February 10, 201115 yr I don't know if this belongs here or somewhere else or if it's something specific too the Beta or not. I am now running 5.0 b4 as of a few minutes ago...had been Running b2 since it was released... I am using Cache and Disk Shares. There were a bunch of files sitting on the cache disk for weeks and weeks that the Mover wasn't moving so I manually moved them to a disk with the most space. Now all of the files and folders in the Shares is 0 Byte, but i can still see and access the files directly from the disk share... I am getting lots of these types of message on the logs... Feb 8 21:30:34 Media shfs: duplicate object: /mnt/disk4/TV/TVShowxxx/Sxx/Exx/<show name>.xxx Did I screw it up by manually moving these files, and how to make things right? Sw2
February 10, 201115 yr I don't know if this belongs here or somewhere else or if it's something specific too the Beta or not. I am now running 5.0 b4 as of a few minutes ago...had been Running b2 since it was released... I am using Cache and Disk Shares. There were a bunch of files sitting on the cache disk for weeks and weeks that the Mover wasn't moving so I manually moved them to a disk with the most space. Now all of the files and folders in the Shares is 0 Byte, but i can still see and access the files directly from the disk share... I am getting lots of these types of message on the logs... Feb 8 21:30:34 Media shfs: duplicate object: /mnt/disk4/TV/TVShowxxx/Sxx/Exx/<show name>.xxx Did I screw it up by manually moving these files, and how to make things right? Sw2 You did nothing wrong. While the file was being moved it was on two different disks. Hopefully the message have stopped now the file is moved. Joe L.
February 10, 201115 yr hi i have same issue when the mover wants to move something from within the protected array to another share in the protected array nothing wrong with it .... jut ignore as long as it comes along once
February 10, 201115 yr Author I don't know if this belongs here or somewhere else or if it's something specific too the Beta or not. I am now running 5.0 b4 as of a few minutes ago...had been Running b2 since it was released... I am using Cache and Disk Shares. There were a bunch of files sitting on the cache disk for weeks and weeks that the Mover wasn't moving so I manually moved them to a disk with the most space. Now all of the files and folders in the Shares is 0 Byte, but i can still see and access the files directly from the disk share... I am getting lots of these types of message on the logs... Feb 8 21:30:34 Media shfs: duplicate object: /mnt/disk4/TV/TVShowxxx/Sxx/Exx/<show name>.xxx Did I screw it up by manually moving these files, and how to make things right? There are MANY files/folders like this (everything that I manually moved)...here is one example. Sw2 You did nothing wrong. While the file was being moved it was on two different disks. Hopefully the message have stopped now the file is moved. Joe L. Negative... I only see the file on a single Disk Share (Disk 4 in the case of the one I copied), but all of those files I manually moved (when the Mover wasn't moving them) show up as Zero Byte files on my User Share... (Which Includes Disk 1-16, as do all my User Shares) I noticed this behavior days after I manually moved the files, not during the move process itself. Files in the User Share Same files on the Disk Share
February 10, 201115 yr I suspect the mover copied the files to the array but didn't delete the cache drive versions. So, you then moved them again and created duplicates on multiple disks. So, look for duplicate files on different disks. I'm not sure why the size would show as 0 though. Peter
February 10, 201115 yr Author A quick glance showed some Zero Byte files on a different disk with the same name... Possible that this is where the mover first tried and failed to move them too before I manually did it. I will search all the other disks for Zero Byte files too... I think I'm going to disable the mover process if it's going to mess things up.
February 10, 201115 yr A quick glance showed some Zero Byte files on a different disk with the same name... Possible that this is where the mover first tried and failed to move them too before I manually did it. I will search all the other disks for Zero Byte files too... I think I'm going to disable the mover process if it's going to mess things up. It should not have messed anything up, and not many (if any) have reported something like this happening. Something else must have happened to cause this.
February 10, 201115 yr Author There are hundreds of zero byte files on nearly everyone of my Disk, some which have not been accessed, touched or changed in years.
February 10, 201115 yr Hum... Are you using a plain vanilla install of unRAID or are you running some other addon's? If running addon's I suggest disabling them, clearing the system of the 0 byte files, and then running plain unRAID for a while. See if this issue stops happening. A full syslog might also help shed some light on what might be happening.
