CobraPL
-
Posts
51 -
Joined
-
Last visited
Content Type
Profiles
Forums
Downloads
Store
Gallery
Bug Reports
Documentation
Landing
Posts posted by CobraPL
-
-
So this "xfs_repair -v" destroys data in case of wrong disk/parition selected. Unlike chkdsk for Windows, which refuses to work in case it can't detect supported filesystem.
And I guess there is no backup of hdd's headers and other sensitive data stored on e.g. cache or pen? VeraCrypt for example had special CD/Iso with headers backed up. -
8 minutes ago, itimpi said:
Did you run the xfs_repair command from the GUI (recommended and safest) or from the command line. If from the command line what was the exact command you used as getting it wrong can potentially cause damage.
GUI had xfs_repair option only for cache, not for array disk. For array one it was simply absent. So I used command line xfs_repair -v but probably on wrong partition.
-
[edit: summary: xfs_repair used on wrong partition or whole disk. Instead of refusing to repair because RAW/unknown file system (like chkdsk does) it just destroyed the data. ]
Greetings,
I started array in maintenance mode, list disks and did xfs_repair -v on my HDD (array, not cache).
To my surprise, multiple errors appeared. And after restart the disk is "not encrypted".
What now? How can I at least copy data and reformat?
The HDD physically is OK, just "xfs_repair -v" destroyed my data. How to recover it? Are there any header/partition backups in unRAID?
"Unmountable: Volume not encrypted"
Perhaps I selected whole disk instead of specific partition to repair, hmm. Under Windows in case of trying to use chkdsk on unsupported disk or partition it simply refuses to check./dev/sdb1: UUID="78521ded-6ac7-40ca-84e6-7095eb84e6d7" TYPE="crypto_LUKS" PARTUUID="44a21b7f-09aa-4683-acd6-8e1762050e43"
/dev/md1p1: UUID="78521ded-6ac7-40ca-84e6-7095eb84e6d7" TYPE="crypto_LUKS" -
https://www.reddit.com/r/unRAID/comments/139h5i5
Most important/critical:
-
Global C-State Control (crashes in unRAID) - multiple sources, both Reddit and forum are swarmed with such reports!
Some reports are that Global C-State Control crash affects ZEN 1-2 only. unRAID could just warn, that in case of random crashes, try disabling Global C-State Control.
Optional: -
CPPC (may be in two places in BIOS) - performance loss under Windows, not sure if under Linux/unRAID
source: https://www.reddit.com/r/Amd/comments/pbbqd0/cppc_enabled_vs_disabled/ -
Aggressive SATA Sleep (some SSDs degradation/failures). I had problem with this setting too.source:
https://www.reddit.com/r/DataHoarder/comments/8q0cg8/comment/e0fflz - Rather disable fTPM in case you don't use it. It caused some problems in the past, not sure if it is actually fixed.
-
Global C-State Control (crashes in unRAID) - multiple sources, both Reddit and forum are swarmed with such reports!
-
On 2/5/2023 at 6:41 PM, lory995 said:
Hi, sorry for the late response but i was, and still am, really caught up with work.
Anyway i can't launch corefreq-cli it gives me an error atmDaemon connection error code 2 corefreq-ro-shm: 'No such file or directory' @ line 21300
but i didn't configure it nor looking into it so i don't know.
I'm gonna look into windows power consumption when i'll have my VM back running again, hopefully soon, but right now in unraid with 4 VMS up and running (2cores x 4 = 8 cores used , avg 30% usage all the time really low demanding tasks) and a total average of 10-15% of all CPU power used i have like 65/70W of real power draw ( at the power outlet ). I got amd_pstate drivers installed.
I guess the power draw in windows would be lower thanks to better drivers, after some search it seems that amd is working a lot in reducing their power usage but linux until kernel 6.x it's really behind.
Actually i have both of them up and am not noticing hangs nor resets at all ( at least with pstate driver and schedutil governor online for 14 days ). Same for performance.
Still, i'd like to ask you why you suggest such a thing. FWIK C-states and CPPC only function it's optimize, that's to say use the least amount of power for the most of performance. Of course if you are gonna disable them you got more performance, but that's not the topic of this thread. Actually it's the opposite.
I've written of all the above just to say: are you implying that they don't work adeguately with unraid so we can get more power saving disabling them instead of keeping them enabled? Can you post the source of your information? i'd like to know more about it before going to mess in the bios.
1.
2. -
Disable Global C States and CPPC. First one caused hangs/resets, second one degrades performance a lot (there was a thread on Reddit with benchmarks).
-
On 12/23/2022 at 6:25 AM, Hexenhammer said:
Hello,
Im trying to optimize power usage.
12700 non-K, in default 65W mode.
Right now HT disabled, so i have 8+4 cores and 12 threads.
