Jump to content
XPEnology Community

s3urat

Member
 View Profile  See their activity

  • Posts

    24

  • Joined

  • Last visited

 Content Type 

Posts posted by s3urat

    Gen 8 Microserver

    in DSM 7.x

    Posted · Edited by s3urat

    I've done some digging, since the same problem manifests in Ubuntu (both server and desktop).

     

    My guess is that the problem/incompatibility is the same as this problem in Unraid. If I disable IOMMU in the BIOS or add amd_iommu=off to the Ubuntu GRUB config (amd_iommu=pt doesn't work!) the disks are detected just fine. Otherwise theyre not.

    Gen 8 Microserver

    in DSM 7.x

    Posted

    So, the output of ./rploader.sh satamap now:

      

    tc@box:~$ ./rploader.sh satamap now

Found "00:11.0 Advanced Micro Devices, Inc. [AMD] FCH SATA Controller [AHCI mode] (rev 49)"
Detected 1 ports/1 drives. Override # of ports or ENTER to accept <1>

Found "01:00.0 Marvell Technology Group Ltd. 88SE9230 PCIe SATA 6Gb/s Controller (rev 11)"
Detected 8 ports/0 drives. Override # of ports or ENTER to accept <8>

Computed settings:
SataPortMap=18
DiskIdxMap=0001

Should i update the user_config.json with these values ? [Yy/Nn] y
Done.
tc@box:~$

     

    The user_config.json after the build:

      

    tc@box:~$ cat user_config.json
{
  "general": {
    "model": "DS3622xs+",
    "version": "7.2.0-64570",
    "smallfixnumber": "0",
    "zimghash": "a4af69c2097b18a5a370768431cf5d84067eaf41833c45088834ae527e5d5da5",
    "rdhash": "a0c580a2a5af1bb5a6cc475f4dc06bd2c8ae94a45e5c1badc2dc62cff3f75f34",
    "usb_line": "withefi earlyprintk syno_hw_version=DS3622xs+ console=ttyS0,115200n8 netif_num=1 pid=0x5591 earlycon=uart8250,io,0x3f8,115200n8 syno_port_thaw=1 mac1=0011323BBDCD sn=2150SQRDJYYBT vid=0x0781 elevator=elevator loglevel=15 HddHotplug=0 DiskIdxMap=0001 syno_hdd_detect=0 vender_format_version=2 syno_hdd_powerup_seq=0 log_buf_len=32M root=/dev/md0 SataPortMap=18 ",
    "sata_line": "withefi earlyprintk syno_hw_version=DS3622xs+ console=ttyS0,115200n8 netif_num=1 pid=0x5591 earlycon=uart8250,io,0x3f8,115200n8 synoboot_satadom=1 syno_port_thaw=1 mac1=0011323BBDCD sn=2150SQRDJYYBT vid=0x0781 elevator=elevator loglevel=15 HddHotplug=0 DiskIdxMap=0001 syno_hdd_detect=0 vender_format_version=2 syno_hdd_powerup_seq=0 log_buf_len=32M root=/dev/md0 SataPortMap=18 ",
    "redpillmake": "prod",
    "friendautoupd": "true",
    "staticboot": "false",
    "hidesensitive": "false"
  },
  "ipsettings": {
    "ipset": "",
    "ipaddr": "",
    "ipgw": "",
    "ipdns": "",
    "ipproxy": ""
  },
  "extra_cmdline": {
    "pid": "0x5591",
    "vid": "0x0781",
    "sn": "2150SQRDJYYBT",
    "mac1": "0011323BBDCD",
    "netif_num": "1",
    "SataPortMap": "18",
    "DiskIdxMap": "0001"
  },
  "synoinfo": {
    "internalportcfg": "0xffff",
    "maxdisks": "16",
    "support_bde_internal_10g": "no",
    "support_disk_compatibility": "no",
    "support_memory_compatibility": "no",
    "maxlanport": "8",
    "netif_seq": "0 1 2 3 4 5 6 7"
  },
  "ramdisk_copy": {}
}
tc@box:~$

     

    If I select, in GRUB, "RedPill DS3622xs+ 7.2.0-64570 (USB, Verbose)" I get stuck in a recovery loop - I can connect to the NAS, it offers to recover, reboots fairly quickly then back to step one.

