For what its worth turning off Cache Dirs Plugin seems to have stopped the errors and all the drives are spun down that aren't currently being used.  i wonder if the continuous scans could have been a factor in the slowdown spikes.  Seems that there were no slowdowns past 6tb either before i turned off the plugin and that slowdowns continued after 4tb and to some degree after 5tb.

 

 

 

Latest update - rebuild will be done in about 10 hours for a total of a little over 7 days.  I can no longer view my log - i get this error "

Fatal error: Allowed memory size of 134217728 bytes exhausted (tried to allocate 134037536 bytes) in /usr/local/emhttp/plugins/dynamix/include/DefaultPageLayout.php(418) : eval()'d code on line 73".

 

Looks like i have filled up my log file's max of 128mb - results of df -h /var/log are 

Filesystem      Size  Used Avail Use% Mounted on

tmpfs           128M  128M     0 100% /var/log

 

I have shut off all dockers, disabled cache dirs and am just hoping that everything is ok after the rebuild is done late tonight.  As far as i can tell i can't reach the shares via samba at moment, but unraid webgui is working fine. 

 

Just waiting for the finish of rebuild and then instructions on what to do after.  So far i have that it may all be fine after a reboot.  Check cables after shutdown for reboot.  

Can or should i clear/reset the log in the meantime so that it continues to log? Or best just not to touch anything until done?

The rebuild finished and all data appears to be intact.

On 1/21/2019 at 1:14 PM, johnnie.black said:

No, if I had to guess my number one suspect would be one the disks, there have been previous issues with other users and those Seagate disks, seems some of them can be duds and perform very poorly, you could try rebuilding one a a time a see if it's one of them.

Is it safe to do test rebuilds? ie start one for a while, see if slowdown problem goes away, if not stop the rebuild, shut down the array and put the original disk back in? using process of elimination?  Assume that turning off mover schedule would be necessary.

 

  I am guessing it is not one of the 4tbs because problems continued through the 5 and even after the 5, but no problems at all from 6 to 8 in random slowdowns.

It would be best if you did it before, when both disks were invalid, you could then cancel and repeat with another disk, if you start a rebuild now you should let it finish.

Thanks Johnnie, I am going to try diskspeed docker first @10% increments on the disks and then smaller.  Perhaps that will highlight a problem disk. 

 

I am interested in finding new hardware/storage rack that will support 20 to 30 disks but has newer and faster to increase speed of rebuilds etc.  Price would be a consideration within reason.  If anyone has any suggestions please feel free to post them.. Or should i post this question in a different part of the forum?

2 hours ago, FrozenGamer said:

I am going to try diskspeed docker first @10% increments on the disks and then smaller.  Perhaps that will highlight a problem disk. 

Worth a try, but the slow performance might be limited to writes.