I want to know how many P-Cores can be disabled without affecting unraid performance, mainly parity [x10 14tb HDDs + two 16tb parity HDDs, plus some SSDs for cache]
Im not using any VM [service disabled], so far the only docker i used is Krusader.
System is just for NAS + DLNA streaming + something [no idea yet what it is] to backup everything to GDrive,
no encoding, no working on it, no virtual machines [unless there is a way to take an NVMe that has windows 11 installed and move it to unraid VM, otherwise if i ever need a backup PC i can dual boot into windows 11, but ill lose the hard drive access than]
Probably even Pentium/modern Atom will be enough. 12700 is an overkill already.
I use 5900X, but I host game servers etc. For pure NAS, nothing fancy is needed. -
I ask since some people complained about "btrfs crash".
Topic from 2015 is full of complains about btrfs:
What is the best practice for latest unRAID? -
22 hours ago, ghost82 said:
If you google enable gpu acceleration over rdp you will find solutions for discrete gpu by tweaking the windows registry if I remember well..this could work also for the igpu...or not...
I know you are now thinking, if you don't know why you reply?
I would not consider RDP at all for your use, if you passthrough a gpu/igpu you are enabling hardware acceleration, so your apps/games should use hardware acceleration but rdp will render the screen output and it will send it over the network; this is how rdp works, vnc for example works in a different way.
I would consider parsec or any other protocol studied for game streaming and avoid rdp, and obviously with a passed through gpu/igpu.
THX. Except I can't passthrough AMD APU.
What are your recommendations for non-passthrough (software) scenario? -
So are we ready to announce 4650G passthrough does not work and it is waste of time for many people?
For example I've lost surely over 10h total time of my life.
OFC I can't extract bios with SI's script, because "access denied". I tried various bioses.
QuoteAfter fixing the Iommu Group I get this: internal error: Unknown PCI header type '127' for device '0000:0d:00.0'
A BIOS dump of the internal gpu is not possible.
Same here, no progress.
-
Greetings,
I plan to move work-related stuff to remote VM. I have Ryzen 4650G (5700G5750G considered) as unRAID server. I could pass-through iGPU (someone succeeded finally for 5700G!). But I want to benefit from iGPU on RDP.
AFAIK normally RDP uses software rendering, but I googled how to enable iGPU. Questions:
1. What OS is needed? W10P is enough or I need at least W2016S or W10Enterprise?
2. Do I need RemoteFX?
3. Is AMD's iGPU supported?
4. Any other considerations?
Thanks
EDIT: WARNING! 5700G does NOT support ECC. It looks confirmed. One needs 5750G. -
On 4/26/2022 at 4:13 PM, onyxdrew said:
I was able to get my 5700G to pass through, I'm on a Gigabyte X570 Gaming X
I did have to use SeaBios and Q35, along with the bios posted above:
Hi, I have Renoir GPU. Where did you get working BIOS for AMD APU? I am VERY interested. MANY people have problems (actually you are first one who succeeded).
-
I see no-one answers you.
I can't say for sure (I am beginner too), but you may not be able to pass-through Radeon cards. Anyway, every report I saw here was about failure to successfully pass-through Radeon integrated to Windows and install drivers. Not sure for Linux or Mac.
I have 4650G and I failed too. -
1 hour ago, doesntaffect said:
I cannot recall where that feature was hidden, since I sold that board already. My current config is in the EPYC watercooled thread.
Regarding your system resets - Did yo reset the bios and did a memtest to see if there are any configuration issues or a faulty memory module?
updated BIOS/BMC, disabled global C-STATE, disabled aggressive SATA, disabled CPPC. I think it stopped resetting, need more testing. RAM is multi-byte ECC, it passed all tests
-
Forget G models, pass-through does not work. AFAIK no-one succeeded (people can't install drivers under VM). So it is waste of cache. Unless 8C G version is FAR cheaper.
I recommend for this MOBO:
- CPU cooler compatibility!!! DOUBLE CHECK. I have Noctua NH-L12S because it fitted. In case of doubt, just buy this version!
- disable global C state, disable aggressive SATA (in two places, one for CPU, one for chipset), disable CPPC- double check ram slots you populate
- rather upgrade BIOS if you have <1.20.
- there are NO USB 2.0 slots at all on these mobos, buy 2.0 pendrive then. Never use 3.0 pen in 3.0 slot (it will fail, matter of days, weeks MAYBE months, but it will fail)- check if second M.2 disables any SATA ports
I have X570D4U and failed to pay extra cash for X570D4U-2L2T. Now it is 2x more expensive
I have Mellanox3, but it took precious PCIe slot.
I have 4650G (bought for a good price). Aiming for 5800X3D in case it ever appears.
Answering your question:
take 5800X or 5700G only if it is cheaper.
Look there, I have almost identical build, and this guy is PRO CLASS: -
Hmm, I've created 72GB VM and installed Srv2022Std. Image should be few GBs on cache drive and but it is 72GB.
