dlandon Posted January 8, 2015 Share Posted January 8, 2015 Adding things in the go file is not recommended, but at times there is no other way. Fromdos takes the carriage return characters off the file and leaves the line feed characters so the file ends up being in Linux format. It's the same as a cp, but insures that the file is in Linux line feed format. A lot of problems people run into is from improper Linux file formatting because they used a non-Linux compatible editor. None of the file formats used in unraid implements the 'discard' option when mounting a drive. There is an open feature request to add trim support to SSD cache drives. If you are looking for more information on how trim works, and its impact on Linux, do a Google search. It is off topic in this forum post. Quote Link to comment
luca2 Posted January 8, 2015 Share Posted January 8, 2015 Hi Unraid, Can you pls confirm if on XFS formatted disks within v6 trim support is enabled by default? Also, if any other special features should be considered regarding trim (i.e. trim not running if snapshots present). I post this here since I could not found a specific XFS thread like the one for BTRFS. Quote Link to comment
mostlydave Posted January 8, 2015 Share Posted January 8, 2015 here is an option: create a cache_backup.sh file in /boot/custom folder and make it executable. cache_backup.sh contents. #!/bin/bash rsync -av /mnt/docker/appdata/xxxxxx /mnt/user/docker_backup/ in the go file add: cp /boot/custom/cache_backup.sh /etc/cron.weekly replace "cron.weekly" with "cron.hourly", "cron.daily" or "cron.monthly" This is really cool, might have to give this a go later when I have time to mess around. Thanks! I still stand in the camp that these kind of forum based knowledge and script answers aren't very user friendly. Frankly I never even thought about using rsync to back up my dockers/ and app data until now. If these was a built in feature and part of the webgui, I would have been all over this on day one... I fully agree here, I tried the setup above and it seems to be working, but I have no idea if there are errors or any indication that it is even running as scheduled. If this could be built into the webgui with a summary on the dashboard that would be the ideal docker backup solution in my opinion. Quote Link to comment
Berg Posted January 8, 2015 Share Posted January 8, 2015 Can xfs be used on the ssd for a cache with TRIM support? Yes. There is no longer a requirement for a BTRFS drive for Docker since the Docker "image" is BTRFS. I have been running an XFS cache for a while. Very useful conversation guys. Thank you. So it seems to me that the one of the only key differences between running XFS vs. BTRFS for a cache drive is that the only way you can pool cache drives is to use BTRFS. Given I way hoping to use two 250 Gb SSDs hard drives in a cache pool, so it seems like BTRFS is the way to go for me. Quote Link to comment
gundamguy Posted January 8, 2015 Share Posted January 8, 2015 here is an option: create a cache_backup.sh file in /boot/custom folder and make it executable. cache_backup.sh contents. #!/bin/bash rsync -av /mnt/docker/appdata/xxxxxx /mnt/user/docker_backup/ in the go file add: cp /boot/custom/cache_backup.sh /etc/cron.weekly replace "cron.weekly" with "cron.hourly", "cron.daily" or "cron.monthly" This is really cool, might have to give this a go later when I have time to mess around. Thanks! I still stand in the camp that these kind of forum based knowledge and script answers aren't very user friendly. Frankly I never even thought about using rsync to back up my dockers/ and app data until now. If these was a built in feature and part of the webgui, I would have been all over this on day one... I fully agree here, I tried the setup above and it seems to be working, but I have no idea if there are errors or any indication that it is even running as scheduled. If this could be built into the webgui with a summary on the dashboard that would be the ideal docker backup solution in my opinion. Doing a little reading on rsync, you could pass it the logfile paramater so that rsync creates a log for you to review. All you would have to do is add --log-file=/xxxx to your call of rsync. It should look like the following I believe. rsync -av --log-file=/xxxx /source /destination Where xxxx is the location you want the log created (Rsync's example is /tmp/rlog). /source is the source you want rsync to sync, and /destination is where you want the sync to go. Earlier the example was source "/mnt/docker/appdata/xxxxxx" destination "/mnt/user/docker_backup/" Quote Link to comment
