• Content count

  • Joined

  • Last visited

  • Days Won


IG-88 last won the day on December 6

IG-88 had the most liked content!

Community Reputation

83 Excellent


About IG-88

  • Rank
  1. Issues with DSM 6.1.3 and TerraMaster F2-220

    install that package, it does need a additional kernel module and some scripts (the synology power button is not acpi) i guess some additional vendor specific micro controller board managing this? there is a good article about hat on the synology faq, can be plugins (like vmm, ...) or other activities synology has its own custom stuff, fan control does not work ootb, usualy it work like in bios predefined no, just the three version, every additional version would need to be individually maintained (like specific hack and drivers) jun did provide the three and thats it for now (some people seem to try to get into the 918+?)
  2. that sounds more like a r&d project and a appliance nas like dsm/xpenology is not the best choice if it not clear how that will be done in detail i would do it on a normal linux with all the freedom it offers after finding out whats needed and how to configure it, it has to be compiled for the destination platform and then the decision to create spk packages (more universal) or just *.dep and install manually and then using cable and analog for distribution over 4 rooms no, that's so last century but we are getting off topic here, driver might be the easiest part maybe and thats it back to business hpsa does not work for p400 at all so i cant test and cciss (4.6.28-22) driver did not compile, loads of errors and i cant see whats wrong so the alternative for up to p420 is also a no go (even the cciss driver from the kernel did not recognised a raid0 array disk on the controller)
  3. Anleitung DSM 6.x?

    klemm doch einfach die platten ab und arbeite mit einer einzelnen platte (die nicht benutzt wird) am hp controller, du willst doch nur sehen ob die platte am controller gefunden wird und ob man sie als basic disk in dsm einbinden kann (aka drauf schreiben) theoretisch sollte es reichen den bootloader von stick und eine platte am smart array und booten, wenn man dsm installiren kann geht der controller und wenn er keine platten findet wars nix ging mit übrigens am wochenende so, der p400 ist ungeeignet als testobjekt denn er wird vom hpsa treiber nicht unterstützt, nur vom alten cciss treiber wenn man ein laufendes dsm auf einer platte hat und den hp als zusätzlichen controller benutzt kann man leicht im log nachsehen was beim laden passiert (bei mir konnt man beim hpsa treiber deutlich lesen das die controller id nicht unterstützt wird) du musst nicht wild rumprobieren, die img files kann man mit 7zip öffnen und im 2. part liegt die extra.lzma, die kann man auch mit 7zip öffnen (einfach weiter klicken) darin findet sich dann /usr/lib/modules und wenn es da keine hpsa.ko oder cciss.ko (alter treiber bis p420?) gibt dann wirds nichts weil kein treiber bei quicknick ist der aufbau ähnlich d ist es die 0.img, //image/ds3615xs/drivers.lzma und da gibt es eine hpsa.ko (bei jun ist die nicht dabei)
  4. it does not have to be on windows 10, you also can use a chroot enviroment on the nas or in a vm even if you get the driver to work, you also have to create the alsa stuff (does not look like as there is enough in dsm like alsactl or alsamixer, aplay, ...) and you will have to compile and install shairpoint-sync, also configuring and using that without a gui(?) or is there a plugin doing this multiroom over alsa?
  5. yes serial cable is needed to see it during boot but if the system can be reached in network and has a working disk with dsm (like on sata ahci), ssh should be enough to get the logs (hp controller only as "additional" with one disk for testing), might be easier then just one disk on hp controller and trying to install dsm on this disk
  6. im using the latest version available 15152, thats closer to the destination we want to use (6.1.3/6.1.4) but for testing 15047 will be ok, also juns loader uses kernel from dsm 15047 as i was reading the hpsa driver doku again i found out my p400 is not supported so good thing if you can test i will go for the older cciss driver that supports only up to p420 but works for my p400 and i did not test it the last time(?), also all questions about p4xx where p410 and p420 so if cciss driver work most people will be happy i geuss
  7. Anleitung DSM 6.x?

