On 8/9/2018 at 7:27 AM, juanamingo said:

but also don't see any explicit 'trim' log entries  other than

 

Aug 9 00:00:02 Guardian root: /etc/libvirt: 923 MiB (967876608 bytes) trimmed
Aug 9 00:00:02 Guardian root: /var/lib/docker: 12.2 GiB (13096083456 bytes) trimmed

 

but i don't believe those are actual trim related.

 

They are trim related.  Dynamix SSD Trim trim's all filesystems that support trim.  In this case, your libvirt image and docker image are both available to be trimmed.  Your SSD however is not being trimmed.  If it was you would see a line that said /mnt/cache

 

The real difference between those images and the drive itself is that the images are virtual file systems.  Even if they're hosted on a spinner, they still get trimmed  (even though the trim on the images actually does nothing - even when its hosted on a SSD)

Edited August 11, 20187 yr by Squid

21 hours ago, Squid said:

They are trim related.  Dynamix SSD Trim trim's all filesystems that support trim.  In this case, your libvirt image and docker image are both available to be trimmed.  Your SSD however is not being trimmed.  If it was you would see a line that said /mnt/cache

 

That's what I wasn't 100% sure of - i figured i'd see something related to cache for SSD trimming, but i'm not.  Does that mean that it's not available/supported for trim? or not running?

I have it the SSD trim schedule to Daily at 00:00

 

5 minutes ago, juanamingo said:

 

That's what I wasn't 100% sure of - i figured i'd see something related to cache for SSD trimming, but i'm not.  Does that mean that it's not available/supported for trim? or not running?

I have it the SSD trim schedule to Daily at 00:00

 

Not available / Not supported / Not whatever

 

You can do the exact same thing as the plugin does at the command line:

fstrim -a -v

 

If you don't see a line for /mnt/cache, then it's not being trimmed.

Edited August 12, 20187 yr by Squid
my keyboard sucks

5 minutes ago, Squid said:

 

fstrim -a -v

 

Ran that and got back

/etc/libvirt: 924.1 MiB (969015296 bytes) trimmed
/var/lib/docker: 12.3 GiB (13221044224 bytes) trimmed

 

So then it's Not available / Not supported / Not whatever

 

Not having paid much attention to it in the past (always assumed it was working), is that something i should worry about / investigate & get working (the cache ssd trim)?

 

Edited August 12, 20187 yr by juanamingo
clarification of 'get working'

Easiest solution is to move the cache drive off of the HBA, and onto the motherboard

3 hours ago, Squid said:

Easiest solution is to move the cache drive off of the HBA, and onto the motherboard

Thanks! Next maintenance performed I'll give that a try!

 

@juanamingo so at the end of the day, your rig was not running TRIM on the ssd's attached to your 9305 HBA.  

 

I too will just move the ssd to a mb sata port.  I was hoping I could avoid using the mb and have everything on the 9305.

 

I guess the only other option is to try the fw update, but I didn't see anything in the list of fixes that would help.

 

Well, the other guy did also suggest to just NOT run the TRIM plugin, so that is a option too.  I have it disabled right now.

 

On 8/12/2018 at 2:21 PM, Switchblade said:

so at the end of the day, your rig was not running TRIM on the ssd's attached to your 9305 HBA.  

 

Correct, i thought it was because i never saw any errors pop up in the logs and never looked any further.  

 

I'm in the same boat as you, wanting to keep all my drives on my HBA, but it is what it is.

 

If I have some time this week, I'll swap to the mbd and when i do i'll post back with results.

Not sure when i'll have time tho.....

 

I installed a 2nd 1tb ssd drive and connected both the new one and the existing ssd to my mb sata ports.  Enabled Trim to be scheduled and no more errors.

 

output that is in the log:

Sep 23 06:00:32 Tower root: /var/lib/docker: 34.4 GiB (36966330368 bytes) trimmed
Sep 23 06:00:32 Tower root: /mnt/cache: 1.7 TiB (1853984260096 bytes) trimmed

I did not bother trying to upgrade the firmware 9305.  At this point, I yield and do what most others have done, and be forced to put ssd's on mb ports.

Edited September 23, 20187 yr by Switchblade

@limetech  I have run several tests that write to the cache pool and then the Mover transferring them to the array.  I also ran a full Parity check, all good.  Syslog shows no major errors and as mentioned in my last post, trim command ran successfully.

 

However, I just noticed that my Used Space on my cache in the gui is now showing about half of what it was before.  It was showing I had about 84gb in use and this is when the Mover has ran, and now it shows about 44gb.  I have checked and all my files that should be on the cache drive are there.  My dockers are running on the cache drive and working great.  

 

With that, it seems that the gui is just reporting the wrong information.  Maybe splitting the total in use by 2 x drives ??  Again, no errors in syslog.

 

 

I did a quick search and nothing came up that matched my issue.  Should I be worried about this?

 

Edited November 21, 20187 yr by Switchblade

Hi All

 

Trying to use this card because of PCIe problems (To many cards, so moving to a single card for all 24 drives)

But I just cant get into the Bios of the card Ctrl+C have no effect?

 

Unraid Works with all the drives except the two "top Ones on the card" I suspect some setting because the cables are okay (Tried moving them around - its just the top two channels?

 

Any idea on how to get into the "Bios UI" of the card? I tried pulling all drives and the UNraid USB but then the Intel boot manager keeps starting up, looking to boot from LAN?

 

If I dont get this working soon I would have to rool back to the old HW also I dont know if its the Card or the configuration?

 

 

I'm a little confused by your post, but I'll try to offer some help.  You will get more help here if you provide more details, motherboard, unraid version, memory, etc.

 

You say that unraid works and all drives are visible expect those using ports 16-23, the top two ports JT5.  If correct, then I would agree that checking the SAS controller bios settings is a good idea.  That is strange that you are unable to enter the bios menu.  If you haven't already, try a wired usb keyboard plugged directly into your motherboard usb port, and do not use wireless keyboard or usb hub of any kind.  I have found that wireless keyboards  and usb hubs sometimes are not active at the very time you need to press the keys to enter bios.

 

Based on the user guide, it looks like all ports are active by default, so you really shouldn't have to change anything.  Check your bios and firmware versions, as I had to upgrade mine before the controller worked properly in my system.  My system hung at boot trying to initialize the controller until I upgraded the bios and firmware.  That said, I'm not using JT5 in my system, no need.  I have all 12 drives plugged into JT1 and JT4.  Even after the bios update, I didn't need to make any changes in the SAS controller bios.

 

As for your system trying to boot from LAN, you need to use the motherboard bios / UEFI to change the boot device order.  In my system, I disabled LAN boot, and have the first device to be USB unraid flash drive.   I'm confused as to why you say it works fine except the top ports,  but then you say it's trying to boot from LAN.  In my opinion, I disable everything on the motherboard that is not in use, wifi, hard drive controllers, serial ports, etc.

 

Take a close look at section 1.8 of the User Guide, I've attached a screenshot of that section.

 

Good luck!

 

 

 

 

 

 

 

This may be of some help:

 

Art of Server (youtube): - https://www.youtube.com/c/ArtofServer/videos

How to fix LSI firmware settings - https://www.youtube.com/watch?v=nLEVpU8u_Ls

How to troubleshoot your LSI IT mode HBA - https://www.youtube.com/watch?v=TC4DovIcGLU