Drive performance testing (version 2.6.5) for UNRAID 5 thru 6.4

[mr-hexen]

I'd be more concerned about the results for your SSd's 

[opentoe]

I'd be more concerned about the results for your SSd's 

 

I think it is a bug. The graphical chart shows the correct SSD speeds being 545MB/sec but the text isn't correct.

 

[dikkiedirk]

It seems one of my data drives, disk 11, has degraded performance. Graph was taken with diskspeed 2.4.

 

Can someone explain this degradation to me? Or is age the cause? Perhaps a mechanical problem?

 

Will this cause reduced parity check speeds?

 

I downloaded 2.5 but still get 2.4.

diskspeed.zip

[JorgeB]

It seems one of my data drives, disk 11, has degraded performance. Graph was taken with diskspeed 2.4.

 

Can someone explain this degradation to me? Or is age the cause? Perhaps a mechanical problem?

 

Will this cause reduced parity check speeds?

 

I downloaded 2.5 but still get 2.4.

 

If the drive degraded it will affect you parity check speed and more importantly, in my experience, when they degrade like that there’s a good chance you’ll get some bad sectors sooner or later.

 

To confirm the issue I would run diskspeed for that drive only with fast scan and more scan points, you can scan every 1% like this:

 

diskspeed.sh –s 101 –f –x sdb,sdc,etc…

 

After –x list all your disks excluding disk11

 

[dikkiedirk]

It seems one of my data drives, disk 11, has degraded performance. Graph was taken with diskspeed 2.4.

 

Can someone explain this degradation to me? Or is age the cause? Perhaps a mechanical problem?

 

Will this cause reduced parity check speeds?

 

I downloaded 2.5 but still get 2.4.

 

If the drive degraded it will affect you parity check speed and more importantly, in my experience, when they degrade like that there’s a good chance you’ll get some bad sectors sooner or later.

 

To confirm the issue I would run diskspeed for that drive only with fast scan and more scan points, you can scan every 1% like this:

 

diskspeed.sh –s 101 –f –x sdb,sdc,etc…

 

After –x list all your disks excluding disk11

 

Good tip! I'll try that.

[dikkiedirk]

Using this command line to only test sdu

./diskspeed.sh –s 101 -f –x sdc,sdd,sde,sdf,sdg,sdh,sdi,sdj,sdk,sdl,sdm,sdn,sdo,sdp,sdq,sdr,sds,sdt,sdv,sdw,sdx

 

But it still tests all disks with an interval of 10%

 

Did I make typo?

 

Can't I just pick sdu?

[JorgeB]

Using this command line to only test sdu

./diskspeed.sh –s 101 -f –x sdc,sdd,sde,sdf,sdg,sdh,sdi,sdj,sdk,sdl,sdm,sdn,sdo,sdp,sdq,sdr,sds,sdt,sdv,sdw,sdx

 

But it still tests all disks with an interval of 10%

 

Did I make typo?

 

Can't I just pick sdu?

 

I never tried the –x option, but saw it on a reply above, better to wait for an answer from the author, the –s 101 should work, I used it.

[dikkiedirk]

This disk is sooo bad upto 2500 GB and will have an impact on parity build and check speed ! See the attachment.

 

Might a preclear cause reallocated sectors and rejuvenate this drive? Wouldn't trust it in my server anyhow.

 

Might test it anyway after exchanging the drive.

 

diskspeed_sdu.zip

[JorgeB]

I would replace that disk ASAP, in my experience you’ll get some bad sectors sooner rather than later.

 

The problem with this kind of symptom is that until there are some bad sectors most disk diagnose software will consider it healthy, as will a smart test.

 

You parity check average speed will also improve a little.

[dikkiedirk]

I would replace that disk ASAP, in my experience you’ll get some bad sectors sooner rather than later.

 

The problem with this kind of symptom is that until there are some bad sectors most disk diagnose software will consider it healthy, as will a smart test.

 

You parity check average speed will also improve a little.

 

Disk will be replaced asap. Preclearing some disks now. Never saw this happening BTW. But then I am not experienced in disk health.

[dikkiedirk]

Using this command line to only test sdu

./diskspeed.sh –s 101 -f –x sdc,sdd,sde,sdf,sdg,sdh,sdi,sdj,sdk,sdl,sdm,sdn,sdo,sdp,sdq,sdr,sds,sdt,sdv,sdw,sdx

 

But it still tests all disks with an interval of 10%

 

Did I make typo?

 

Can't I just pick sdu?

 

I never tried the –x option, but saw it on a reply above, better to wait for an answer from the author, the –s 101 should work, I used it.

 

I got it working. The first time I tried it I took the easy way by creating the command line in wordpad and then copying this command line to the console in putty. Somehow this doesn't work.

 

The second time I typed the command line directly in putty and I saw several disks skipped and correct disk being tested.

diskspeed_sdu.zip

[jbartlett]

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.

[dikkiedirk]

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.

