IG-88

even if, loader 1.03b does work with 6.2.2 it "only" needs new drivers and that can be done in general i'd say its a combination of hard and software, hotplug with standard sata cables (power and data) is not "safe", if you try to connect hot and do it the wrong way with the right angle and force you can short circuit or over power, definitely not recommanded the "elegant" way and keeping all settings is still to boot up a recovery linux assemble the mdadm raid1 of all system partitions and change the version file "\etc.defaults\VERSION" but that's to complicated for most people

going to 6.2.2 with loader 1.03b is possible with a new driver package (extra.lzma) at the moment only drivers coming with dsm do work on 6.2.2 (like nic's using the e1000e driver), that can be solved that only means when going beyond 6.2.0 to 6.2.2, there is no guarantee to use 7.0 using a hardware thats already part of the platform used for a loader makes it more likely that if there is a loader that you will be able to use it, but we dont know what loader will come and what modell it will use its a good thought to look what synology is selling as optional cards for all/most models, this hardware will most likely be supported in the next 2-3 years ahci is a safe bet, so when buying storage extensions then a 4 or 8 port ahci compatible controller is very likely to work in the future when it comes to nic's the e1000e seems to be a safe choice for 1G, for 10G cheaper aquantia/tehuti base cards are sold by synology but only support certain phy's chips but there is a great variety on this cards an not all with that chip will work with the driver that comes with dsm, intel ixgbe based cards look like a good alternative but even there it can happen that there is a newer chip or card revision that does not work with dsm as synology only sells/supports the old version (uses a older intel driver version for dsm) the 918+ was a good case as the original hardware uses a realtek r8168 based chip onboard and that's the most used chip in a lot of cheaper boards, but thats a model specific thing

exactly (real3x and Hostilian did the same) olegin also had a extra.lzma that was acknowledged to work where the original fails and kept hardware acceleration it was changing the order on loading modules, i adapted this in my test extra.lzma for 918+ but got no feedback if its working the way intended if this is a problem then i would do two versions of extra.lzma, one with the hardware acceleration and one without

for a new test installation just change vid/pid in the grub.cfg (SN and mac are optional and can also be changed later) replace the files on the 2nd partition with the new ones

in general i'd say no in the one i will release soon (i'm writing the text how to install and doculentation) both driver are from the kernel or from synology had limited success on dsm 6.1 with newer mpt2sas/mpt3sas drivers and as my lsi sas controller does not work on my test hardware i dont want to add drivers that are likely to crash maybe a can find a different old hardware later that does work with the sas controller newer drivers for realtek, intel aquantia and tehuti, some people with performance problems on realtek cards might get better results but in general its more about getting newer hardware to work (newer chip revisions, different phy's) but the 6.2.2 extra.lzma is atm mainly to support a wider base as 6.2.2 breaks the drivers that come in jun's extra.lzma so atm only people with hardware supported by dsm natively can rum 6.2.2

yes but your ram problem might be gone on a 918+ as it uses a newer 4.4 kernel (3617 uses 3.10) if you create a new usb and start in in your system nothing will happen to your disks, as long as you dont install, you could check if your amd cpu does work with a 918+, if yes it might be worth testing further, disconnecting all disks with 3617, add a empty disk and install 918+ to check if it works better with your 16GB ram

if its a ryzen it should work with 918+ may be just try it, remove you working usb, crate a new one with 1.04b for 918+, boot it u p and look with synology assistant if you can find it in network (if you dont install for now you dont have to set vid/pid for usb ,that's only needed for installing the pat file and for now its just about to know if the kernel boots up)

what cpu? it its at least 4th gen intel (haswell) then you could test if this happens there too

yes, but i'd suggest to do a 1st test with a fresh install to a single empty disk and check that all hardware is working with 6.2.2

latest 6.2 (6.2.2) uses different kernel settings resulting in drivers from jun's extra.lzma not working, 6.2 with the vanilla 1.03b loader is only for 6.2.0 aka v23739, in your state (driver for <6.2.2 and loader/kernel 6.2.2 only the original dsm drivers are working) your driver is e1000 and that's not part of the original dsm as you are already in a semi brick i can send you my test version of a extra.lzma for 6.2.2 what system, 3615 or 3617? to "install" it you just need to plug you usb to a windows computer and in the mounted partition (2nd partition of the image) you just replace the extra.lzma the you plugin back in you xpenology box, boot up and wait/see what happens, it should boot and restart 2 times (1st finishing 6.2.2 installation, second for automatic update to u3 or u4, 2nd will not happen when the box does not have internet) alternative to the new extra.lzma would be downgrading https://xpenology.com/forum/topic/12778-tutorial-how-to-downgrade-from-62-to-61-recovering-a-bricked-system/

