MAM59

AAAH! I've mixed this (6.12) up with the UNRAID version.... now I see that the path is recent. My fault, sorry! I once tried it (long ago), but the fans did not work. So I deleted the plugin. Still it looks for this driver at boot time. I'm just puzzled. "lsmod | grep nct" gives me nothing. Wheras "modprobe -c | grep nct" gives a lot (but only aliases): root@F:~# modprobe -c | grep nct alias i2c:nct1008 lm90 alias i2c:nct210 lm90 alias i2c:nct214 lm90 alias i2c:nct218 lm90 alias i2c:nct72 lm90 alias i2c:nct7802 nct7802 alias i2c:nct7904 nct7904 alias of:N*T*Connn,nct1008 lm90 alias of:N*T*Connn,nct1008C* lm90 alias of:N*T*Connn,nct214 lm90 alias of:N*T*Connn,nct214C* lm90 alias of:N*T*Connn,nct218 lm90 alias of:N*T*Connn,nct218C* lm90 alias of:N*T*Connn,nct72 lm90 alias of:N*T*Connn,nct72C* lm90 alias tty_ldisc_14 ppp_synctty alias symbol:HWMON_NCT6775 nct6775_core alias symbol:drm_helper_disable_unused_functions drm_kms_helper alias symbol:ieee80211_emulate_add_chanctx mac80211 alias symbol:ieee80211_emulate_change_chanctx mac80211 alias symbol:ieee80211_emulate_remove_chanctx mac80211 alias symbol:ieee80211_emulate_switch_vif_chanctx mac80211 alias symbol:iwl_puncturing_is_allowed_in_bios iwlwifi alias symbol:iwl_uefi_get_puncturing iwlwifi alias symbol:mt76x02_mcu_function_select mt76x02_lib alias symbol:mt792x_assign_vif_chanctx mt792x_lib alias symbol:mt792x_unassign_vif_chanctx mt792x_lib alias symbol:nct6775_probe nct6775_core alias symbol:nct6775_reg_is_word_sized nct6775_core alias symbol:nct6775_show_alarm nct6775_core alias symbol:nct6775_show_beep nct6775_core alias symbol:nct6775_store_beep nct6775_core alias symbol:nct6775_update_device nct6775_core (I CAN load that nct6775 driver, but at boot time it wants the nct6687 which does not exist (anymore?)) And I cannot find ANY reference to 6687 on the boot stick!

sorry to kick in a bit offtopic, but I am too stupid to reach @ich777 per direct message: Since a lot of versions UNRAID always tries to load a NCT Driver here (and always fails), the message scrolls offscreen too fast to be able to read the whole name. But right now I was able to produce it manually (by reading and trying this thread): root@F:~# modprobe nct6687 modprobe: FATAL: Module nct6687 not found in directory /lib/modules/6.12.24-Unraid Yeah we are on 7.1.2 now already, so very likely the above path does not exist anymore. but for an unknown reason the path still exists (seems to be recreated on boot) ?!?!? But still he tries to load it at boot time and I am not able to locate WHERE and HOW TO TO TURN IT OFF again! Some hint for me ? (since the driver is not found, nothing is loaded, nothing is harmed, its just this message that appears and obviously I am not able to load the driver even if I liked too)

My Guess would be that your NanoKVM (whatever it is) adds some kind of additional drive to UNRAID And then, in total, you exceed the drive limit of your license. From the specs: At the same time, using the extra storage space of TF card, it can be mounted as a USB flash drive device. See if you can turn this off, else you need to upgrade the license.

Dann würde es ja nicht mehr drinstehen. Guck nochmal nach, ob es wirklich weg ist (und anschliessend neu Booten, sonst lungert es weiter im Speicher rum). Lösch zu Sicherheit die Zeilen auf GO.TXT und boote dann neu.