 

I'm glad diskspeed helped me to find this troublesome disk.

It interferes with my plans to upgrade a 4 TB parity disk to 6 TB. I wanted to add the 4 TB disk back to the array. It would be wiser to replace the troubled 3 TB disk by that 4 TB disk though. Would I still be able to built proper parity on a new parity disk with the troubled disk still in place? And if building parity fails can I still add the old parity disk back as parity? Sorry for the clumsy description. 

 

[JorgeB]

I'm glad diskspeed helped me to find this troublesome disk.

It interferes with my plans to upgrade a 4 TB parity disk to 6 TB. I wanted to add the 4 TB disk back to the array. It would be wiser to replace the troubled 3 TB disk by that 4 TB disk though. Would I still be able to built proper parity on a new parity disk with the troubled disk still in place? And if building parity fails can I still add the old parity disk back as parity? Sorry for the clumsy description.

 

You still should to be able to replace the parity disk without issues, but it’s impossible to be certain, disk can fail tomorrow or can last years, if the parity sync fails, and as long as you don’t write anything to the array, you can put the old parity disk back, do a new config and trust parity to rebuild that disk.

[dikkiedirk]

I'm glad diskspeed helped me to find this troublesome disk.

It interferes with my plans to upgrade a 4 TB parity disk to 6 TB. I wanted to add the 4 TB disk back to the array. It would be wiser to replace the troubled 3 TB disk by that 4 TB disk though. Would I still be able to built proper parity on a new parity disk with the troubled disk still in place? And if building parity fails can I still add the old parity disk back as parity? Sorry for the clumsy description.

 

You still should to be able to replace the parity disk without issues, but it’s impossible to be certain, disk can fail tomorrow or can last years, if the parity sync fails, and as long as you don’t write anything to the array, you can put the old parity disk back, do a new config and trust parity to rebuild that disk.

 

I guess I have to rely on Hitachi's error correcting. Parity built will be slow at certain points is my guess?

[JorgeB]

I'm glad diskspeed helped me to find this troublesome disk.

It interferes with my plans to upgrade a 4 TB parity disk to 6 TB. I wanted to add the 4 TB disk back to the array. It would be wiser to replace the troubled 3 TB disk by that 4 TB disk though. Would I still be able to built proper parity on a new parity disk with the troubled disk still in place? And if building parity fails can I still add the old parity disk back as parity? Sorry for the clumsy description.

 

You still should to be able to replace the parity disk without issues, but it’s impossible to be certain, disk can fail tomorrow or can last years, if the parity sync fails, and as long as you don’t write anything to the array, you can put the old parity disk back, do a new config and trust parity to rebuild that disk.

 

I guess I have to rely on Hitachi's error correcting. Parity built will be slow at certain points is my guess?

 

Should be a little slower but nothing major, I’d think it will add less than an hour to your build time.

 

Just to clarify my earlier post if you are not familiar with the procedure, the new config has to be done with the old parity and the problem disk, i.e, the same exact array you had before the new parity, when you start the array with the trusted parity it will trigger a parity check, stop that asap, stop the array, unassign the problem disk, assign new disk and begin rebuild.

 

There’s also another option, probably simpler, make a backup of your flash drive now, and if there are any errors during the sync, stop, put old parity back in and restore the flash backup, in case you do this make sure the array is stopped before making the backup or it will trigger a parity check after restoring it and starting the array for the first time.

 

[dikkiedirk]

I'm glad diskspeed helped me to find this troublesome disk.

It interferes with my plans to upgrade a 4 TB parity disk to 6 TB. I wanted to add the 4 TB disk back to the array. It would be wiser to replace the troubled 3 TB disk by that 4 TB disk though. Would I still be able to built proper parity on a new parity disk with the troubled disk still in place? And if building parity fails can I still add the old parity disk back as parity? Sorry for the clumsy description.

 

You still should to be able to replace the parity disk without issues, but it’s impossible to be certain, disk can fail tomorrow or can last years, if the parity sync fails, and as long as you don’t write anything to the array, you can put the old parity disk back, do a new config and trust parity to rebuild that disk.

 

I guess I have to rely on Hitachi's error correcting. Parity built will be slow at certain points is my guess?

 

Should be a little slower but nothing major, I’d think it will add less than an hour to your build time.

 

Just to clarify my earlier post if you are not familiar with the procedure, the new config has to be done with the old parity and the problem disk, i.e, the same exact array you had before the new parity, when you start the array with the trusted parity it will trigger a parity check, stop that asap, stop the array, unassign the problem disk, assign new disk and begin rebuild.

 

There’s also another option, probably simpler, make a backup of your flash drive now, and if there are any errors during the sync, stop, put old parity back in and restore the flash backup, in case you do this make sure the array is stopped before making the backup or it will trigger a parity check after restoring it and starting the array for the first time.

 

Thanks for pointing it all out. I will make the backup of the flash drive, but I hope everything turns out great and I won't have to use it.