mlx4/mlx5 from the kernel is present, ixgbe (latest) for intel, also bnx2x, aquantia and tehuti i had hardware for bnx2x and tehuti 10G for testing so yes 10G is supposed to work but i haven't seen much feedback here

it meant to be used after reading the normal tutorial https://xpenology.com/forum/topic/7973-tutorial-installmigrate-dsm-52-to-61x-juns-loader/ you just copy/replace the extra.lzma to the 2nd partiton of juns laoder, can be done fore writing it to usb (with osfmount) or after it by just plugging th eusb to a windows pc all drivers that come with the original extra.lzma from jun int the loader are included in the extended version no problem there is osfmount in that and as written above just copy the new extra.lzma to the usb as last step, windows 10 will only recognize the 2nd partition so you will have no choice what to open, just replace the original file and you are good to go create your usb and test it with a empty disk, if all the hardware is recognized you can do the same with your 8 disks in the new system, boot the usb, "migrate" from 1812 to 3615 or 3617

keine ahnung, habe mir synology VMM vor 2 jahren mal angesehen aber so recht keinen nutzen bei meinem (alten) 8GB system gesehen, mein sehr gut ausgestatteter windows pc ist für sowas mit oracle virtual box besser geeignet (mehr ram, nvme ssd)

6 x sata ahci - driver ahci in kernel of dsm 8 x sata by c600 SAS - driver isci, not fom dsm and not from jun, needs additional extra.lzma with driver! 1 x nic intel 82579V, driver e1000e, in dsm 1 x realtek 8111 nic, driver r8168, in jun's loader so to get the sas working you will need something from here https://xpenology.com/forum/forum/91-additional-compiled-modules/ 6 ports + 8 ports =14, dsm 3615/17 default is 12 ports and on bigger updates manually edited config is overwritten if you don't need 14 then i'd go with leaving the last two ports unused, so using at least 2 sata ports would be preferable, i'd go 6xsata and 2x from sas definitely prevents trouble when updating if you insist on sas then you can try to disable onbiard sata and check if they also "vanish" in dsm and the 8 ports of the sas get to be ports 1-8 in dsm you can check that before moving you disks, use a usb flashdrive and one disk for installing dsm (as test) and a 2nd disk you connect later (does not need to be part of dsm, so dont initilalize or make it part of a raid), move the free disk to different ports to check if there is anything unusual with the mapping, when the mapping is fixed but different from what to expect just label the cables accordingly (if you dont have a 2nd disk you can also move the one "aktive" dsm disk, should work too) you might loose some plugins anyway as you are changing (migration) the system type to "bromolow" (in case of 3615) having 6.1.7 already does not hurt and might minimize losses but imho its not needed for what you want to do you don't have to at all, dsm will recognize the existing shr and use it (even when "disabled" in the conf), the settings in the conf just prevents you from creating a new shr, so you can change this later when it comes to the shr mode in the conf then there is no difference if its shr1 or shr2, the hybrid mode is for 1 and 2 disk redundancy when shr is enabled you will be able to use the gui the normal way to handle shr1 and shr2

nein, da kenne ich mich nicht aus, google doch mal "kvm raw disk windows 10" https://www.funtoo.org/Windows_10_Virtualization_with_KVM "... Create Raw Disk ..."