February 10, 201115 yr Time to check your filesystems for corruption? http://lime-technology.com/wiki/index.php?title=Check_Disk_Filesystems Peter
February 10, 201115 yr Author Hum... Are you using a plain vanilla install of unRAID or are you running some other addon's? If running addon's I suggest disabling them, clearing the system of the 0 byte files, and then running plain unRAID for a while. See if this issue stops happening. A full syslog might also help shed some light on what might be happening. I had been running unmenu, but honestly haven't touched it in a while, so it probably can go away....I don't think I'm doing any other add-on (or if I am, I don't know how to check) Here is my "Go" file from my flash disk #!/bin/bash # Start the Management Utility /usr/local/sbin/emhttp & cd /boot/llink echo "./llink" | at now + 1 minute sleep 30 for i in /dev/md* do blockdev --setra 2048 $i done # Start up unmenu /boot/unmenu/uu cd /boot/packages && find . -name '*.auto_install' -type f -print | sort | xargs -n1 sh -c
February 10, 201115 yr Here is my "Go" file from my flash disk #!/bin/bash # Start the Management Utility /usr/local/sbin/emhttp & cd /boot/llink echo "./llink" | at now + 1 minute sleep 30 for i in /dev/md* do blockdev --setra 2048 $i done # Start up unmenu /boot/unmenu/uu cd /boot/packages && find . -name '*.auto_install' -type f -print | sort | xargs -n1 sh -c The short answer is you need to disable everything in your go script except for the top 3 lines. llink may be the application that is causing the problems you are seeing. Disabling it and running plain vanilla unRAID for a little while may prove that to be the case. Please disable the addons and then reboot the server.
February 10, 201115 yr Author Here is my "Go" file from my flash disk #!/bin/bash # Start the Management Utility /usr/local/sbin/emhttp & cd /boot/llink echo "./llink" | at now + 1 minute sleep 30 for i in /dev/md* do blockdev --setra 2048 $i done # Start up unmenu /boot/unmenu/uu cd /boot/packages && find . -name '*.auto_install' -type f -print | sort | xargs -n1 sh -c The short answer is you need to disable everything in your go script except for the top 3 lines. llink may be the application that is causing the problems you are seeing. Disabling it and running plain vanilla unRAID for a little while may prove that to be the case. Please disable the addons and then reboot the server. Done...I don't even remember or know what llink is. I've also removed all of the Zero Byte Files from all Disk Shares On reboot I had to redo the "New Slots" thing again to get all the drives Green Again and Array started.. On my User Shares, 2 have Red Circle instead of Green. Is there significance to that *I think it was like that before
February 10, 201115 yr Those are orange, and I think that means there is content on the cache drive that has not been moved to the protected array.
February 10, 201115 yr Author Those are orange, and I think that means there is content on the cache drive that has not been moved to the protected array. True Statement... I'm manually launching the mover now to see what happens now that I"ve removed the other stuff from the Go file and restarted. Some immediate weirdness with the mover, see attached TXT.. I'll add screen shots of my Disks and User Share settings and full syslog as soon as I get wife off of PC at home. mover.txt
February 10, 201115 yr You have a disk acting up on you. (can't tell which one since you did not post a full syslog, even then it is not easy) Feb 10 12:54:20 Media kernel: ata15.00: irq_stat 0x08000000 Feb 10 12:54:20 Media kernel: ata15: SError: { UnrecovData 10B8B Dispar BadCRC Handshk } Feb 10 12:54:20 Media kernel: ata15.00: failed command: READ FPDMA QUEUED Feb 10 12:54:20 Media kernel: ata15.00: cmd 60/00:08:47:44:e0/01:00:29:00:00/40 tag 1 ncq 131072 in Feb 10 12:54:20 Media kernel: res 40/00:08:47:44:e0/00:00:29:00:00/40 Emask 0x10 (ATA bus error) Feb 10 12:54:20 Media kernel: ata15.00: status: { DRDY } Feb 10 12:54:20 Media kernel: ata15: hard resetting link Feb 10 12:54:20 Media kernel: ata15: SATA link up 3.0 Gbps (SStatus 123 SControl 300) Feb 10 12:54:20 Media kernel: ata15.00: configured for UDMA/100 Feb 10 12:54:20 Media kernel: ata15: EH complete Typically, a bad CRC in communicating with the disk is a bad SATA cable, or bad MEMORY, or a noisy power supply. Then, you probably do not have the min-free space setting on the user-share set, so "mover" is ending up running out of space. Feb 10 12:55:26 Media shfs: shfs_write: write: (28) No space left on device Joe L.
February 10, 201115 yr Author Here are Screenshots of my Disks/Shares but can't post syslog since its too big for the Forum SW http://img64.imageshack.us/img64/8478/mainew.jpg[/img] http://img193.imageshack.us/img193/8238/movieshare.jpg[/img] http://img39.imageshack.us/img39/1534/tvshare.jpg[/img] MOD EDIT: shrunk the pictures slightly to make it easier to view
February 10, 201115 yr Author zip the syslog and then attach it. it will compress very well typically. Thanks for the Pic Resize, sorry... Trying to post ziped syslog to this reply Syslog.zip
February 11, 201115 yr ATA15: Feb 10 12:41:53 Media kernel: ata15.00: ATA-7: ST3360320AS, 3.CHL, max UDMA/100 Feb 10 12:41:53 Media kernel: ata15.00: 703282608 sectors, multi 0: LBA48 NCQ (depth 31/32) Feb 10 12:41:53 Media emhttp: ST3360320AS_5QF4WTR4 (sdp) 351641272 => pci0000:00/0000:00:1c.4/0000:02:00.0/host15/target15:0:0/15:0:0:0 The disk with the issues seems to be your cache drive. Try a new cable first, then go from there. Do NOT bundle its SATA cable with any power cable, to preclude its pickup of any induced noise. Joe L.