[Pducharme]

Thank you for the very nice script !

 

Here is my Graph   almost all my disks have similar performance (except my SSD Cache)

 

[dikkiedirk]

It is fun to check a parity checks speed while observing speeddests charts. The parity checks speed follows the slowest disk closely. Just a few MB/s below it. I had a really bad disk in the array that impacted the parity check speed greatly. Still have to check the tunables test. These are really great scripts!

[Dephcon]

Here is my Graph   almost all my disks have similar performance (except my SSD Cache)

 

So lovely, my junk is all over the place.

[clowrym]

I seem to be getting this error when trying to run Diskspeed...

 

root@Tower:/mnt/cache# diskspeed.sh
./diskspeed.sh: line 30: syntax error near unexpected token `('
./diskspeed.sh: line 30: `iterations=1 samples=11 showhelp=0 outputfile="diskspeed.html" skipdrives="" unraid="" ShowGraph1=0 log=0 fast=0 numargs=$# for ((i=1 ; i <= numargs ; i++)) do'

 

latest unraid 6.1.4

 

any idea's?

[jbartlett]

I seem to be getting this error when trying to run Diskspeed...

 

root@Tower:/mnt/cache# diskspeed.sh
./diskspeed.sh: line 30: syntax error near unexpected token `('
./diskspeed.sh: line 30: `iterations=1 samples=11 showhelp=0 outputfile="diskspeed.html" skipdrives="" unraid="" ShowGraph1=0 log=0 fast=0 numargs=$# for ((i=1 ; i <= numargs ; i++)) do'

 

latest unraid 6.1.4

 

any idea's?

 

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".

[opentoe]

Obviously my single SSD drive's speed is reporting wrong speeds. Anyone else? It is the last drive in this list.

 

 

Parity: WDC WD4001FAEX-00MJRA0 WD-WCC1F0127613  4.0 TB   133 MB/sec avg

Disk 1: WDC WD2002FAEX-007BA0 WD-WMAWP0099064  2.0 TB   114 MB/sec avg

Disk 2: HGST HDN724040ALE640 PK2338P4HE525C  4.0 TB   139 MB/sec avg

Disk 3: Hitachi HDS723020BLA642 MN5220F32HHMJK  2.0 TB   118 MB/sec avg

Disk 4: HGST HDN724040ALE640 PK2338P4HEPH8C  4.0 TB   138 MB/sec avg

Disk 5: Hitachi HDS723020BLA642 MN1220F30803DD  2.0 TB   134 MB/sec avg

Disk 6: WDC WD2002FAEX-007BA0 WD-WMAWP0284322  2.0 TB   114 MB/sec avg

Disk 7: HGST HDN724040ALE640 PK1334PCJY9BRS  4.0 TB   139 MB/sec avg

Disk 8: WDC WD4001FAEX-00MJRA0 WD-WCC130263021  4.0 TB   146 MB/sec avg

Disk 9: WDC WD4003FZEX-00Z4SA0 WD-WCC130966733  4.0 TB   148 MB/sec avg

Disk 10: HGST HDN724040ALE640 PK2334PCGYD32B  4.0 TB   140 MB/sec avg

Disk 11: WDC WD2002FAEX-007BA0 WD-WMAY04375059  2.0 TB   110 MB/sec avg

Disk 12: WDC WD2002FAEX-007BA0 WD-WCAY01300087  2.0 TB   118 MB/sec avg

sdg: WDC WD2002FAEX-007BA0 WD-WMAY01552774  2.0 TB   118 MB/sec avg

sdh: Samsung SSD 840 EVO 120GB S1D5NSBFB10859N  120 GB   118 MB/sec avg

 

[jbartlett]

Obviously my single SSD drive's speed is reporting wrong speeds. Anyone else? It is the last drive in this list.

 

sdh: Samsung SSD 840 EVO 120GB S1D5NSBFB10859N  120 GB   118 MB/sec avg

 

If you use the -l or --log option, it'll create a debug log file that I can analyze. Please attach it here or email it to jbartlett at strangejourney.net

[opentoe]

Obviously my single SSD drive's speed is reporting wrong speeds. Anyone else? It is the last drive in this list.

 

sdh: Samsung SSD 840 EVO 120GB S1D5NSBFB10859N  120 GB   118 MB/sec avg

 

If you use the -l or --log option, it'll create a debug log file that I can analyze. Please attach it here or email it to jbartlett at strangejourney.net

 

I'll run it again and use the log option. Hold tight.

 

Log sent to your email. Thanks!

 

EDIT: What's really weird is when I copy a single large file to it the speed tops out of 112MB/sec. Just like the results I get when writing to the array. It is an UNassigned drive. Not part of the array at all. I do have an old speedtest report from late last year and that same drive was getting around 480MB-500MB/sec speeds. So this may have nothing to do with the speedtest script at all. Maybe post a thread in the V6 support area.