. Again, same hardware, same media center, the only change was going from software raid to unRAID. It happens quite often that I'm watching a video and I'll get kicked out of playback due some "glitch" (which I still think is the spin-up issue). Probably doesn't make sense, but I thought I'd send it out there. Your problem is probably related to spin up and the reason you never saw this on the software RAID5 was because no disks would have ever been spun down with that solution. I am fairly sure that if you set the drives to never spin down in unRAID you would not have a problem. Quote Link to comment
daniel.boone Posted November 23, 2009 Share Posted November 23, 2009 Additional output as requested Here is my original post with MB details http://lime-technology.com/forum/index.php?topic=4674.msg43026#msg43026 I stream media via http. I did perform some testing with beta11. Accessing any spun down disks dirrectly did not cause any stutter. I did try the telnet session test but no stutter. Was starting to think all is well with my MB. Then I ran a media scan on the video smb share and playing video stuttered while the drives were spinning up. Presently my video stall is limited only to drive spin up when accessing smb share. Spinning up drives using the GUI while a video is playing does not casue a stall. If I spin up drives before running a media scan, no stutter. db lrwxrwxrwx 1 root root 0 Nov 22 20:29 fd0 -> ../devices/platform/floppy.0/block/fd0/ lrwxrwxrwx 1 root root 0 Nov 22 20:12 hda -> ../devices/pci0000:00/0000:00:1c.4/0000:01:00.0/ide0/0.0/block/hda/ lrwxrwxrwx 1 root root 0 Nov 22 20:29 loop0 -> ../devices/virtual/block/loop0/ lrwxrwxrwx 1 root root 0 Nov 22 20:29 loop1 -> ../devices/virtual/block/loop1/ lrwxrwxrwx 1 root root 0 Nov 22 20:29 loop2 -> ../devices/virtual/block/loop2/ lrwxrwxrwx 1 root root 0 Nov 22 20:29 loop3 -> ../devices/virtual/block/loop3/ lrwxrwxrwx 1 root root 0 Nov 22 20:29 loop4 -> ../devices/virtual/block/loop4/ lrwxrwxrwx 1 root root 0 Nov 22 20:29 loop5 -> ../devices/virtual/block/loop5/ lrwxrwxrwx 1 root root 0 Nov 22 20:29 loop6 -> ../devices/virtual/block/loop6/ lrwxrwxrwx 1 root root 0 Nov 22 20:29 loop7 -> ../devices/virtual/block/loop7/ lrwxrwxrwx 1 root root 0 Nov 22 20:29 md1 -> ../devices/virtual/block/md1/ lrwxrwxrwx 1 root root 0 Nov 22 20:29 md2 -> ../devices/virtual/block/md2/ lrwxrwxrwx 1 root root 0 Nov 22 20:29 md3 -> ../devices/virtual/block/md3/ lrwxrwxrwx 1 root root 0 Nov 22 20:29 md4 -> ../devices/virtual/block/md4/ lrwxrwxrwx 1 root root 0 Nov 22 20:29 md5 -> ../devices/virtual/block/md5/ lrwxrwxrwx 1 root root 0 Nov 22 20:29 md6 -> ../devices/virtual/block/md6/ lrwxrwxrwx 1 root root 0 Nov 22 20:24 sda -> ../devices/pci0000:00/0000:00:1c.0/0000:02:00.0/host1/target1:0:0/1:0:0:0/block/sda/ lrwxrwxrwx 1 root root 0 Nov 22 20:29 sdb -> ../devices/pci0000:00/0000:00:1d.7/usb2/2-6/2-6:1.0/host2/target2:0:0/2:0:0:0/block/sdb/ lrwxrwxrwx 1 root root 0 Nov 22 20:17 sdc -> ../devices/pci0000:00/0000:00:1f.2/host3/target3:0:0/3:0:0:0/block/sdc/ lrwxrwxrwx 1 root root 0 Nov 22 20:21 sdd -> ../devices/pci0000:00/0000:00:1f.2/host4/target4:0:0/4:0:0:0/block/sdd/ lrwxrwxrwx 1 root root 0 Nov 22 20:28 sde -> ../devices/pci0000:00/0000:00:1f.2/host5/target5:0:0/5:0:0:0/block/sde/ lrwxrwxrwx 1 root root 0 Nov 22 20:04 sdf -> ../devices/pci0000:00/0000:00:1f.2/host6/target6:0:0/6:0:0:0/block/sdf/ lrwxrwxrwx 1 root root 0 Nov 22 20:15 sdg -> ../devices/pci0000:00/0000:00:1f.2/host7/target7:0:0/7:0:0:0/block/sdg/ lrwxrwxrwx 1 root root 0 Nov 22 20:09 sdh -> ../devices/pci0000:00/0000:00:1f.2/host8/target8:0:0/8:0:0:0/block/sdh/ Quote Link to comment
wholly Posted November 23, 2009 Share Posted November 23, 2009 Just added my cache disk last night thanks to the new feature in Plus and it seems to be working pretty well. Except now that my cache drive has caught up I seem to have a few strange and erroneous messages in my syslog: Nov 22 22:03:58 NAS2 shfs: cache disk full Nov 22 22:08:09 NAS2 shfs: cache disk full Nov 22 22:09:11 NAS2 shfs: cache disk full Nov 22 22:10:10 NAS2 emhttp: shcmd (42): /usr/sbin/hdparm -y /dev/sdg >/dev/null Nov 22 22:13:37 NAS2 shfs: cache disk full Nov 22 22:16:04 NAS2 shfs: cache disk full Nov 22 22:17:15 NAS2 shfs: cache disk full This is when my cache disk is completely empty (well 33meg of 120G empty) I am doing moves on a user share between two folders when this message is displayed. Rob Quote Link to comment
BLKMGK Posted November 23, 2009 Share Posted November 23, 2009 My second server. This one is ALL SATA except the cache drive. I do not know the hardware in this one - it's been together a long time too! Has a 4port Promise card in it on PCI, the other drives all come off the MOBO. Not sure more data is really beneficial at this point but what the heck. Since this thing serves video and music only the files are generally just coming from one drive anyway. I HAVE noticed more drops than I used to see but I suspect that's XBMC and nothing new with unRAID. <shrug> parity device: pci-0000:00:1f.2-scsi-0:0:0:0 (sda) ata-ST31500341AS_9VS0575B disk1 device: pci-0000:00:1f.2-scsi-1:0:0:0 (sdb) ata-WDC_WD10EACS-00ZJB0_WD-WCASJ0346500 disk2 device: pci-0000:00:1f.2-scsi-4:0:0:0 (sde) ata-ST3750640AS_5QD05DSM disk3 device: pci-0000:00:1f.2-scsi-5:0:0:0 (sdg) ata-ST3750640AS_5QD023CC disk4 device: pci-0000:00:1f.2-scsi-2:0:0:0 (sdc) ata-ST3750640AS_5QD2JLFQ disk5 device: pci-0000:00:1f.2-scsi-3:0:0:0 (sdd) ata-ST31500341AS_9VS0JY4Y disk6 device: pci-0000:04:03.0-scsi-0:0:0:0 (sdh) ata-WDC_WD10EACS-00ZJB0_WD-WCASJ0347187 disk7 device: pci-0000:04:03.0-scsi-1:0:0:0 (sdi) ata-ST3750640AS_3QD0PGBL disk8 device: pci-0000:04:03.0-scsi-2:0:0:0 (sdj) ata-ST3750330AS_3QK034VL disk9 device: pci-0000:04:03.0-scsi-3:0:0:0 (sdk) ata-ST31500341AS_9VS0SK95 disk10 device: unassigned <snip> disk19 device: unassigned cache device: pci-0000:02:00.1-ide-0:1 (hdb) ata-ST3400620A_5QG02J9B and -bash: root@VOID:~#: command not found root@VOID:~# v /sys/block total 0 lrwxrwxrwx 1 root root 0 Nov 23 00:31 hda -> ../devices/pci0000:00/0000:00:1c.4/0000:02:00.1/ide0/0.0/block/hda/ lrwxrwxrwx 1 root root 0 Nov 22 19:41 hdb -> ../devices/pci0000:00/0000:00:1c.4/0000:02:00.1/ide0/0.1/block/hdb/ lrwxrwxrwx 1 root root 0 Nov 23 00:32 loop0 -> ../devices/virtual/block/loop0/ lrwxrwxrwx 1 root root 0 Nov 23 00:32 loop1 -> ../devices/virtual/block/loop1/ lrwxrwxrwx 1 root root 0 Nov 23 00:32 loop2 -> ../devices/virtual/block/loop2/ lrwxrwxrwx 1 root root 0 Nov 23 00:32 loop3 -> ../devices/virtual/block/loop3/ lrwxrwxrwx 1 root root 0 Nov 23 00:32 loop4 -> ../devices/virtual/block/loop4/ lrwxrwxrwx 1 root root 0 Nov 23 00:32 loop5 -> ../devices/virtual/block/loop5/ lrwxrwxrwx 1 root root 0 Nov 23 00:32 loop6 -> ../devices/virtual/block/loop6/ lrwxrwxrwx 1 root root 0 Nov 23 00:32 loop7 -> ../devices/virtual/block/loop7/ lrwxrwxrwx 1 root root 0 Nov 23 00:32 md1 -> ../devices/virtual/block/md1/ lrwxrwxrwx 1 root root 0 Nov 23 00:32 md2 -> ../devices/virtual/block/md2/ lrwxrwxrwx 1 root root 0 Nov 23 00:32 md3 -> ../devices/virtual/block/md3/ lrwxrwxrwx 1 root root 0 Nov 23 00:32 md4 -> ../devices/virtual/block/md4/ lrwxrwxrwx 1 root root 0 Nov 23 00:32 md5 -> ../devices/virtual/block/md5/ lrwxrwxrwx 1 root root 0 Nov 23 00:32 md6 -> ../devices/virtual/block/md6/ lrwxrwxrwx 1 root root 0 Nov 23 00:32 md7 -> ../devices/virtual/block/md7/ lrwxrwxrwx 1 root root 0 Nov 23 00:32 md8 -> ../devices/virtual/block/md8/ lrwxrwxrwx 1 root root 0 Nov 23 00:32 md9 -> ../devices/virtual/block/md9/ lrwxrwxrwx 1 root root 0 Nov 22 19:10 sda -> ../devices/pci0000:00/0000:00:1f.2/host1/target1:0:0/1:0:0:0/block/sda/ lrwxrwxrwx 1 root root 0 Nov 22 19:44 sdb -> ../devices/pci0000:00/0000:00:1f.2/host2/target2:0:0/2:0:0:0/block/sdb/ lrwxrwxrwx 1 root root 0 Nov 22 19:11 sdc -> ../devices/pci0000:00/0000:00:1f.2/host3/target3:0:0/3:0:0:0/block/sdc/ lrwxrwxrwx 1 root root 0 Nov 22 19:14 sdd -> ../devices/pci0000:00/0000:00:1f.2/host4/target4:0:0/4:0:0:0/block/sdd/ lrwxrwxrwx 1 root root 0 Nov 22 19:49 sde -> ../devices/pci0000:00/0000:00:1f.2/host5/target5:0:0/5:0:0:0/block/sde/ lrwxrwxrwx 1 root root 0 Nov 23 00:31 sdf -> ../devices/pci0000:00/0000:00:1a.7/usb1/1-3/1-3:1.0/host0/target0:0:0/0:0:0:0/block/sdf/ lrwxrwxrwx 1 root root 0 Nov 22 23:49 sdg -> ../devices/pci0000:00/0000:00:1f.2/host6/target6:0:0/6:0:0:0/block/sdg/ lrwxrwxrwx 1 root root 0 Nov 22 19:43 sdh -> ../devices/pci0000:00/0000:00:1e.0/0000:04:03.0/host9/target9:0:0/9:0:0:0/block/sdh/ lrwxrwxrwx 1 root root 0 Nov 22 19:46 sdi -> ../devices/pci0000:00/0000:00:1e.0/0000:04:03.0/host10/target10:0:0/10:0:0:0/block/sdi/ lrwxrwxrwx 1 root root 0 Nov 22 19:50 sdj -> ../devices/pci0000:00/0000:00:1e.0/0000:04:03.0/host11/target11:0:0/11:0:0:0/block/sdj/ lrwxrwxrwx 1 root root 0 Nov 22 19:09 sdk -> ../devices/pci0000:00/0000:00:1e.0/0000:04:03.0/host12/target12:0:0/12:0:0:0/block/sdk/ This one is still on beta8. I will fix that tonight! I tend to put the more cutting edge stuff on the other server since this server has data I'm most concerned with. Quote Link to comment