    If I select "Tiny Core Friend" then I get the "No drives detected in DS3622xs+" error.

    Gen 8 Microserver

    in DSM 7.x

    Posted

    I did just that. It correctly detects 1 port and 1 disk on the first controller (HP? the one meant for the DVD drive), and 8 ports and no disks on the Marvell controller. Weird.

     

    Gen 8 Microserver

    in DSM 7.x

    Posted

    On 6/13/2023 at 2:00 AM, djvas335 said:

    Latest tinycore with dsm 7.2 ds3622xs works with no issues on microserver gen7, gen8, gen10 and gen10 plus, I have tested those myself.

    How?

    I can't get anything past a fairly old TCRP version working on my Microserver Gen10. It just doesn't detect any disks connected to the Marvell controller.

    Latest Tinycore Redpill or ARPL on HP Microserver GEN10

    in AMD

    Posted

    Anyone got either TCRP 0.9.2.9 or ARPL 1.0-beta2 working on a HP Microserver GEN10?

     

    I'm currently on TCRP 0.8.0.5 and DSM 7.1-42661 Update 4. I was trying to update TCRP (or switch to ARPL) before upgrading to DSM 7.1.1-42962.

     

    While both loaders build and boot OK (meaning I can reach the NAS over the network), neither one detect any drive connected to the Marvell 88SE92xx controller - DSM complains that no drives are connected. The SSD connected to the other SATA adapter, meant for a CD/DVD drive, gets detected OK.

     

    TCRP 0.9.2.9 reports no drives connected to the Marvell controller when running ./rploader.sh satamap now. ARPL does the same in it's equivalent step.

    DSM 7.1-42661

    in DSM Updates Reporting

    Posted

    - Outcome of the update: SUCCESSFUL

    - DSM version prior update:  DSM 6.2.3-25426 Update 3 w/ Jun's Loader v1.03b DS3615XS

    - Loader version and model: TinyCore RedPill 0.8.0.0 DS3615XS

    - Using custom extra.lzma: NO

    - Installation type: BAREMETAL HPE Microserver Gen10, using integrated NIC

     

    Upgrade cliff notes:

     

    -1. Backed up all important data

    0. Disconnected all the drives (probably not needed)

    1. Downloaded the latest tinycore-redpill.vx.x.x.img.gz from https://github.com/pocopico/tinycore-redpill

    2. extracted the .img file, burned it to USB via BalenaEtcher

    3. boot the server from new USB pen drive

    4. SSH to TinyCore

    5. ./rploader.sh update now
    6. ./rploader.sh fullupgrade now
    7. ./rploader.sh serialgen DS3615xs
    7. a. I added the serial from the original DSM-6.2.3 and both MAC addresses to user_config.json manually
    8. ./rploader.sh identifyusb now
    9. ./rploader.sh satamap now
    10. ./rploader.sh ext bromolow-7.0.1-42218 auto
    11. ./rploader.sh build bromolow-7.0.1-42218
    12. ./rploader.sh clean now
    13. ./rploader.sh build bromolow-7.1.0-42661
    14. ./rploader.sh clean now
    15. rm -rf /mnt/sdb3/auxfiles
    16. rm -rf /home/tc/custom-module
    17. ./rploader.sh backup now

    18. shutdown

    19. reconnect the drives and power up

    20. restore 7.1-42661 (with Update 1) with a .pat file downloaded from https://www.synology.com/en-us/support/download

    22. wait for DSM migration and package upgrade to finish

    23. manually upgrade to 7.1-42661 Update 2 via the control panel

    24. reboot into TinyCore

    25. ./rploader.sh update

    26. ./rploader.sh postupdate bromolow-7.1.0-42661

    27. sudo reboot

     

    All Synology packages get upgraded OK and seem to be working, including Docker and ActiveBackup (it also remains activated!). Some old packages need to be removed, as they're incompatible with DSM7, some packages need to be repaired. The power button package is not needed anymore.

