IG-88

every model comes with its own (~200MB ) *.pat file to install and gets its own set of kernel, drivers, ..., the hack jun created is specific for that model so if you want a different model you would need a different kernel and pat file and a newly matched hack, as jun is the only one with the code (and the knowledge) there is no way of adapting, even if you manage to change the model in bootloader you would end up with a dsm with working protections preventing you from using it on non synology hardware

that was last year and for dsm 6.0 (i still have the last version but never used it, after some testing i migrated from 5.0 to 6.1) there will be a new one for 6.1 and? thats your actual problem? what hardware? if its old and not uefi you will have to use the mbr version of the loader

accessing (read/write) and booting fro m it are two different things, with the win10 creators update ms introduced access to more then one partiton so it was possible to access 1st and 2nd with win10 to modify files, with fall creators update that changed, now you can only access the 2nd parttion (to replace extra.lzma) you can mod the 1st partition before writing to flash drive with osfmount (img file mount) as in polanskiman's howto for installing 6.1 to change grub.cfg on a already created drive you can use a system able to handle it properly (like linux) and mount the partition in question or use win32diskimager to create a img of the flash drive, mod the with osfmount (see above) and write it back to the flash drive booting depends on you hardware where you want to boot the flash drive, uefi and sometimes you will have to choose media or partitons, mbr (old bios) does need a different loader ??? if you ask if modding the grub.cfg is needed, yes it is, without the correct vid/pid of your usb flash drive you might be able to boot but installing dsm *.pat file will fail with error 13 if the vid/pid does not match quicknick seems to be about to release his 6.1 loader, he uses the same files from juns loader but extended it massively so you can boot up with a configuration tool (instead of the usual installer) that can be used with putty over network to directly modify the boot image on the flash drive, like setting vid/pid, mac address or serial, that includes reading the vid/pid and mac of the hardware its running on and writing it (automatic with some questions if its all correct) to the grub.cfg and when choosing reboot to exit the loader will come up with everything made ready, the only thing to do is to write the usb flash drive with his loader, booting it on the hardware you want to use and connecting to it (usually via network using telnet/ssh), so keep checking the forum i guess if he release it it will be in the loaders and/or tutorial/guide section https://xpenology.com/forum/forum/31-loaders/ https://xpenology.com/forum/forum/36-tutorials-and-guides/

in vm config goto serial port, enable com1, modus host-pipe, uncheck connect with host/pipe and for path on windows have this "\\.\pipe\ds3615" (ds3615 is just a string you choose for each vm differently as you can have more then one vm using this) in putty you select serial instead of the usual ssh or telnet, where the ip/host is (now named "serial line") comes the pipe you defined in the vm "\\.\pipe\ds3615" and speed is 115200, after starting the vm the pipe is available and you can connect with putty

dmesg1 is first thing. intel + realtek. dmesg2 is second. rebooting and i connected to realtek , still has intel nic. dmesg3 is 3rd. i ejected intel nic and booting with realtek. by the way, this reatek nic has *.*.*.75 in lanport 1. but worked. looks normal, realtek driver is loaded and there is eth0/eth1 or with just the realtek only eth0 also it works as expected after installation, problem seems to be dhcp related and only in bootloader when installing dsm sorry but i 'm limited in resource and knowledge so i only look for the drivers (works as it should be) and try to help people get things running, well when we see the positive side - at least you got it runnig with the internal nic

ich glaube das mit dem 2.5" hat sich für mich erst mal erübrigt, die 4TB platten die für mich in frage kämen gibts nur von seagate und sind shingled recording ich hatte ohnehin vor das in die 2 jahreshälfte oder weiter zu schieben, mal sehen wie es dann aussieht, ist aber auf jeden fall preislich und vom platz her machbar, der eingesparte strom würde sich auf jeden fall mit den zusätzlichen controllern ausgleichen

you could switch boot in bios to 2nd disk? or leave bios as it is and attach sata cable of 1st disk when boot from 2nd has started, 1st disk will be detected and system can be repaired, after this booting normal should work