BLKMGK Posted November 24, 2009 Share Posted November 24, 2009 Alright stripped out old drives. Parity check is SCREAMING with just a few SATA in it. parity device: pci-0000:00:1f.2-scsi-0:0:0:0 (sda) ata-ST31500341AS_9VS0SMTZ disk1 device: pci-0000:00:1f.2-scsi-1:0:0:0 (sdc) ata-ST31500341AS_9VS0KRYR disk2 device: pci-0000:00:1f.2-scsi-0:0:1:0 (sdb) ata-ST31500341AS_9VS0T46P disk3 device: pci-0000:00:1f.2-scsi-1:0:1:0 (sdd) ata-ST31500341AS_9VS0X1T5 disk4 device: pci-0000:00:1f.5-scsi-0:0:0:0 (sde) ata-WDC_WD15EADS-00P8B0_WD-WMAVU0433335 I'm now out of SATA ports so if there are add-in cards that do not cause this pause issue I'm all ears! Also, I see no AHCI listed in this BIOS but I did see refrence to an "enhanced" mode I will look for a BIOS update but I won't hold my breathe. Quote Link to comment
limetech Posted November 24, 2009 Author Share Posted November 24, 2009 Alright stripped out old drives. Parity check is SCREAMING with just a few SATA in it. parity device: pci-0000:00:1f.2-scsi-0:0:0:0 (sda) ata-ST31500341AS_9VS0SMTZ disk1 device: pci-0000:00:1f.2-scsi-1:0:0:0 (sdc) ata-ST31500341AS_9VS0KRYR disk2 device: pci-0000:00:1f.2-scsi-0:0:1:0 (sdb) ata-ST31500341AS_9VS0T46P disk3 device: pci-0000:00:1f.2-scsi-1:0:1:0 (sdd) ata-ST31500341AS_9VS0X1T5 disk4 device: pci-0000:00:1f.5-scsi-0:0:0:0 (sde) ata-WDC_WD15EADS-00P8B0_WD-WMAVU0433335 I'm now out of SATA ports so if there are add-in cards that do not cause this pause issue I'm all ears! Also, I see no AHCI listed in this BIOS but I did see refrence to an "enhanced" mode I will look for a BIOS update but I won't hold my breathe. Pretty sure the Adaptec 1430SA and 1220SA don't have this "problem". AHCI will make no difference, btw. Code is all in place to support 'spinup groups'. Should be able to post very soon. Quote Link to comment
Joe L. Posted November 24, 2009 Share Posted November 24, 2009 Code is all in place to support 'spinup groups'. Should be able to post very soon. Will spin-up-groups be able to handle more than 2 drives in a given group? (Can I put all my drives with "Movies" in a single group, so they all spin up together if any is accessed, all with "Music" in a second group, etc.) Quote Link to comment
sosdk Posted November 24, 2009 Share Posted November 24, 2009 I also have the spin up problem parity device: pci-0000:00:1f.2-scsi-0:0:0:0 (sda) ata-SAMSUNG_HD103UJ_S13PJ1EQ504937 disk1 device: pci-0000:00:1f.2-scsi-1:0:0:0 (sdb) ata-SAMSUNG_HD103UJ_S13PJ1EQ504936 disk2 device: pci-0000:00:1f.2-scsi-2:0:0:0 (sdc) ata-SAMSUNG_HD103UJ_S13PJ1BQA07919 disk3 device: pci-0000:00:1f.2-scsi-4:0:0:0 (sde) ata-SAMSUNG_HD103UJ_S13PJ1EQ504935 disk4 device: pci-0000:00:1f.2-scsi-5:0:0:0 (sdf) ata-SAMSUNG_HD103UJ_S13PJ90Q909841 disk5 device: pci-0000:00:1f.2-scsi-3:0:0:0 (sdd) ata-WDC_WD10EADS-00M2B0_WD-WCAV51184422 From what Tom said earlier you should not be having this problem, if I understood him correctly. Please provide a little more details for us and Tom. Your Motherboard, amount of RAM, version of unRAID (I assume you are running the current beta since this is the beta thread), and anything else you can think of. Sorry for the late reply. I'm running 4.5 beta 11 and I know that from what Tom wrote I shouldnt have the problem but I have. I've foiund 2 scenarios where it happens everytime I try. 1: When all drives are spin down besides the one I play any media music/movie from and I delete or start a copy that have to spin up data and parity drive . 