     

    Stuff I needed to recreate/reapply:

    • /usr/syno/synosdk/texts/enu/events gets reset. If you use it to write custom messages to syslog you need to re-add your custom templates.
    • re-run the script to free ports 80/443 and reboot.
    • re-apply changes for having static IPs in OpenVPN for users (https://www.wundertech.net/setup-a-static-ip-address-for-openvpn-clients-on-your-synology-nas/)
      • only /volume1/@appstore/VPNCenter/etc/openvpn/radiusplugin.cnf got reverted to default, all the other changes remained.
    • acme.sh (https://github.com/acmesh-official/acme.sh/wiki/Synology-NAS-Guide) seemed to be working, including forced certificate renewal and deployment via --deploy-hook synology_dsm, but running /usr/local/share/acme.sh/acme.sh --force --upgrade --nocron --home /usr/local/share/acme.sh probably doesn't hurt.
    • the Portainer container needed to be recreated but Docker volumes and networks created with it remained and are working.
    • Nginx Proxy Manager started working once ports 80 and 443 were freeed.

    DSM 6.2.3-25426 Update 3

    in DSM Updates Reporting

    Posted

    - Outcome of the update:  SUCCESSFUL

    - DSM version prior update: DSM 6.2.3-25426 Update 2

    - Loader version and model: Jun's Loader v1.03b DS3615XS

    - Using custom extra.lzma: No

    - Installation type: BAREMETAL - HP ProLiant MicroServer GEN 10

    - Additional comment: Using integrated NIC. Upgraded via DSM, reboot required.

    DSM 6.2.3-25426

    in DSM Updates Reporting

    Posted

    - Outcome of the update:  SUCCESSFUL

    - DSM version prior update: DSM 6.2.2-24922 Update 4

    - Loader version and model: Jun's Loader v1.03b - DS3615xs

    - Using custom extra.lzma: NO

    - Installation type: Baremetal - HP Microserver GEN10

    - Additional comments:

    DSM 6.2.2-24922U4 used a custom extra.lzma with Jun's Loader 1.03b (internal NICs weren't working without it)

    1. uploaded the stock extra.lzma, rd.gz and zImage from Jun's Loader 1.03b to the server via SCP
    2. SSH to the server
    3. mount /dev/synoboot2
    4. copy the stock extra.lzma rd.gz and zImage to /dev/synoboot2, replacing the custom files
    5. unmount /dev/synoboot2
    6. performed a manual DSM update via the Control Panel and reboot
    • Like 1

    hp microserver gen10 + hp nc360t + dsm 6.2.2-24922 = unstable?

    in AMD

    Posted

    On 1/4/2020 at 2:42 PM, beniusc said:

    Hi

    same problem of instability with the intel Nc360T, i modify the .cfg with the instructions.

     

    now i think it’s ok because no more reboot in long transfert with lan since 2 hours.

     

    thanks for the solution.

     

    ben

    Just a heads up: for me the problems reoccured after a day or two and I downgraded to 6.1. Perhaps I have a defective network card or perhaps it is an incompatibility with the GEN10 Microservers, since it appears to work with the N54L variants, I don't know. It would be nice if a GEN10 owner would chime in :).

    hp microserver gen10 + hp nc360t + dsm 6.2.2-24922 = unstable?

    in AMD

    Posted

    On 11/16/2019 at 3:39 PM, Benson said:

     

    Could you show me the detail config. from grub.conf about common_args_3615 ?  i dont know whick postion should add pcie_aspm=off to common_args_3615 

     

    You just type pcie_aspm=off (or maybe pcie_aspm=0? I couldn't find much useful information) at the end of set common_args_3615='...'

     

    Something like this:

    set common_args_3615='disable_mtrr_trim syno_hdd_powerup_seq=0 HddHotplug=0 syno_hw_version=DS3615xs vender_format_version=2 console=ttyS0,115200n8 withefi elevator=elevator quiet syno_port_thaw=1 pcie_aspm=off'

     

    Note that it didn't do anything for the problems I had.