if it somehow did then on updating it writes to the usb drive, you can check it with your usb drive from the system, an the first partition there should be "GRUB_VER" and in that you would find what dsm version is installed no, it need a uart chip, the usb is often mentioned for using with the system receiving the serial data does not look like it on the pictures i've found, iLO might have a virtual com port but N40L seems not to have iLO (Gen8 does) i guess checking the GRUB_VER from the usb and checking the hardware be using a recovery linux will hopefully tell you enough you can also try to boot up from usb (?) with open media vault or if there is a free sata port boot a disk with OMV to have access to you data on the datavolume of your dsm disks (that's my fallback plan when dsm is failing and i need access over network to my data)

da das KVM als virtualisierungsplattform ist kannst du auch außerhalb des synology umfelds nach vorschlägen suchen generell sollte man wohl virtio für die hardware nehmen und eine raw disk wäre auch gut virtualiserung ist sehr ram hungrig aber ich denke mal du hast in deiner kiste wenigstens 8GB ram und die vm hat min. 4GB ram

no the output from the synology kernel goes to the serial port are you sure it did not update to 6.2.2? to check the hardware you can disconnct the disks and try a different usb with a fresh 6.1 loader (as long as you dont want to install with that loader you even dont have to change vid/pid jost booting up to network is fine that way), then check if you can find it in network you might check bios settings to (C1E can block booting)

habe sowas nicht am laufen aber es gibt da etwas vergelichbares wie die vmware tools, da sind evtl. auch extra treiber mit dabei die für mehr geschw. sorgen https://www.synology.com/en-us/releaseNote/WinVirtioDriver

https://n40l.fandom.com/wiki/Cpu_gen8 The Gen8 Microserver differs from its predecessors by offering a socketed CPU (socket 1155), making user CPU upgrades possible. The Gen8 supports both Sandy Bridge and Ivy Bridge CPUs. WARNING: The Gen8 G1610T/G2020T heatsink is rated for a 35w TDP, the newer i3-3240 SKU is rated for a 55w TDP. Caution is advised if using a higher TDP CPU due to increased heat ouput. also vorsicht wenn man eine 65W cpu mit dem org. kühlkörper nutzen will, das kann zum einen laut werden (wenn der lüfter eine temp. reglung hat) und kann auch die cpu in themisches throttling schicken weil es zu heiß wird (ich denke da vor allem an transcoding über software das dann über die normalen cpu funktionen läuft) außerdem kann es zu höheren innen temperaturen im gehäuse führen was sich auf die temperatur alller anderen componenten auswirkt (spanungsregulatoren, kondensatoren, festplatten, ...) das kann längerfritig auch folgen haben (geringere lebensdauer als "üblich") anbetracht der hohen preise von xeon cpu's kommt man evtl .besser den microserver zu verkaufen und ein passendes system mit der cpu zu bauen, die boards mit Jxxxx sind recht günstig aber es kann auch ein einfaches micro atx board mit neuerem "pentium" sein (z.b. G5400)

i dont think its a good idea to put a backup on a computer that in case of a problem does not starting anymore not that it would help much if you have a backup but i'd suggest using the same program as used in the tutorial, Win32DiskImager https://sourceforge.net/projects/win32diskimager/files/latest/download that 1.0 version also can detect the only a small part of the usb flashdrive is used and only backs up this part for mounting the boot media directly on the running dsm i find the following method easier to handle mkdir -p /mnt/synoboot1 mkdir -p /mnt/synoboot2 echo 1 > /proc/sys/kernel/syno_install_flag mount /dev/synoboot1 /mnt/synoboot1 mount /dev/synoboot2 /mnt/synoboot2 and if there is a windows 10 computer near you can just stick the usb flashdrive into it and the 2nd partition (with the extra/extra2.lzma on it) will be there as a drive letter i really don't recommend someone who even don't knows the basics to mess around with experimental stuff - i have not seen or received much feedback about the 1.04b extra.lzma, so i still see it as more or less untested

just check the connections with a multimeter if its crossed, the description in the Wikipedia article contains all information

serial output is just normal for xpenology, does not depend on the hardware type (server/desktop) but you need a null modem cable, the usb2serial is the equivalent of a normal serial cable i use a usb2serial adapter and connected to it a null modem cable A null modem cable is a RS-232 serial cable where the transmit and receive lines are crosslinked. https://en.wikipedia.org/wiki/Null_modem

https://sourceforge.net/projects/dsgpl/files/

wenn sich jeamnd findet der das wifi zeug mal dokumentiert/pflegt würde es nicht unbedingt an den treibern für neuere pcie karten oder usb dongle scheitern, die würde ich versuchen zu liefern aber ich kann das nicht auch noch supporten, das kostet einfach zu viel zeit und ich brauche das auch nicht, bei 10G netzwerk sehe ich da mehr sinn und nutze es auch selbst