no, most ide will not work as the kernel module for this (pata_*.ko) doe not load, synology made changes to the kernel for driver they need and as they dont need pata_*.ko the drivers are not adapted so they will not work no thats a older ARM processor based unit, xpeonology is based on x86 processors (now day's 64bit capable processors)

newer linux system (depending on distribution) might keep a ethX reserved even when card is missing and new cards get e new ethX but the key value for this is the mac address and the mac is still the same as it was set in the kernel config in grub.cfg, same mac, same ethX

that's a little bit more then just "a file", but it's already done, the realtek drivers are part of my extra.lzma https://xpenology.com/forum/topic/9508-driver-extension-jun-102bdsm61x-for-3615xs-3617xs-916/ that's not going to work, its not like windows, the driver has to be compiled specific for the kernel and in our case its not just kernel 3.1.102, its a sysnology modified kernel of this version so it hast to be compiled against the source from synology i wrote a how to about that, but you can just use the extra.lzma from above, no need to build your own driver https://xpenology.com/forum/topic/7187-how-to-build-and-inject-missing-drivers-in-jun-loader-102a/

this video does a good job in explaining it,it should be possible to make any drive external, is just about getting the bit's right, internal drives don't need to be all in a row, you could just define one in the middle as external when you do the bin/hex math correct (but its easier to use the last as external) https://www.youtube.com/watch?v=2PSGAZy7LVQ

je länger ich über 2,5" nachdenke .... der preis pro tb ist nicht so unterschiedlich, 2.5" hat geringen stromverbrauch, wenig gräuschpegel, mehr spindeln und die laufwerke werden evtl. nicht so warm der zusätzlche controller braucht wieder strom aber insgesamt wird sich das nicht viel nehmen nur das es leiser sein wird und mehr kabel hat ich glaube das werde ich mal genauer durchrechnen, die packungsdichte ist bei den 6er drive cages garnicht mal so schlecht

so you still can try installing with intel nic, after install connect both nic's to network, use intel to ssh and check /var/log/dmesg and webgui for 2nd nic, if both nic's are present and getting a dhcp address i would think it will work to remove the intel nic and using the onboard realtek, if the intel works and the realtek does not you can post errors from dmesg and messages about he realtek nic

that was 7 years before kernel 3.10 we are using, i can't see how thats supposed to work netlink is something socket based to communicate between kernel and unserspace if you look in Kconfig in /drivers/scsi/ you will see that activating scsi_transport_fc triggers netlink_scsitransport so i guess its not optional, and there is no gui option for netlink_scsitransport i'm not going to look into this further anyway what do you want with fc drivers, dsm is not build in any way to support fc (afaik), so even if you get the driver to load there will be tons of stuff missing for handling fc depending on whats your goal there might be other systems better suited for this purpose

the scsi_priv.h is included in scsi_transport_fc.c and this contains scsi_priv.h ... /* scsi_netlink.c */ #ifdef CONFIG_SCSI_NETLINK extern void scsi_netlink_init(void); extern void scsi_netlink_exit(void); extern struct sock *scsi_nl_sock; #else static inline void scsi_netlink_init(void) {} static inline void scsi_netlink_exit(void) {} #endif CONFIG_SCSI_NETLINK is set to yes in the kernel config, so it looks ok the scsi_netlink.c (providing the code in question) is not compiled with "make modules" but the scsi_netlink.o is present after a "make bzImage" so its something that needs to be in the kernel but is not part of the kernel synology uses so i guess "FC" can't be used

i'd add two things - i'm not sure you can use a external (removable) disk as backup destination, when using things like rsync (like a backup destination in hyperbackup), but you will see a removable drive in a sbm/cifs network as a share and you can copy data to it (if thats enough) - you can also define a internal sata port as esata, so that dsm is using it as removable disk, that way there is no dsm system on this disk and you can eject/unmount it in the gui like a usb drive and switch it off and on as you like you can use a internal drive cage for this or use a esata backet that is connected to the internal sata port to have a hardware esata connector on the back of your nas (thats what i did to speed up backup) you would have to change the synoinfo.conf for this

