jbartlett

Kicked off a copy of around 3 TB of files to a system with only half of that space available (didn't notice) and letting it run overnight. I woke up to find the system full, all the user shares gone, the user tab in the admin empty. A reboot brought the shares back but unfortunately I didn't think to copy the syslog off prior. Trying to duplicate, no luck yet.

7 Gaming VMs, one rig with UNRAID

OEM licenses don't like any changes to the motherboard which seems to include the CPU.

I'm trying to figure out the best way to back up files on one unraid server to another in a different geographical location, but I had the thought to see if anyone else has been down this route before and see if they're willing to share. The process should be automatic in that new files on Server A are sent to Server B but not a mirror in which files deleted aren't deleted on the remote server. Have you? Will you?

I have a script that does this if you want to utilize it.

Did it used to work? I just installed 6.1.4 on my backup server and it didn't have any issues. If you just downloaded it, your version may be incomplete. If you edit the file, it should be 818 lines long and end with "done".

Interesting. If you want to play around with it, try adding "-i 3" to the command line to test each sample point 3 times. In my mind, this will check to see if the problem is with the physical media or the controller board. Understandably, it will take three times longer to run. What's interesting is the two different reports you gave appear identical. I've never had two separate in-depth tests come back with matching numbers, there's always a slight variation somewhere.

Note that if you use MakeMKV, you're not recompressing the audio/video streams. It is a bit for bit copy of the video but in a different container format.

Pro tip: You can click on a drive in the legend to hide ones that push the other lines down.

For shitz-n-grins for maxing out a bus, I have four SDD's in a RAID0 on my new X99 PC build and the throughput caps out at 1.5 GByte/sec, which is just marginally better than three SDD's in a RAID0.

Try running the tunables script at http://lime-technology.com/forum/index.php?topic=29009.0

This ... is AMAZING

I'd try a manual scan and specify rages on the far edges.

What are the drives? I'd doubt the problem is with them since they're all scoring nearly the same assuming they did better previously.

Can you please run it with the following options, then zip up the file and email it to [email protected]? This will create the file diskspeed.log and only test your cache drive. NOTE: Please verify that your cache drive is sdk and alter the command below to accommodate (add sdk, remove current id) diskspeed.sh -l -x sdb,sdc,sdd,sde,sdf,sdg,sdh,sdi,sdj,sdl,sdm

One potential thing to add to the release notes: If you receive access denied errors on shares that have access rights enabled but you have access, reboot the Windows PC you received the access denied error on.

I'm contemplating creating a tool that looks for slowdowns in the drives performance which from another user on here indicates that it can be a sign of trouble even before the SMART data starts reporting issues. Two things are mostly needed, the ability to scan and time the read of specific areas of the hard drive (dd can do that) and knowing the physical geometry. Physical geometry is pretty much useless for accessing data since drives will silently remap bad spots on the drive but the timings to read a given area can indicate a remap took place. Having the physcal geometry would allow a heat map of the drive with specific platters given which would be more clear of problem areas on a platter. If a drive has four platters, it has 8 heads. It reads from head 1 to 8 in sequence before advancing the drive head. If any platter has a bad/remapped spot, the drive head must move to another location in order to read the remapped data which would affect the read time. Being able to read & time each track looking for slowdowns and then read each side on each platter spanning tracks would present a heat map for each side of each platter that would be more clear if areas of a disk are going bad vs one heat map over the entire drive. So my question - does anyone know of a way to programically determine the actual drive geometry or would it be simpler to rely on user input? Example of a heat map: http://www.highcharts.com/demo/heatmap-canvas

Drives I want to test have data on them, can’t run preclear, and I’m not sure it would give me the info I want anyway. In my experience once a drive starts getting this slower sectors it will have some bad sectors sooner rather than later, but until then S.M.A.R.T. looks fine without any errors, I’ve had a few of these older Samsung drives fail with read errors and when I use MHDD on them almost always have a lot of these slower areas. My plan is to use diskspeed as an early warning as I still have about 20 of these drives on 3 different servers and can’t afford to replace them all preemptively. What program is this? It's possible I can duplicate this functionality and display the data in a browser using a heat map.

I plan to go that route, once I figure out how to do it.

Version 2.5 Fixed computation for percentages less than 10% Reverted to 1 GB scans for better results but slower Added -f --fast to scan 200 MB instead of 1 GB, same as version 2.3 & 2.4

Thank you for letting me know, I'll take a look. This is fixed.

Thank you for letting me know, I'll take a look.

I can do this.

Version 2.4 released, download link in first post Change Log If the drive model is not able to be determined via fdisk, extract it form mdmcd Add -l --log option to create the debug log file speeddisk.log Modified to not display the MB sec in drive inventory report for excluded drives Modified to compute the drive capacity from the number of bytes UNRAID reports to support external drive cards. Added -g --graph option to display the drive by percentage comparison graph Added warning if files on the array are open which could mean drives are active Added drive spin up support by reading a random sector in the first quarter of the drive via background tasks and then performing a sync Upcoming version 3.0 Benchmark all drives at the same time at every 2% of the drive's capacity. Overall test duration will be equal to one pass over the slowest drive.

Looking at the log you sent previously, that's what the system is reporting via the dd command and not an issue with the script itself. But that's one of the big reasons for this script, to look for invisible things like that. I'd replace the SATA cable or swap the SATA plug with another device to see if the issue stays with the drive or migrates.