    und hast du mal 6.0.x mit quicknick's probiert? davon abgesehen wollte ich dieses wochenende noch mal für die hp controller anlauf nehmen, ich habe in der reste kiste auf arbeit einen p400 gefunden und mal in einen pc eingrbaut den ich noch rumstehen habe, so kann ich schnell testen ob ein test treiber beim laden mit vorhandener hardware abstürzt und ob er dann auch platten findet - aber keine versperechen das dabei irgendwas rauskommt, ist ein reines spaß projekt, ich brauche das nicht für mich ist nur weil es (unverständlicherweise) so häufig nachgefragt wird (ein raid0 pro platte zu definieren ist ziemlich unschön, würde ich nicht machen - ab p420 scheint es aber einen per software aktivierbaren hba mode zu geben, das macht eher sinn, hat aber mit dem default treiber von kernel 3.10.102 auch nicht funktioniert, also mal sehen was der alternative treiber bringt falls ich ihn zum laufen bekomme)
  8. Atheros Alt1c.ko

    thanks i will replace it in all the versions with a freshly compiled one in the 4.2 version (planed for this evening)
  9. nice coincidence, a few weeks ago i've given up on that but last week i stumbled across a old p400 at work (in the recycling bin) as i still have a old pc for data recovery with ide, sata, pci and pcie i can now test the hpsa.ko driver myself main problem was that the compile switches the make/compiling process uses by finding out on what driver it is compiles (3.10.102) did not work, there are some switches 0/1 that are set according to kernel version and my guess is that some of the back porting/modification synology did does interfere with that simple kernel detection (it does not check the functions in kernel source, just the kernel number and does extra checks for suse and red hat by version number) the source is from here http://cciss.sourceforge.net the switches are in hpsa_kernel_compat.h kernel class comes from Makefile, it will be 3B for kernel 3.10.102 to test changing these switches one can tinker in the 3B section starting with #elif defined(KCLASS3B)/************ Kernel Class 3B ************/ the first two in the default seem to collide with "our" kernel with the set default to 0 (indicated by collision warnings wehen compiling with the default settings) #define KFEATURE_HAS_WAIT_FOR_COMPLETION_IO_TIMEOUT 1 #define KFEATURE_HAS_WAIT_FOR_COMPLETION_IO 1 if someone else finds out under what conditions it does deliver a working driver i would redo it with this settings and hpsa,ko would be in so if you have time and its important for you feel free to try to build a working driver about possible alternatives, imho open media vault is more like dsm/xpenology
  10. Anleitung DSM 6.x?

    erst mal 6.x.x ist zu unspezifisch, 6.0.x und 6.1.x brauchen absolut unterschiedliche bootloader und entsprechend auch treiber (unterschiedlicher linux kernel) soweit ich das sehen kann braucht der P212 den gleichen treiber wie der P410, das wäre hpsa.ko und den habe wir bisher für 6.1 nicht zum laufen bekommen mit 6.0.x und quicknicks loader müsste es imho gehen davon abgesehn frage ich mich wie du da 4 weitere platten in das gehäuse bekommen hast, sind die microserver nicht recht begrenzt? gibt das nicht probleme mit der stromversorgung?
  11. mac and serial are not important when it comes to installing dsm even the vi/pid is not needed to "see" the booted loader in network (if it works), there are a lot of people that forgot to change vid/pid and they usually failed when trying to install the dsm file (implication with that is the booted system from usb was "seen" in network before) his network connection and dhcp server looks ok, he posted a picture of the system running windows and there where valid dhcp settings like ip, dns and defaul gateway, also it clear from that picture that it is a realtek nic the thread started with a working 1.02a usb stick, that one should at least work and when its not available a newly created usb stick with that version is worth a try, best if you still have the old file you used to create it
  12. Setup unsupported NVMe Cache

    if you overwrite files of your dsm the self check on next boot will recognize it and warn you, for nor you just copied files that where not present and where not checked you can make copies before doing it and if it does not boot anymore or something like that its also possible to reinstall or start a live linux from usb and undo he changes with the copies you made before as long as its not a completely synology written part then it should be source available to compile for the syste in question in the same way the kernel drivers are compiled, that would be the clean way so first step is about to find out what process is reading the synoinfo.conf and what that process does with the variable "supportnvme" it its set to yes copying stuff and hope something will happen has not the best chance to work, best way is to understand how its implemented and then try to mimic it
  13. Atheros Alt1c.ko

    on what loader of the three hardware types did you test? whats the latest source for you? the actual synology kernel source 15152 or a source from external like atheros? the module in question is provided by jun, i only exchange them if there are reported problems like the ixgbe.ko with the new 4.2 extra.lzma if its just the 15152 from synology i already have the modules as they are in the synology default config when compiling, its just a few seconds to make them part of the 4.2 release and you can test it then
  14. did you tried without additional extra.lzma, just the 1.0b as it comes? if you still have the "original" 1.02a, that one should at least bring the system to the point where synology assistant finds it in network
  15. installation problems (can't find server)

    if its PATA/IDE then dsm 6.1 is not a good choice as the pata_* drivers are still broken (same as the sata_*) depends on you point of view, afair the largest ide drives where 750GB and thats more a "small" for a NAS