Skip to content
View in the app

A better way to browse. Learn more.

Unraid

A full-screen app on your home screen with push notifications, badges and more.

To install this app on iOS and iPadOS
  1. Tap the Share icon in Safari
  2. Scroll the menu and tap Add to Home Screen.
  3. Tap Add in the top-right corner.
To install this app on Android
  1. Tap the 3-dot menu (⋮) in the top-right corner of the browser.
  2. Tap Add to Home screen or Install app.
  3. Confirm by tapping Install.

Speed Issues: Data Transfer and Data Rebuild Rates

Featured Replies

BTW, when unRAID reports that its parity-checking rate is X MB/s, the system actually transfers #drives times X MB/s off of the hard disks. With 15 drives and a reported parity-checking rate of 14 MB/s that's 210 MB/s of reads.

 

Additionally, the system writes as much as X MB/s onto the parity drive, in case there are problems.

  • 2 weeks later...

Did not post my syslog before...

Did not post my syslog before...

 

According to this syslog, the system was booted on December 12 at 1pm?  The last log entry is on Dec 13.  I just mention this in case it is not the syslog you intended to attach.  You have not turned on the new NTP feature, so perhaps the unRAID server date and time are wrong?

 

Your system looks fine, with 6 drives on the onboard controller, 4 on each TX4, and 1 on the JMB, and all appear to be optimally configured.  The only issue is that the parity drive is on one of the TX4's, on the PCI bus, so I wouldn't expect very good performance.  14MB/s sounds too slow though, even for this setup with multiple full TX4's, but there is no apparent reason for it, in this syslog.  There is no parity check covered in this syslog, so I can't say anything about errors then, that may occur behind the scenes during it.

 

Although you reported no difference, I honestly think you should see an improvement, by moving the parity drive to one of the 6 motherboard ports, or a JMB port.  Could you post another syslog, with the parity drive moved, and after a parity check?  We can check for any errors that may be affecting performance.  Sometimes a failing drive can slow everything way down.  You might check SMART reports for all of the drives.  The SMART View in the MyMain plugin of UnMENU makes it very easy to check SMART status for all drives.

 

One thing to remember, ignore the parity check speed for the first 3 to 5 percent, it actually takes quite a while before it becomes accurate.  I've often seen ridiculously low numbers at first, before it rose to over 65MB/s, and slowly decreased, until it finished with a true overall average of 51MB/s, on my all SATA system.

Thanks for all replies. Will do another parity check and post a new syslog this weekend.

I should do the parity check and then directly get the syslog right?

 

Used unmenu with myMain to watch the smart status of my drives. One drive in particular seems to be a bit under the weather.

 

Will this decrease the speed?

Time to change it?

 

www.earselect.se/smart.gif

Both drives sdp and sdn look 'troubled'.  You should click the sm link at the far right of each MyMain drive line, to produce a SMART report for each of those drives.  Save these reports, so you can compare them later for changes, especially the SMART attributes colored red.  You might check these SMART reports for how long ago the errors occurred.  Post them here if you aren't sure how to read or interpret them.

 

Drive errors can really affect drive performance, and the errors should show up in the syslog.

 

For more info about SMART, including attributes, commands, testing, discussions etc, see the UnRAID Topical Index, SMART section.

Thanks for fast reply!

 

never heard of the ata_error_count attribute

 

www.earselect.se/sdp.txt

www.earselect.se/sdn.txt

 

I guess I play it safe and order two new ones

never heard of the ata_error_count attribute

That is probably a 'translation' of this in the SMART report:  ATA Error Count: 42

 

I would not give up too quickly on these 2 drives.  Now that we can see their SMART reports, it looks to me as if the few problems they display are not recent.  I would definitely keep an eye on them.  In fact, it would be good to run a SMART long test on each, then get and compare the SMART report with your current report for each.  If no surprises, then they should be good for awhile.  See this for the SMART long test command.

 

If you are still getting too slow performance, then you should check the syslog for errors during the slow operation.

  • Author

Sorry it took so long to update, but here's my final report: After adding the cache drive, I'm getting sustained write speeds of over 35MB/second.

 

So, in the end, my reads are over 60MB/second and my writes are over 35MB/second. I am VERY happy. Thanks to all who pitched in.

 

Archived

This topic is now archived and is closed to further replies.

Account

Navigation

Search

Search

Configure browser push notifications

Chrome (Android)
  1. Tap the lock icon next to the address bar.
  2. Tap Permissions → Notifications.
  3. Adjust your preference.
Chrome (Desktop)
  1. Click the padlock icon in the address bar.
  2. Select Site settings.
  3. Find Notifications and adjust your preference.