hi, und ich spare mir driver cages und kaufe platten die lange halten :-))) (HGST, habe in 4 jahren noch keine tauschen müssen) aber grundlegend hängen gehäuse (und damit formfaktor des boards), design/waf und geräuchpegel direkt zusammen und haben am ende viel einfluss auf die wahl wenn es z.b. in ein bestimmtes regal oder zum hifi passen muss schränkt das die auswahl ein, ich habe damals hotswap zugunsten von tiefe und geräuschpegel gestrichen aber auch die platten größe und menge hat einfluss, 2,5" wären mir zu klein ich denke im moment über 10 oder 12 TB platten nach, das geht nur mit 3,5", außerdem ist die menge an ports auch ein kostenfaktor, kontroller und kabel kosten auch geld z.B. gibts mit dem Silverstone DS380 eins mit 8xhotswap von der stange ansonsten empfehle ich mal bei Streacom, LianLi und Silverstone nachzusehen caseking.de hat einen guten gehäuse-"finder" mit dem man bei festen parametern (höhe, tiefe, ...) ganz gut sondieren kann

hallo, dann hast du vermutlich nicht meine extra.lzma verwendet da ist ein r8152 treibe rmit dabei zum deinem jdownloader problem kann ich aber nichts sagen das ist ziemlich speziell und eigentlich sollte netzwerktreiber bei netzwerkzugriffen von java programmen nicht abrauchen, da liegen mehrere lagen api's/schichten dazwischen

hi, did some testing and there was no problem with the driver in general, it does work/load in the same way is it does for the 3615 one odd thing (also seen with the 3615, did test it when seeing it with 916+) is that at first try after first boot with flash drive (no dsm installed) and realtek on the net, synology assistant showed a apipa address (like seen in your logs), as i was giving my desktop a additional address from this range i could access it the over the web-interface and install dsm (manual, dsm from local file) i also tried installing dsm with a intel nic and after this inserting the realtek as 2nd nic and there was nothing special to notice, just showed up and got a dhcp address, nothing unusual in log, also removing the intel after install was no problem, realtek came up as 1st nic in dsm after this i tried a fresh install without the intel card and only the realtek, with the same loader i used for the intel nic (mac address of intel in grub.cfg) and the dhcp worked as expected, no apipa dress just the same address the intel got before, but i could not reproduce it, after changing the mac in grub.cfg the realtek nic got a normal dhcp address so the driver works, just there seems to be a strange dhcp problem with the boot loader on first install, but when using the synology assistant it shows the correct ip address (apipa range) and when using the apipa address with the browser the install process worked (also did tests with first 6.1(.0) and 6.1.5 as first install but there was no difference, worked as expected) /var/log/dmesg ... [Sun Feb 25 20:31:43 2018] e1000e: Intel(R) PRO/1000 Network Driver - 3.3.4-NAPI [Sun Feb 25 20:31:43 2018] e1000e: Copyright(c) 1999 - 2016 Intel Corporation. [Sun Feb 25 20:31:43 2018] e1000e 0000:00:1f.6: setting latency timer to 64 [Sun Feb 25 20:31:43 2018] e1000e 0000:00:1f.6: Interrupt Throttling Rate (ints/sec) set to dynamic conservative mode [Sun Feb 25 20:31:43 2018] e1000e 0000:00:1f.6: irq 139 for MSI/MSI-X [Sun Feb 25 20:31:43 2018] e1000e 0000:00:1f.6 eth0: registered PHC clock [Sun Feb 25 20:31:43 2018] e1000e 0000:00:1f.6 eth0: (PCI Express:2.5GT/s:Width x1) dc:4a:3e:80:1c:ce [Sun Feb 25 20:31:43 2018] e1000e 0000:00:1f.6 eth0: Intel(R) PRO/1000 Network Connection [Sun Feb 25 20:31:43 2018] e1000e 0000:00:1f.6 eth0: MAC: 12, PHY: 12, PBA No: FFFFFF-0FF ... [Sun Feb 25 20:31:43 2018] r8168 Gigabit Ethernet driver 8.045.08-NAPI loaded [Sun Feb 25 20:31:43 2018] r8168 0000:01:00.0: irq 140 for MSI/MSI-X [Sun Feb 25 20:31:43 2018] r8168: This product is covered by one or more of the following patents: US6,570,884, US6,115, 776, and US6,327,625. [Sun Feb 25 20:31:43 2018] r8168 Copyright (C) 2017 Realtek NIC software team <nicfae@realtek.com> This program comes with ABSOLUTELY NO WARRANTY; for details, please see <http://www.gnu.org/licenses>. This is free software, and you are welcome to redistribute it under certain conditions; see http://www.gnu.org/licenses/>.

