uldise
-
Posts
962 -
Joined
-
Last visited
-
Days Won
3
Content Type
Profiles
Forums
Downloads
Store
Gallery
Bug Reports
Documentation
Landing
Posts posted by uldise
-
-
I'm having this issue as well after upgrading from 6.1.9 final to 6.2 final. I never had the issue with 6.1.9.
I tried changing the guest OS from 'Other Linux (64-bit)' to 'Ubuntu Linux (64-bit)' but it's still occurring.
i'm still in the same boat with you - just wait for unRAID version upgrades with kernel changes.. for me early 6.1.x versions had issue, but it was gone on 6.1.9 if i remember correctly...
-
Boot to unRAID USB via PLOP unRAID Guest
Create an ISO that boots directly to the unRAID USB Stick.
How-Tos:
Pros:
- Can use unRAIDs built in OS Update.
Cons:
- Slower Boot Time
So, just wanted to contribute back to this thread with info that might help (and offer a chance for someone to correct me!).
A few years back I had virtualized unRAID via PLOP and could first-hand attest to the slooooowww boot time and issues with that; so much so that I had gone baremetal. With a new build, I decided I'd revisit virtualizing unRAID under ESX, and ran across plopkexec (actually mentioned elsewhere on this forum for a different purpose). I can say that this drastically cuts down on the boot time (for all perceived purposes, same as bare metal). With passed through M1015, I now have effectively the same config as my prior bare metal (and even moved it over seamlessly) without the hassles I recall of the AIO solutions prior.
Granted, this is now a day in, so still waiting for the other shoe to drop...
I am still running plop as I could not figure out how to get kexec to work. Can you share?
i'm running plopKExec on my offsite server - just download plopKexec iso image, add it to your unRAID VM, and thats all.. plopKexec boots, then it scans flash drives, finds unRAID flash, then automatically boots first boot choice of the flash..
- Can use unRAIDs built in OS Update.
-
First, thanks TheOne for a great plugin.. seen this one long before, but installed it today for first time..
looks like all devices are displayed ok.. some questions:
- i'm running unRAID on top of ESXi. it will be cool to manage all server hdd's in this cool plugin - so, are there an easy way to add devices manually?
- my cache drive is RAID1 array and occupies 2 slots. any ideas how to add one device to two slots or maybe another idea ?
thanks again for plugin!
Uldise
-
Fixed my error!
VM -> Edit Settings -> Options -> Guest Operating System
I had this originally set to "CentOS 4/5/6/7 (64 bit)" [force of habit - we have many vms of this type at work]
Shutdown vm, modified to "Ubuntu Linux (64 bit)", booted and the errors disappeared.
This is booting via plop, first USB device attached is the unraid USB flash drive, second USB device is blank (& unmounted).
This gem was hidden a few pages back. I had "reset high-speed" errors with 6.1.9 and 6.2 RC1, simply creating a new VM with Ubuntu Linux (x64) compatibility fixed my issue.
(I use PLOP and Host USB device passthrough)
Hi, tested some more combinations with no luck:
- changed Guest Operating System to "Ubuntu Linux (64 bit)", tried some other linux x64 with no luck
- added second USB controller, still no luck
- created new VM from scratch and used PLOP, still no luck..
looks like it's something hardware dependent.. as i wrote, with 6.1.9 no usb resets. this resets was on 6.0/6.1 early versions too..
-
hi, i just upgraded my test server from 6.1.3 to 6.1.9 and my system are running about an hour without this issue. i'm on esxi ESXi-5.5.0-20140902001-standard.
i plugged in my UPS too, then removed it from unRAID VM, still no issue anymore..
Hi, i just upgraded my test server from 6.1.9 to 6.2RC1 and this issue is back again
no changes on test server hardware/host software..and looks like workaround with second USB/UPS on USB is still working.
-
I just tested the latest stable-6.1.9 and 6.2b21 without either workaround and still see the errors.
I can use 6.1.9 (via Plop) if i have my usb UPS plugged in or if I use my card reader.
6.2b21 did not work with the 2 usb workaround (PLOP). (haven't tested with card-reader yet.)
If I boot either version with just Unraid usb i see the errors and ESXI stops seeing the flash drive until i physically remove the flash drive and plug back in.
I am currently running ESXI 6.0.0 3380124.
If anyone that had this problem and is now running error free without either workaround(2 USB/card-reader) could you please list your setup along with ESXI build+Unraid version!
Thanks
-Dave
hi, i just upgraded my test server from 6.1.3 to 6.1.9 and my system are running about an hour without this issue. i'm on esxi ESXi-5.5.0-20140902001-standard.
i plugged in my UPS too, then removed it from unRAID VM, still no issue anymore..
-
I tried some of the workarounds mentioned, none of them worked.
Can someone list all the workarounds?
please describe, what you tried ? and give us some info about your setup..
-
I've tried with the extra USB device both mounted & umounted. The errors continue in the log file.