    Unstable network on DSM 6.2.2-24922

    in Answered Questions

    Posted

    I'll repost the question here, since I didn't get any info on the post installation subforum.

     

    Anyone else experienced this?

     

    I've got an HP Microserver gen10, an HP NC360T NIC and DSM 6.2.2-24922. Ever since the upgrade from 6.1 to 6.2 the system will drop of off the network about once a week. Sometimes when adding large torrents to the download station (but maybe that's just a coincidence).

     

    Apparently it happens because the network card stops/resets and the dhcp-client can't (or won't) start again.

     

    Relevant portion from /var/log/messages: 

    2019-05-15T13:23:52+02:00 diskstation [228138.340179] init: dhcp-client (eth0) main process (8937) killed by TERM signal
2019-05-15T13:23:52+02:00 diskstation dhcp-client: stopped on eth0
2019-05-15T13:23:52+02:00 diskstation dhcp-client: started on eth0
2019-05-15T13:23:52+02:00 diskstation synonetd: net_default_gateway_set.c:34 failed to set default gateway

    Found in /var/log/kern.log: 

    2019-05-15T17:47:58+02:00 diskstation kernel: [15716.768357] e1000e 0000:02:00.0 eth0: Reset adapter unexpectedly

    After some Googling (especially this discussion), the solution would seem to be an upgrade of the e1000e drivers and/or disabling ASPM (PCIe Active State Power Management). I've added "pcie_aspm=off" to common_args_3615 in my grub.cfg but it didn't help. I've since downgraded to DSM 6.1 but the problem still bugs me.

    hp microserver gen10 + hp nc360t + dsm 6.2.2-24922 = unstable?

    in AMD

    Posted

    So, apparently it happens because the network card stops/resets and the dhcp-client can't (or won't) start again.

     

    Relevant portion from /var/log/messages:

      

    2019-05-15T13:23:52+02:00 diskstation [228138.340179] init: dhcp-client (eth0) main process (8937) killed by TERM signal
2019-05-15T13:23:52+02:00 diskstation dhcp-client: stopped on eth0
2019-05-15T13:23:52+02:00 diskstation dhcp-client: started on eth0
2019-05-15T13:23:52+02:00 diskstation synonetd: net_default_gateway_set.c:34 failed to set default gateway

     

    Found in /var/log/kern.log:

      

    2019-05-15T17:47:58+02:00 diskstation kernel: [15716.768357] e1000e 0000:02:00.0 eth0: Reset adapter unexpectedly

     

    After some Googling (especially this discussion), the solution would seem to be an upgrade of the e1000e drivers and/or disabling ASPM (PCIe Active State Power Management). I've added "pcie_aspm=off" to common_args_3615 in my grub.cfg. We'll see how it goes.

    hp microserver gen10 + hp nc360t + dsm 6.2.2-24922 = unstable?

    in AMD

    Posted

    Anyone else experienced this?

    I've got an HP Microserver gen10, an HP NC360T NIC and DSM 6.2.2-24922. Ever since the upgrade from 6.1 to 6.2 the system will drop of off the network about once a week. Sometimes when adding large torrents to the download station (but maybe that's only coincidence).

     

    Is there any way to diagnose this? Any system logs besides the Log Center?

     

    The only abnormality I'm able to see in the Log Center is that LAN 1 goes down and receives a 169.254.xxx.xxx IP, like the cable was disconnected. It stays like that indefinely.

    DSM 6.2.2-24922

    in DSM Updates Reporting

    Posted

    - Outcome of the installation/update: SUCCESSFUL

    - DSM version prior update: DSM 6.2.1-23824U6

    - Loader version and model: Jun's loader v1.03b, DS3615xs

    - Using custom extra.lzma: NO

    - Installation type: BAREMETAL -  HP MicroServer Gen10

    - Additional comments: Internal NIC disabled, added a HP NC360T dual port NIC.

×
×
  • Create New...