Na ja, bei 0,5m und Cat7 ist die Wahrscheinlichkeit, dass es die Ursache ist, nicht sooo hoch 🙂 Also nach Durchsicht Deiner Logs vermute ich einen Fehler beim Powermanagement. Du versuchst irgendwo zuviel Strom zu sparen, die LAN Karte wird abgeschaltet nur um 5s später zu merken, dass die Idee strohdumm war und sie schnell wieder aktiviert wird. und hier haben wir auch schon den "BÖSEN" (in Deinem /boot/config/go.txt File): # ------------------------------------------------- # Powertop & ASPM # ------------------------------------------------- powertop --auto-tune Schmeiss das Plugin raus und es wird wieder gut sein... (oder mach zumindest --auto-tune weg, das ist echt tödlich für einen Server)

vielleicht doch erstmal ne diagnostics posten, damit man mal genau sieht, welch Hardware da vergraben ist? Ansonsten, die nächsthäufigste Ursache für Symptome wie Deine sind "minderwertige Kabel". Auch wenn man bei 2,5G sagt, man könne bedenkenlos die alten CAT5.e Kabel weiterverwenden, so gibt es da doch oft ein paar Unverträglichkeiten. Besonders bei alten Kabeln in der Wand wäre ich vorsichtig, die Dosen sind oft recht schlampig (Abschirmung) angeschlossen...

Another trick would be to install the plugin "appdata backup" and run it it creates backups from your docker settings. You can then switch and use the plugin again to restore previous settings stresslessly 🙂

No. Use that plugin, it works well. UNRAID is more or less just a Fileserver. All things like Dockers and VMs come from other sources and are not really in the scope of LimeTech. They just take them and put them in, but they do not develop them or build tools around them.

why bother with it over and over again? Give it a directory on a pool device and it can grow and grow (until the device is full) without beeing pampered by you every few weeks.

Sagen wir mal so: Du bist mutig! Bei einem kastrierten LAN ohne Flow Control zu arbeiten ist echt sportlich. Da wundert es mich wenig, wenn es andauernde Port Resets gibt. (merke: 2,5G ist "10G mit Pausen", nur ohne Flow Control weis niemand, was Pause und was Daten sind...)

Ja, das passiert eigentlich automatisch, wenn Du den Host bei NPM anlegst und von dort auch ein Zertifikat anforderst. Das Problem wird eher sein, dass niemand (auch NPM nicht) Deinem selbstgestrickten Zertifikat aufh der ...:8000 vertrauen wird und somit kein Tunnel in Betrieb geht. Probiers einfach aus. Wenns klemmt, dann mal ernsthaft darüber nachdenken, warum Du INTERN ein Zertifikat verwenden willst/musst/kannst? Schmeiss das auf 8000 weg und verbinde über http.

I would not say it is a "problem", just "work to do". "lost+found" is automatically created by a filesystem check each time. You are urged to inspect it, move out files that you want to keep, delete the rest of the files and finally delete the directory too. So, afterwards it does not exist anymore and also is no "problem".

I would more think they did not even know about the possible combinations (yet). But anyway, it should be clear that UNRAID is still far away from a proper ZFS implementation. (BTW: the marketing division only tells that UNRAID is now capable of importing "many" foreign ZFS pools. They never have said that they can be used afterwards :-))) )

TrueNAS is only a port of a port. The real origin is Solaris 11, which does not exist anymore. The next sibling was FreeBSD from which TrueNAS and others derived it. And, as usual, ports are never really complete. The more they copy, the more gets lost. Crossing from BSD to Linux gives a lot of troubles because their Kernels differ a lot and not all features are available. The Linux strategy was to implement new stuff fast, BSD is more conservative, it tries to be back-compatible as long as it goes. Therefor there are always some missing structures/function calls or something and a cross-implementer has to make compromises to get things running. Often then they skip complicated stuff ("we will implement this later..." and it never happens) and if then later on others take this work to port it again, they dont even know anymore that something essential is missing. So, it is not really a surprise that "ZFS on UNRAID" is far from beeing a real ZFS. But, if people can live with it, who cares?

no need to wait. if the errors are always in the front (this was the same here too), you can safely cancel, reboot and start a new check. If it shows the usual errors again then, you know what to blame and to do. Just an hour run should be enough to know. (The check does not change anything normally, it just reads the data and marks the problematic sectors. If you set the "correct me" hook it then "repairs" these faults by overwriting them with the supposly correct value) Write down the sector numbers that it complains about, if they are always the same (or at least very close to each other from the last run) its very likely that you have a controller problem.