I currently have it attached and mounted in a separate vm (to test the scenario above with a UPS attached vaia USB to a separate vm). Errors continue in the log file.
not sure i'm understand you correctly, but one of tests it worked for me was add UPS USB device and setup apcupsd in unRAID.. - then you have two USB devices - unRAID USB key and UPS USB.
can you try different USB ports? and please post some info about your hardware, at least mobo..
one more thing to give it a try - boot from VMDK.
-
I boot from plop and was seeing the errors until I attached a second (blank) USB stick to my unraid VM. Note: I am not passing through a USB controller, only assigning USB devices to the VM. After getting a UPS and configuring a VM for it (with USB cable connected) I was able to remove the second USB from my unraid VM and the errors have not returned.
Poor choice of words on my part; a USB controller was added to the vm and then the 2 USB flash drives were attached.
have you tried to mount this stick with unassigned devices? when i tested this with USB UPS, UPS daemon was loaded first to stop error messages..
-
I have two M1015s and a Supermicro motherboard with a LSI 2008 integrated on it. I used Supermicros files to flash this one to IT mode, their latest is P16. The M1015 were flashed a while back with files I found here. They are P14s. Are there any known defects in these fws? Or performance differences? Would it be advisable to upgrade to P19 or P20?
sorry, but don't fix until it works
-
I tested adding my (APC)UPS to UnRaid(6.1.3) and booted from usb flash and continue to get the resets/excessive reads. Not sure what I'm doing different from the people that have this working. Do I need to enable the UPS setting in UnRaid before rebooting? I couldn't even complete the boot process with this setup and had to re-plug in the flash drive to see it in ESXI again.
Could anyone that has the (2) usb setup working verify there are no errors in :{ESXI IP}/host/vmkernel.log ?
I kept seeing "ALERT: URB timed out - USB device may not respond"
Going Back to the card reader for now!
Does anyone know where/how to report this trbl to the linux kernel people?
when i tested with APC UPS, you must start UPS deamon, to "mount" this USB
according to your USB timeouts - i have never seen this message on my servers..
-
I noticed this on my server last night. Flash drive had over 2 million read hits on it over the last ten days.. ugh.
Just upgraded to 6.1.3 and issue is still there.
Has there been any progress on this aside from plugging in a second USB?
i don't think so.. i tried VM with Ubuntu 15.10 with kernel 4.2 on my test server last weekend, and issue are still there.
-
These are NFS exporting options; NFS export is already in place, please take a look on Settings>Unassigned Devices.
What I'm implementing now is NFS mount.
Argh.. sorry, my mistake.. have not updated yet cos i'm still on V6B15 cos USB dropout issue..
-
I have some good news: as requested, I'm adding NFS share mount to Unassigned Devices.
Right now, i'm using only defaults,nolock as mount options; are there any other mounting options I need to add?
cat /etc/exports to look how unRAID exports NFS.. one of my disks for example:
"/mnt/disk1" -async,no_subtree_check,fsid=11 *(sec=sys,rw,insecure,anongid=100,anonuid=99,all_squash)
-
Other question:
What memory do I need for this board, would the Kingston UDIMMS from my X9SCM work?
according to http://www.supermicro.com/products/motherboard/QPI/5500/X8DTH-6F.cfm supported memory types is:
Supports up to 192 GB 1333 / 1066 / 800MHz DDR3 ECC Registered memory
Supports up to 48 GB 1333 / 1066 / 800MHz DDR3 ECC / Non ECC Unbuffered memory
i'm using Registered ones, but i think your UDIMMs will work just fine.
you can always check your memory compliance at kingston's website too!
-
The board can be seen working before buying it. It will be pulled from a Supermicro chassis. It has 2 low power Xeon CPUs, L5630.
Maybe that is a drawback? But they sell the memory separately.
i don't think so, these low power procs will work just fine i think..
mine are second powerful in all range, and uses much power - when idle it uses about 100W without any hdd/controllers. at full load it uses over 200W.
-
The x4 slot only has 4 disks. I saw that your board also has a LSI SAS controller onboard. Do you use it? Or can't it be flashed to IT mode? I saw some guy here in Holland offering the same motherboard with some CPUs for little over 100 Euros. I wrote him an email offering 75 Euros, lets see how that goes.
I got much better speeds of close to 120 MB/sec on an entirely different platform with 6 disks and G2020 Pentium CPU which ran V6.0.0. All disks were on onboard SATA sockets.
I got mine board from ebay (DE) at 200 Eur without CPU/RAM, so your offer is very cheep(may be board are semi-broken ? remember to test.).. mine was with unknown bios password on arrival, so needed to re-flash latest BIOS, and yes flashed on-board SAS 2008 with IT firmware too - works like a charm. i'm not using it with unRAID but in different VM.