2: When playing a HD movie from 1 disk and I start another HD movie from another disk. Hardware is Asus board Limetec used to use, 1GB ram and Celeron 430. Quote Link to comment
Joe L. Posted November 24, 2009 Share Posted November 24, 2009 The release notes for beta-8/9/11 state: - Upgrade pkgtools in order to support slackware packages which have .txz extension. To use that capability, the "xz" compression utility needs to be included as well. Currently, in beta-11, it is not. I'm guessing this is an oversight. Joe L. Quote Link to comment
limetech Posted November 25, 2009 Author Share Posted November 25, 2009 Code is all in place to support 'spinup groups'. Should be able to post very soon. Will spin-up-groups be able to handle more than 2 drives in a given group? (Can I put all my drives with "Movies" in a single group, so they all spin up together if any is accessed, all with "Music" in a second group, etc.) Yes, and you can have the same drive in multiple groups. Quote Link to comment
limetech Posted November 25, 2009 Author Share Posted November 25, 2009 I also have the spin up problem parity device: pci-0000:00:1f.2-scsi-0:0:0:0 (sda) ata-SAMSUNG_HD103UJ_S13PJ1EQ504937 disk1 device: pci-0000:00:1f.2-scsi-1:0:0:0 (sdb) ata-SAMSUNG_HD103UJ_S13PJ1EQ504936 disk2 device: pci-0000:00:1f.2-scsi-2:0:0:0 (sdc) ata-SAMSUNG_HD103UJ_S13PJ1BQA07919 disk3 device: pci-0000:00:1f.2-scsi-4:0:0:0 (sde) ata-SAMSUNG_HD103UJ_S13PJ1EQ504935 disk4 device: pci-0000:00:1f.2-scsi-5:0:0:0 (sdf) ata-SAMSUNG_HD103UJ_S13PJ90Q909841 disk5 device: pci-0000:00:1f.2-scsi-3:0:0:0 (sdd) ata-WDC_WD10EADS-00M2B0_WD-WCAV51184422 From what Tom said earlier you should not be having this problem, if I understood him correctly. Please provide a little more details for us and Tom. Your Motherboard, amount of RAM, version of unRAID (I assume you are running the current beta since this is the beta thread), and anything else you can think of. Sorry for the late reply. I'm running 4.5 beta 11 and I know that from what Tom wrote I shouldnt have the problem but I have. I've foiund 2 scenarios where it happens everytime I try. 1: When all drives are spin down besides the one I play any media music/movie from and I delete or start a copy that have to spin up data and parity drive . 2: When playing a HD movie from 1 disk and I start another HD movie from another disk. Hardware is Asus board Limetec used to use, 1GB ram and Celeron 430. I think the way I outlined "how to tell" if multiple drives are on the same controller is not reliable. Please post the output of this command: v /sys/bloock Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.