Sparse images should grow with data written and should have option to shrink/compact them, especially after using "sdelete -z" on VM. Exactly like in Virtualbox and VMWare.
How to enable such option? -
"This is all client side and you'd have to investigate there."
Really? The VM is hibernated or paused. RDP port is probably not open at all. How to wake such VM up via client side?
Disconnecting will do nothing, Closing will cause hibernating after changing unRAID option.
OK
Only problem is what to do to easily start hibernated/paused VM w/o logging every time by hand. -
1 hour ago, Squid said:
Pause just pauses execution on the VM and keeps it powered up. Hibernate is the same as Windows. Dumps the contents of RAM, its current state to the vdisk and powers down.
That's hibernate. You set the default action in unraid to be hibernate instead of shutdown and install the virtio-tools from the ISO into the VM so that it supports that function
Because a VM is for all intents and purposes the same thing as a baremetal machine. When you log out of your Windows box. It stays on also.
Thanks for your answer, so I want my VM to:
a) pause when I simply disconnect
b) hibernate when last user logs out
c) resume upon RDP connection attempt, sth. ike wake on lan.
How can I do that exactly?
EDIT:
"You set the default action in unraid to be hibernate instead of shutdown"
Where is this option? I have both drivers and tools installed on VM. -
1. What is difference between "pause" and "hibernate"?
2. Is there any "suspend to disk"?, so after I restart unRAID I can continue work on VM w/o losing anything? Is such suspending default behavior in unRAID's graceful restart/shutdown?
I had sth like that in Virtualbox and VMWare.
3. VM, even when I am not logged, takes some CPU power. How to handle pausing/resuming in clever way? So VM is paused after e.g. I hit log out (in VM's Windows) and is resumed when I try to RDP to VM? Or maybe there is another solution like some mobile app to pause/resume by hand, but easily? -
Here is same processor two times (two cores for unRAID, 4 cores for VM).
"power plan" "performance" to the right, energy efficient to the left. 8.5% single core performance.
I'll check wattage after I obtain the proper device to check it.
-
Guys, I am back to try unRAID as backup server for various people over WAN. I also plan to have my VN for busness stuff (e-mail etc. as RDP over Wireguard) Some questions/issues:
1. UEFI or MBR? In case I ever want to pass-through GPU I think I saw that UEFI is a must. In case I ever want to move to new hardware, I may be forced to use UEFI in X years. Is there any easy switch from MBR made unRAID stick to UEFI one?
Now I have MBR unRAID unable to boot to UEFI, but it is not too late for change.2. I think everyone failed trying to pass-though 4xxxG/5xxxG iGPU. I saw some threads, everyone seem to lost the battle. Should I give up? I have 4650G.
3. I had resets w/o trace in log during unRAID trial work (once per ~1day, in the night, during HDD read or write tests). After that I did following:
a) disabled global C-state somethingb) disabled CPPC
c) disabled aggressive something for SATA
Now I'll check for resets again. Any other Ryzen recommendations?
RAM is ECC, tested/OK.
4. Any general unRAID recommendations before I fresh start? Recommended SAFE apps?
5. What are recommended VM settings for:
a) W10Pb) Server 2022 Standard (one looks like 10, not one looking like W11/next)
And which is better for RDP daily work? Money aside.6. Any help for setting following options in my AsrockRack?
Spoiler
7. I plan to RDP to VM, but OFC over VPN, Wireguard probably. I could handle Wireguard via Mikrotik router or via unRAID (preferrably).
BUT! If I set br0 how can I put RDP behind unRAID's Wireguard?
8. i440fx or Q35? Should I always select latest version?
EDIT:
9. Both for Virtualbox and Vmware I could do following steps:
a) delete temp files inside VM (under Windows)(optionally defrag inside Windows to compact all files to beginning of the disk)
b) "sdelete -z" to fill empty space with zeores
c) compact image file
How to do c) point in unRAID? Is image file sparse?
How to copy image file to remote destination every X days? -
I have system reset/"unsafe shutdown"(it is reset) once per 24h of running server. I just updated BIOS/BMC and disabled global C-STATE. Any other recommendations?
Ram is ECC, I checked it anyway, no problems.
BTW where is exactly ECO mode in BIOS 1.20? I saw only general option: Power plan - Performance mode or Energy efficient mode. Is it the same?
-
I have latest RC install, almost clean state. No other plugins (except IPMI). ECC Ram, platinum-grade 550W. I had unraid restart ~@2AM w/o reason, all logs lost. Clearing HGST/WD 18TB HDD was already past 24H.
I also have such thing in LOG:
root: error: /plugins/preclear.disk/logger.php: wrong csrf_token -
Anyone tried BIOS 1.20?
Complicated (DOS) BMC update needed.
I have Mellanox3, but it took precious PCIe slot.
[Support] Djoss - JDownloader 2
in Docker Containers
Posted · Edited by CobraPL
How can I access/locate the "output" directory? I tried for last 30min.