I would propose that you do another check very soon (should give 0 errors again), then reboot and recheck. If there are 5 errors again after the reboot, you sadly have the wrong SATA adapters. You have to consider then if you can live with those errors (that should be harmless, but of course there is no guarantee for this) or throw out the sata controllers. But in any case, stay away from those >4port cards! They use a multiplexer intern this halfs the speed instantly. You slow down your whole system and even risk blocking (accessing 2 drives on the different side of the multiplexer is not possible, a write to the array will always need this).

I remember I was facing the same problem. Since I only used the onboard SATA 1-8 the errors are gone. But the 2 cheap SATA controllers in the slots (dunno which chiptype, too long ago) had the problem on server shutdown. They seemed to modify some sectors before power turned off. The sector numbers were all quite "early" on the disk and mostly the same each time. (It was harmless, did never kill any data. Maybe a lazy flush of some cached directory or something.) Its only noticeable with array drives. So I moved the drives "to the front" and everything is fine again. (But this does not mean that this is the same for you, just give it a try if you can)

Beware that to an yet unfixed bug, the notification mail/msg shows the results of the semi last run. They say, they have fixed it in 7.1 We will see...

Take a look at Baikal, works like a charm for me https://github.com/ckulka/baikal-docker (also available directly from UNRAID's APP page)

Nein, dadurch würde es nicht schneller, aber vielleicht langsamer. Du hast also ne NVME mit nur einer Lane? Da bist Du dann im Speedbereich eines Raspberry Pi 5 🙂

ja, besser zu 2*2 Tb greifen, wird Dich nicht viel ärmer machen heutzutage

empty your browser's cache (CTRL-F5 usually) (or wait a day until it times out by itself)

ALL current (7000/9000) AMD Cpus have an IGPU as far as I know. So you do not need the 16x slot and can use it for other things like 100G LAN 🙂 There are other motherboards, some even come equipped with 10G Ethernet already. Mine for instance (ASUS ProArt X670E CREATOR WIFI) comes with 10G+2.5G onboard (but both are twisted pair only, so I still use my Mellanox SFP+ card in a slot). These "better" boards usually offer "bifurbication" support for the 16x slots, so it can be split into 4*4x and be used to drive up to 4 additional NVMes in an extra card (there are "up to" 4 NVMe slots onboard already). But beware! read the manuals carefully, there are certain restrictions on using onboard devices and populating slots! Before buy, download the manual and read it carefully. The normal ads won't show you them. Warning: Although it comes close to your desires, I would not recommend "my" Motherboard as a base for UNRAID. It's a typical gaming thing, it does not care about power consumption, this box (inluding fast 16core CPU and a medium fast graphics card) draws 150W when idle and up to 500W when heavily gaming. This is nothing that you like to run 24/7. This box is for film cutting/encoding and gaming only. Even for normal surfing I use a different minipc with just <20W

Unwahrscheinlich, wer kastriert schon eine NVMe??? Die beiden werden wohl auf dem Mutterbrett sein und er hat nur die Parity per USB extern angeschlossen. Quasi als typischen Denkfehler.

For streaming usually the builtin IGPU is enough. No, older Intel are usually worse when it comes to spare Lanes Yes. Usually AMD offers more lanes to the boards Total no-go. Most of these cards will not even show up with less than 4 lanes connected. Like HBA they need the bandwidth. There is no option to slow down. You dont need to calculate the possible max speed, its either 10G or ZERO.