December 5, 201411 yr Hi, just upgraded from 10a to 12 without problems. I don't use plugins. Unraid here is a pure file server. I've a suggestion for the new interface. Is it possible to use more colors and not only percentages for USED disk space/utilisation in the dashboard and main tab ? For example 0-70% = green 70 - 90% = orange 90 - 100% = red Thanks and keep up the good work!
December 5, 201411 yr IMHO red should only ever be used for errors and a 90% full disk is not an error. e.g. to the casual observer the array below is "mostly red" but the reality is he has over 2TB of free space and is perfectly healthy
December 5, 201411 yr The use of colored bars is certainly possible, it should be introduced as an option though - not everybody would want/like it. Also the thresholds must be made programmable, to define your own level of "healthy". Attached a teaser in the black theme ... (used the same color codings as are used in the optional 'stats' plugin)
December 5, 201411 yr This is a nice idea, red seems OK to me for a nearly full drive but I understand your point nas
December 5, 201411 yr The colors and thresholds should be user selectable to be suitable for the color blind.
December 5, 201411 yr Make the colors configurable, example. This could be loaded into some array from an .ini file. Values > then a theshold number use the color until you run out of thresholds. 90% = #FF0000 80% = #FFFF00 This could go as far as Individual drive thresholds [disk1] 90% = #FF0000 80% = #FFFF00 [total] 90% = #FF0000 80% = #FFFF00 Point is, I don't really want to see RED on this screen unless I HAVE to react to something. Active system drives usually require a different threshold then archive drives. Your total array threshold may be different then the individual drive threshold. For the color blind, they can use 50 shades of grey or whatever colors suite them best.
December 7, 201411 yr ... Point is, I don't really want to see RED on this screen unless I HAVE to react to something .... Well summarised. It would be nice if we could informally standardise on this
December 8, 201411 yr It would also be nice if you can have the option of free space left instead of percentage. I know that there are people on this forum that like their drives as full as possible for greater read speeds on the outer edges of the disk. This way you can set it to your min free space setting so that you know that your files will copy over to that disk.
December 8, 201411 yr there are people on this forum that like their drives as full as possible for greater read speeds on the outer edges of the disk. That statement makes my brain hurt. Can you explain what you mean? Maybe a link to a page explaining this phenomenon?
December 8, 201411 yr there are people on this forum that like their drives as full as possible for greater read speeds on the outer edges of the disk. That statement makes my brain hurt. Can you explain what you mean? Maybe a link to a page explaining this phenomenon? I for instance use my server for Blu-ray backups so I have a min free space set to 50 GB. Before I did that, if I tried to copy a 25 GB file to a share that all disks that had less than 25 GB free the file will not copy to the array, and if use a cache drive the file will not move off the cache drive. unRAID works just like any other storage solution in the fact that there is not enough room for a file it will not copy over. Being able to easily see that a disk is running low on a given amount of free space would be great. Using a percentage is not as useful for people that use different sized disks. 50 GB of free space is a larger percentage of a 1.5TB (3.33%) drive than a 3TB (1.66%) drive.
December 8, 201411 yr I was wondering about this statement. there are people on this forum that like their drives as full as possible for greater read speeds on the outer edges of the disk. I am perfectly aware of the disadvantages of a full disk, I just was wondering about this speed increase you are talking about. AFAIK, every hard drive manufactured recently places the fastest outer platters at the beginning of the mapped space, so if the drive is full, you are writing to the slowest part of the drive, not the fastest. I may be mistaken, and I always welcome the opportunity to expand my knowledge, so I was wondering if you could help educate me on the speed increases when the drive is full?
December 8, 201411 yr I was wondering about this statement. there are people on this forum that like their drives as full as possible for greater read speeds on the outer edges of the disk. I am perfectly aware of the disadvantages of a full disk, I just was wondering about this speed increase you are talking about. AFAIK, every hard drive manufactured recently places the fastest outer platters at the beginning of the mapped space, so if the drive is full, you are writing to the slowest part of the drive, not the fastest. I may be mistaken, and I always welcome the opportunity to expand my knowledge, so I was wondering if you could help educate me on the speed increases when the drive is full? I have seen multiple posts in this forum by well known members explaining that the outer edges of the disk spin faster than the inner portion causing data to be read at a faster rate and that is why they leave their disks so full. I have no first hand knowledge of this but I have seen this multiple times on this forum. But what I have seen in my personal system is as my parity checks get closer to the end they go up from 80 to 120 and have an average parity check of 100. If parity checks go from inner to outer this may explain what I see and what others have claimed.
December 9, 201411 yr AFAIK read speed is faster on the outer edge of a drive which is the first to fill up. The inside closest to the spindle is when a drive is near full. case in point. 3TB 7200 RPM - ST3000DM001 - These are very fast drives for spinners Raw DD of outer tracks. root@unRAID1:# dd of=/dev/null bs=1024 count=10240000 if=/dev/sdc 10240000+0 records in 10240000+0 records out 10485760000 bytes (10 GB) copied, 58.0042 s, 181 MB/s Block count root@unRAID:# cat /proc/partitions | grep sdc 8 32 2930266584 sdc root@unRAID:# /boot/bin/calc ; 2930266584-10240000 2920026584 Raw DD of blocks near end of drive. root@unRAID:# dd of=/dev/null bs=1024 count=10240000 skip=2920026584 if=/dev/sdc 10240000+0 records in 10240000+0 records out 10485760000 bytes (10 GB) copied, 110.131 s, 95.2 MB/s
December 11, 201411 yr This enhancement has been implemented in a recent merge of code and should be available in the next release.
December 12, 201411 yr Author This enhancement has been implemented in a recent merge of code and should be available in the next release. Great ! Can't wait.
Archived
This topic is now archived and is closed to further replies.