dlandon Posted January 8, 2015 Share Posted January 8, 2015 Can xfs be used on the ssd for a cache with TRIM support? Yes. There is no longer a requirement for a BTRFS drive for Docker since the Docker "image" is BTRFS. I have been running an XFS cache for a while. Very useful conversation guys. Thank you. So it seems to me that the one of the only key differences between running XFS vs. BTRFS for a cache drive is that the only way you can pool cache drives is to use BTRFS. Given I way hoping to use two 250 Gb SSDs hard drives in a cache pool, so it seems like BTRFS is the way to go for me. While cache pooling is implemented in unraid, I think it not advisable to pool cache drives at this time. BTRFS is still in its infancy and it seems to be stable for single drive implementations, but may not be reliable for cache pooling. Quote Link to comment
luca2 Posted January 8, 2015 Share Posted January 8, 2015 I use a SSD drive formatted with XFS and I set up a daily cron to run trim. Hi again, In a setup where my ssd is mounted outside of the array to only store the vm´s images (in an SSD formatted as XFS), would you advise to run a daily cron to perform trim? I am thinking of my vm´s running and the cron running at the same time. Quote Link to comment
mostlydave Posted January 8, 2015 Share Posted January 8, 2015 here is an option: create a cache_backup.sh file in /boot/custom folder and make it executable. cache_backup.sh contents. #!/bin/bash rsync -av /mnt/docker/appdata/xxxxxx /mnt/user/docker_backup/ in the go file add: cp /boot/custom/cache_backup.sh /etc/cron.weekly replace "cron.weekly" with "cron.hourly", "cron.daily" or "cron.monthly" This is really cool, might have to give this a go later when I have time to mess around. Thanks! I still stand in the camp that these kind of forum based knowledge and script answers aren't very user friendly. Frankly I never even thought about using rsync to back up my dockers/ and app data until now. If these was a built in feature and part of the webgui, I would have been all over this on day one... I fully agree here, I tried the setup above and it seems to be working, but I have no idea if there are errors or any indication that it is even running as scheduled. If this could be built into the webgui with a summary on the dashboard that would be the ideal docker backup solution in my opinion. Doing a little reading on rsync, you could pass it the logfile paramater so that rsync creates a log for you to review. All you would have to do is add --log-file=/xxxx to your call of rsync. It should look like the following I believe. rsync -av --log-file=/xxxx /source /destination Where xxxx is the location you want the log created (Rsync's example is /tmp/rlog). /source is the source you want rsync to sync, and /destination is where you want the sync to go. Earlier the example was source "/mnt/docker/appdata/xxxxxx" destination "/mnt/user/docker_backup/" I appreciate the help, but I gave up to on the rsync solution, in my opinion it shouldn't be more complicated to setup and monitor the backup of the appdata folder than is is to setup docker or any of the apps I have installed with it. My time is valuable to me and while rsync may work perfectly fine for others I just don't have any confidence in this solution. I would really love to see this incorporated into gui and supported by LT. Until or in case that never happens I am planning on adding a second cache drive, that's not ideal but it's better that nothing for now. Quote Link to comment
dlandon Posted January 8, 2015 Share Posted January 8, 2015 I use a SSD drive formatted with XFS and I set up a daily cron to run trim. Hi again, In a setup where my ssd is mounted outside of the array to only store the vm´s images (in an SSD formatted as XFS), would you advise to run a daily cron to perform trim? I am thinking of my vm´s running and the cron running at the same time. The trim cron job would have no impact on the VMs. I run a daily trim on my cache drive with two Dockers and four (4) VMs. No impact at all. Quote Link to comment