-
The max overall speed that I saw was 90 MB/sec. But that is with a 20 disk array on 3 M1015s on a single CPU board. It might be the most you can expect from the X9SCM with M1015. I doubt a 2 disk RAID0 parity would help much. It might improve when I move to all 4 or 6 TB disks.
when i was on X9SCM-iif with same configuration of disks, the results are pretty the same. so i think it's not CPU related.. i have 4 vCPUs assigned to unRAID VM, and i see, parity check utilize approx one vCPU fully.. what is your CPU usage when doing parity check?
and sorry - i have no more hdd's there, so can't test with more added..
one more question - you have 3 M015 - this mean one is working on x4 slot. - how many hdds you have connected to that slot? with 8 hdds using all at once, you will suffer for some bandwith bottleneck..
-
Looking at your sig, I see you have a RAID 0 array as your parity. I am wondering if that is giving you the better than expected results. I am sure it would have a positive impact, I just don't know how much.
Yes, i have Raid0 array as parity, but for parity check it should be as fast as the slowest drive. for me it's not parity, but i have three 4TB data drives - all 1TB/platter "green" drives, so it works as expected.
Two more messages, just for reference:
Total size: 4 TB
Elapsed time: 6 hours, 59 minutes
Current position: 3,50 TB (87,6 %)
Estimated speed: 105,2 MB/sec
Estimated finish: 1 hour, 18 minutes
Sync errors detected: 0
Parity check finished (0 errors)
Description: Duration: 8 hours, 30 minutes, 38 seconds. Average speed: 130,5 MB/sec
-
Wow... I would be interested in seeing the final results once the parity check is complete. That is definitely way higher than I start, though I usually start around 120MB/sec, but end up at 82MB/sec. I do see I have 2 drives with the 750GB platters, which I know are not ideal and I will try and replace down the road, but it would be great to get somewhere near where you are.
For you and dikkiedirk, what firmware are you on? I am on P20 as it looked stable after the fix, but I am curious if this makes a difference at all (as well as the UnRAID version). It may be worth trying to go back to a beta build and see what happens.
Can anyone confirm, if I am on 6.1.2 is there any potential issue with me moving back to a beta build for testing? I only have my one production box and don't want to risk my data at all.
Current progress:
Total size: 4 TB
Elapsed time: 3 hours, 19 minutes
Current position: 1,84 TB (46,1 %)
Estimated speed: 145,1 MB/sec
Estimated finish: 4 hours, 7 minutes
Sync errors detected: 0
i'm on FW version 18 - flashed it 2.5 years ago and it's working
so no need for change i think.. -
Seems that M1015 cards don't allow higher speeds. I have 3 of them running on my mainserver and I don't get much better results. I run them on the "old" X9SCM-F mainboard with an old E3-1240 CPU. The M1015s are in PCI-E 2.0 x8 slots. Yours are in PCI-E 3.0 8x slots which should result in better performance, but not so it seems.
Might well be a driver problem that limits speed on the M1015.
I have M1015 cards in my setup too, just started a parity check, after 9 minutes the results are:
Total size: 4 TB
Elapsed time: 9 minutes
Current position: 83,6 GB (2,1 %)
Estimated speed: 162,3 MB/sec
Estimated finish: 6 hours, 42 minutes
Sync errors detected: 0
see all hardware details in my sig below. i'm on unRAID v6B15 - did not upgrade further cos i'm on esxi with USB drops issue with V6 final..
-
You can let the gust SSH into the host (passwordless with keys) and shutdown the whole thing.
Is there any documentation on how this is done with regard to UPS's attached to unRAID guest VMs?
i'm connected my APC unit to Windows VM, but this is similar to unRAID i think.
first, i'm telneting to unRAID and power it down with sbin/powerdown
then wait some time (1minute for me) and ssh to esxi host and excute "/sbin/shutdown.sh && /sbin/poweroff"
look at unRAID /etc/apcupsd/doshutdown
you can ssh from here to esxi host and execute command above..
if you have VMWare tools installed on unRAID, esxi will do clean shutdown i think.. (not tried yet..)
-
Has anyone who is still have issues with resets had a chance to test with the new version 6.1? I would but I have to shut my entire VM server down to remove the passthrough configuration and then have to shut it down again to configure it again if it doesn't fix the issue.
I'm doubting the issue is resolves since none of the 6.1RC's seem to fix it but it's worth a shot.
ok, i just tested it on my test server (X9SCM-iif board) with 6.1 final and no change for me - issue are still there..
for me, a quickest workaround is to add second USB device - flash or UPS and then no issues..
-
Well here we go with some weirdness. I tested on my test server and it works fine. I'm going to try some more things on my main server and see if I can track it down. The MBs are the same in both my servers, so I don't get it. Geez, I hate these kinds of issues.
have you connected more USB devices or this single one?
looks something similar to esxi USB dropouts - when i add second USB device to unRAID, no more dropouts to me..
no changes on test server hardware/host software..
ESXI: usb 1-1: reset high-speed USB device number 2 using ehci-pci
in Virtualizing Unraid
Posted
you can use one of the workarounds mentioned early on this thread.. for example, attach second usb drive or your UPS usb cable(in this case you should start apcupsd)..
both worked for me on 6.1.x early releases.. i had not tested how this work with 6.2, but you can try