February 11, 201115 yr Author ATA15: Feb 10 12:41:53 Media kernel: ata15.00: ATA-7: ST3360320AS, 3.CHL, max UDMA/100 Feb 10 12:41:53 Media kernel: ata15.00: 703282608 sectors, multi 0: LBA48 NCQ (depth 31/32) Feb 10 12:41:53 Media emhttp: ST3360320AS_5QF4WTR4 (sdp) 351641272 => pci0000:00/0000:00:1c.4/0000:02:00.0/host15/target15:0:0/15:0:0:0 The disk with the issues seems to be your cache drive. Try a new cable first, then go from there. Do NOT bundle its SATA cable with any power cable, to preclude its pickup of any induced noise. Joe L. My Cache drive is ESATA with External enclosure, and likely has been dropped at one time or another...easily replaced from the pile of other smaller (<500GB) drives. Working for a HD company has at least some perks, just wish they had a better employee purchase program.
February 11, 201115 yr ATA15: Feb 10 12:41:53 Media kernel: ata15.00: ATA-7: ST3360320AS, 3.CHL, max UDMA/100 Feb 10 12:41:53 Media kernel: ata15.00: 703282608 sectors, multi 0: LBA48 NCQ (depth 31/32) Feb 10 12:41:53 Media emhttp: ST3360320AS_5QF4WTR4 (sdp) 351641272 => pci0000:00/0000:00:1c.4/0000:02:00.0/host15/target15:0:0/15:0:0:0 The disk with the issues seems to be your cache drive. Try a new cable first, then go from there. Do NOT bundle its SATA cable with any power cable, to preclude its pickup of any induced noise. Joe L. My Cache drive is ESATA with External enclosure, and likely has been dropped at one time or another...easily replaced from the pile of other smaller (<500GB) drives. Working for a HD company has at least some perks, just wish they had a better employee purchase program. The issue is not likely the drive itself (although I suppose it could be), but the SATA link to the drive. Are you using a shielded SATA cable designed for an external drive enclosure? The serial communications from the drive to the disk controller on the SATA cable is having CRC errors.. Joe L.
February 11, 201115 yr Author ATA15: Feb 10 12:41:53 Media kernel: ata15.00: ATA-7: ST3360320AS, 3.CHL, max UDMA/100 Feb 10 12:41:53 Media kernel: ata15.00: 703282608 sectors, multi 0: LBA48 NCQ (depth 31/32) Feb 10 12:41:53 Media emhttp: ST3360320AS_5QF4WTR4 (sdp) 351641272 => pci0000:00/0000:00:1c.4/0000:02:00.0/host15/target15:0:0/15:0:0:0 The disk with the issues seems to be your cache drive. Try a new cable first, then go from there. Do NOT bundle its SATA cable with any power cable, to preclude its pickup of any induced noise. Joe L. My Cache drive is ESATA with External enclosure, and likely has been dropped at one time or another...easily replaced from the pile of other smaller (<500GB) drives. Working for a HD company has at least some perks, just wish they had a better employee purchase program. The issue is not likely the drive itself (although I suppose it could be), but the SATA link to the drive. Are you using a shielded SATA cable designed for an external drive enclosure? The serial communications from the drive to the disk controller on the SATA cable is having CRC errors.. Joe L. Yes It's a shielded eSATA cable... I have a few spares, I will also check the connection form the eSATA PCI port to the MB, I'm not 100% sure how it's wired. I could easily eliminate the Cache drive all together without too much pain.
February 12, 201115 yr Author Ok, this might be a separate issue, but I cannot understand why UnRaid thinks I'm out of space when trying to write to one of my User Shares... Feb 11 20:37:08 Media shfs: shfs_write: write: (28) No space left on device I have hundreds of GB's free (I just added another 1TB), have my User Shares covering ALL 17 of of my Disk, Have tried multiple Allocation Methods (currently at "Most Free with split level of 2). But I'm still getting "Disk Full" Errors and Zero Byte files being generated because the array thinks its full... I also seem to have quite a few DUPE files all over the place for some reason too. I Current have the following structure on my Array... is there some other setting I should have on my allocation to prevent the User Shares from thinking they are full? Movie\MovieName\Movie.avi or VIDEO_TS TV\TVShowName\Sxx\Exx\Episode.AVI/MKV, etc BACKUP\Multiple SubFolders Games\Multiple Subfolders Music\Genre\Artist\Album\Songs.MP3
February 12, 201115 yr Probably your split level is too low. You will likely have to do some investigating and delete any empty folders on your disks. Read the info from this link down and see if it helps; http://lime-technology.com/wiki/index.php?title=Un-Official_UnRAID_Manual#Allocation_method Peter
Archived
This topic is now archived and is closed to further replies.