damn it and i thought i was off the hook with the drivers

kind of good it happens with jun's extra.lzma too, so i did not messed up anything with my newer driver from realtek 916+8168G_NIC_onboard.txt show that the driver was loaded and a eth0 was created so driver works (mostly) only thing is that it does not get a dhcp adress, its just an APIPA adress 169.254.14.204 it looks the same with the intel (dualport) nic, loads, eth0/eth1 present but only gets a APIPA adress the thing i'm interested in is the /var/log/dmesg from a system that uses intel (or any working nic), was installed with dsm and after this realtek card added (lan connected on both), the dmesg should contain information about the r8168.ko driver loading, finding hardware and maybe commenting about anything thats wrong i'll have to get me a better/newer test hardware for a few days, i guess i will see that it is a intel based that can handle 916+ and i will borrow that realtek pcie card too so i might be able to test that combination the next days myself

we need to be precise about the cips in question as there is a lot of problems with the naming of the driver (as you are using it) so from your discretion its: PC1. RTL8111F PC2. RTL8111H PC3. RTL8111G did you try to use jun's loader without my extra.lzma? also if you installed with a intel nic, what was in the log about the realtek nic? if its not working there should be at least a error message i cant test the 916+ image with baremetal install (just as vm in virtual box), i testet the drivers i used with 3615 image and DN-10130 (RTL8168E based) pcie card an the driver source is the latest http://www.realtek.com.tw/DOWNLOADS/downloadsView.aspx?Langid=1&PNid=13&PFid=5&Level=5&Conn=4&DownTypeID=3 so if you can provide a log i can try to fix it

if quicknicks loader comes out there will be no need for my extra.lzma anymore i guess and if a driver from his collection does not work we can easily use the source i used for my extra.lzma and fix it

hallo, lies doch vieleicht erst mal die faq und das tutorial (hast du vermutlich noch nicht sonst wüssest du das es normal ist das außer kernel laden nichts weiter auf dem bildschirm zu sehen ist), ich habe zwar keine N40 aber aus dem gedächtnis sage ich mal schau nach C1E im tutorial (und vergiss nicht vid/pid des usb sticks, das ist spätestens wichtig wenn man das pat file auf die platte installiert) https://xpenology.com/forum/forum/83-faq-start-here/

not directly you can find a list of drivers in my extra.lzma extension thread, if you open juns's extra.lzma and compare it with the *.ko files (the name in the list are the names of the *.ko files) you can find out what jun supports directly, you can do the same with dsm itself to find out what synology supports directly if you need a 10G driver, you can ask for it in my driver extension thread, as 10gbit is the "new normal" i will try to compile a driver as needed (as done lately for the cheap aquantia cards) if you want to be "future proof" (if something like that is even possible) then you can go with cards/chipsets supported directly by dsm (as i have done in the past, i use BCM57840 based, bnx2x.ko and Tehuti based, tn40xx.ko - but there are more supported by dsm itself)