luca2 Posted January 8, 2015 Share Posted January 8, 2015 Thx for replay. I will test it too. Quote Link to comment
gundamguy Posted January 9, 2015 Share Posted January 9, 2015 I use a SSD drive formatted with XFS and I set up a daily cron to run trim. Hi again, In a setup where my ssd is mounted outside of the array to only store the vm´s images (in an SSD formatted as XFS), would you advise to run a daily cron to perform trim? I am thinking of my vm´s running and the cron running at the same time. The trim cron job would have no impact on the VMs. I run a daily trim on my cache drive with two Dockers and four (4) VMs. No impact at all. I implemented this last night. Thought I'm not sure that this is doing anything when run on btrfs formated SSDs. I was testing my implementation so I ran the comand a few times. The first time it said it trimmed a large number of bytes, or bits, or whatever? The second time it said the same thing... so it seemed like fstrim might not have done anything? I did a bit of research and found that this was the expected result, but couldn't decipher if this means fstrim is working or not. In comparison if you fstrim an xfs files system twice the first run should give you a number of blocks or bits or bytes or whatever to be trimmed and the second should say 0. Is fstrim on btrfs just not a thing? Quote Link to comment
luca2 Posted January 9, 2015 Share Posted January 9, 2015 Pls, how do you check (command) for the results after running trim? Quote Link to comment
gundamguy Posted January 9, 2015 Share Posted January 9, 2015 Pls, how do you check (command) for the results after running trim? This is a good question. I used the syntax that dlandon wrote to create my script. fstrim -v /mnt/cache | logger The -v option makes fstrim verbose, and | logger I assume adds the output of fstrim to the system log. If you check the system log the verbose output from fstrim should show up. Now for my question: I'm not a linux expert so "| logger" was not somethign I had done before. Does adding "| logger" to the end of your command work the same for other commands? How do you know which commands "| logger" will work with? IE if I did an rsync -av (archive and verbose) command with | logger at the end would all the verbose output from rsync show up in the system log? Quote Link to comment
StevenD Posted January 9, 2015 Share Posted January 9, 2015 unfortunately, I cant get fstrim to run. root@nas:~# fstrim -v /mnt/cache fstrim: /mnt/cache: FITRIM ioctl failed: Operation not supported It appears that the drive supports TRIM: root@nas:~# hdparm -I /dev/sdb /dev/sdb: ATA device, with non-removable media Model Number: Crucial_CT512M550SSD1 Serial Number: 14230C370A19 Firmware Revision: MU01 Transport: Serial, ATA8-AST, SATA 1.0a, SATA II Extensions, SATA Rev 2.5, SATA Rev 2.6, SATA Rev 3.0 Standards: Used: unknown (minor revision code 0x0028) Supported: 9 8 7 6 5 Likely used: 9 Configuration: Logical max current cylinders 16383 16383 heads 16 16 sectors/track 63 63 -- CHS current addressable sectors: 16514064 LBA user addressable sectors: 268435455 LBA48 user addressable sectors: 1000215216 Logical Sector size: 512 bytes Physical Sector size: 4096 bytes Logical Sector-0 offset: 0 bytes device size with M = 1024*1024: 488386 MBytes device size with M = 1000*1000: 512110 MBytes (512 GB) cache/buffer size = unknown Form Factor: 2.5 inch Nominal Media Rotation Rate: Solid State Device Capabilities: LBA, IORDY(can be disabled) Queue depth: 32 Standby timer values: spec'd by Standard, with device specific minimum R/W multiple sector transfer: Max = 16 Current = 16 Advanced power management level: 254 DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 udma5 *udma6 Cycle time: min=120ns recommended=120ns PIO: pio0 pio1 pio2 pio3 pio4 Cycle time: no flow control=120ns IORDY flow control=120ns Commands/features: Enabled Supported: * SMART feature set Security Mode feature set * Power Management feature set * Write cache * Look-ahead * Host Protected Area feature set * WRITE_BUFFER command * READ_BUFFER command * NOP cmd * DOWNLOAD_MICROCODE * Advanced Power Management feature set SET_MAX security extension * 48-bit Address feature set * Device Configuration Overlay feature set * Mandatory FLUSH_CACHE * FLUSH_CACHE_EXT * SMART error logging * SMART self-test * General Purpose Logging feature set * WRITE_{DMA|MULTIPLE}_FUA_EXT * 64-bit World wide name * IDLE_IMMEDIATE with UNLOAD Write-Read-Verify feature set * WRITE_UNCORRECTABLE_EXT command * {READ,WRITE}_DMA_EXT_GPL commands * Segmented DOWNLOAD_MICROCODE * Gen1 signaling speed (1.5Gb/s) * Gen2 signaling speed (3.0Gb/s) * Gen3 signaling speed (6.0Gb/s) * Native Command Queueing (NCQ) * Phy event counters * NCQ priority information * unknown 76[15] * DMA Setup Auto-Activate optimization Device-initiated interface power management Asynchronous notification (eg. media change) * Software settings preservation unknown 78[8] * SMART Command Transport (SCT) feature set * SCT Write Same (AC2) * SCT Features Control (AC4) * SCT Data Tables (AC5) * reserved 69[4] * reserved 69[7] * Data Set Management TRIM supported (limit 8 blocks) * Deterministic read ZEROs after TRIM Security: Master password revision code = 65534 supported not enabled not locked not frozen not expired: security count supported: enhanced erase 2min for SECURITY ERASE UNIT. 2min for ENHANCED SECURITY ERASE UNIT. Logical Unit WWN Device Identifier: 500a07510c370a19 NAA : 5 IEEE OUI : 00a075 Unique ID : 10c370a19 Checksum: correct root@nas:~# Quote Link to comment
dlandon Posted January 9, 2015 Share Posted January 9, 2015 Let's move the trim discussion here: http://lime-technology.com/forum/index.php?topic=37596.msg347746#msg347746 It is way off topic for this beta release. Quote Link to comment
Edrikk Posted January 19, 2015 Share Posted January 19, 2015 Hi, Any sense of when the next beta build will be released? Given that it's been roughly 7 weeks since beta 12, wouldn't it be good to release a new build to: 1) Correct the issues that have been fixed [from those reported] for those using Beta 12 2) Level-set again to know what's still really an issue vs. not? (I'm new to Unraid, but I ask this question knowing/having read/understanding the the unraid beta cycles last a long time, and go a while between releases).... Thanks, Quote Link to comment
BRiT Posted January 19, 2015 Share Posted January 19, 2015 Any sense of when the next beta build will be released? "Soon" Quote Link to comment
Edrikk Posted January 20, 2015 Share Posted January 20, 2015 LOL! Yup... Well played. I also saw "this week" in a git comment for Dynamix (regarding the Unraid backend). That was about 3 weeks ago.. I've been holding my breath since... Quote Link to comment
SSD Posted January 20, 2015 Share Posted January 20, 2015 For most things, patience is a virtue. With unRAID, it's a necessity. Quote Link to comment
squirrellydw Posted January 20, 2015 Share Posted January 20, 2015 For most things, patience is a virtue. With unRAID, it's a necessity. But it's worth it Quote Link to comment
rick.p Posted January 21, 2015 Share Posted January 21, 2015 ..... wouldn't it be good to release a new build to: 1) Correct the issues that have been fixed [from those reported] for those using Beta 12 2) Level-set again to know what's still really an issue vs. not? This. Quote Link to comment
PeterB Posted January 21, 2015 Share Posted January 21, 2015 For most things, patience is a virtue. With unRAID, it's a necessity. Indeed! If I'm not mistaken, this is the 2014 Q3 release that we're waiting for. Anyone who was holding their breath will have expired by now. Quote Link to comment
dalben Posted January 21, 2015 Share Posted January 21, 2015 Speaking words of wisdom, let it be Quote Link to comment
BRiT Posted January 21, 2015 Share Posted January 21, 2015 Speaking words of wisdom, let it be Perhaps you mean: Quote Link to comment
lars Posted January 21, 2015 Share Posted January 21, 2015 yep well, maybe it would be not a bad idea to have sth like a beta12b version. which actually just focuses on fixing the most common encountered bugs in beta12, instead of introducing new features which may work flawless or not. don't get me wrong, i am pretty happy with b12! only real issue i have is drives not spinning down when they should (i guess thats better than not spinning up, but still annoying). but there seem to be more probs for other users. so why not take a breath, stop introducing all kind of new and improved features, get a grip on the existing probs and try to solve them... than go with new steam on b13 (or rc1). i am easy going using only one plugin and one docker-app. i would think for others the issues are a lot worse. cheers, L 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.