Need help to upgrade old NAS. Lots of questions.

    • Official Post

    Unfortunately, I had to RMA the motherboard. It happens that the NMVe connector was somehow faulty since the beginning. I don’t know if it’s a mechanical, electronic or logical problem, but this was the reason I thought Netac SSD had issues (and I changed it for no reason).

    Unfortunately, hardware issues are the hardest to find. There's no decent diagnostics software that covers all of the various brands of Motherboards and peripherals, and rarely will OEM software diagnose a specific fault. Assuming the power supply is up to spec (at least the outputs can measured), hardware "guess work" often begins with Motherboard isolation and burn-in software, then adding memory sticks and peripherals one by one. Most of it is off-line and time intensive.

    The above is why, when building a new box, it's best to start with good quality Motherboard from an OEM with a good reputation. (Supermicro, Intel, MSi, ASUS, Gigabyte, etc.) Similarly, a good quality power supply is not wasted money.

  • Unfortunately, hardware issues are the hardest to find. There's no decent diagnostics software that covers all of the various brands of Motherboards and peripherals, and rarely will OEM software diagnose a specific fault. Assuming the power supply is up to spec (at least the outputs can measured), hardware "guess work" often begins with Motherboard isolation and burn-in software, then adding memory sticks and peripherals one by one. Most of it is off-line and time intensive.

    The above is why, when building a new box, it's best to start with good quality Motherboard from an OEM with a good reputation. (Supermicro, Intel, MSi, ASUS, Gigabyte, etc.) Similarly, a good quality power supply is not wasted money.

    At least my PSU, although a bit old and not too strong (360W), is from a good brand (Zalman, SPI Electronics made). Being older than PLUS 80 seal, it lacks this certification, but it fits all the requirements to PLUS 80 White.


    I have two other PSU in the shelf, a Corsair AX650 and a Seventeam ST-750z-AF, that has better PLUS 80 scores, but they are too strong for a NAS, I think. But using a PLUS 80 Gold or Bronze PSU in a machine intended to remain on 24/7 should save me in energy consumption in the long term, besides allowing a better cable management.


    Best regards.

    • Official Post

    my PSU, although a bit old and not too strong (360W), is from a good brand (Zalman, SPI Electronics made)

    For use in a typical home NAS, 360W is fine. That should save you some power as well. (A big PS will use more power, even if there's very little draw.)

    With the AX650 and ST750z, at least you have substitutes if you suspect something is wrong with the 360W.

  • I finally received a new motherboard from the seller to replace the one with the damaged M.2 slot.

    AM-JKLUjGQPwV74_M6jbmo3iZB3jQoNopOOozqWa4Uv9XsXJQ7IQyPqasB4E_veRUGli6cYUls6fUqh5Hxbm7M4AMbHttTvBeU40bmkXCozxFZlGW5P_HfrGXn3a20I7EjvNRNlCCLvS9dQMJ2f4Ebe-aXH8=w1199-h897-no?authuser=0


    The board is identical to the previous one, with the same poorly written manual of just one page.

    AM-JKLVJ8PqpTX0WJJGpEZHxHcw0YT66dpNlI7_TdPONHDPnnBYhuOF-kqjZCDrc-yBzfLgQbpaqUSbaoSwUWEnp2SmZeov1v8Sx4_0XdT1tQiT7dWWnPU3AsejnASxyDZiwFQn-KzVvphTyKu0X6w3PIg-m=w1199-h897-no?authuser=0


    AM-JKLV-x-MLsWbZjtFEN9i5LWq2mtAo5YP9JeINMtu73l4Ysnw4a21PzHoqnPSuLDrNKULa76XlqIF13PuqZ9q9NQ9DpjJtV4H9r2ukzl_E6WMT2ofq7Z_uqvGXBV_58WXXYjp1MIm5-Dj58YyON_xtAgLe=w1199-h897-no?authuser=0


    Before installing the card, I made two small changes in the case (I replaced the standard Cooler Master fan with a pwm Noctua and replaced the blue Power LED housed in the Power Button with a white one). The installation of the board in the case was uneventful. I also tried, with little hope, to use the NVMe driver from the previous install, but it didn't work, so I used a fresh install of the OMV. I haven't configured all the features yet, but so far everything is going well. However, my media player computers still cannot see the Samba share in OMV that contains my audio files. Very odd.

    AM-JKLUpmlvI0r7-h6-LP38kVHyTzZ2RA5mp1D_aWUa5-4buFzxOUmORpguHFXtURL0fDiFRrets1OdcBbuK2-Ded9mLjhZ8udYjKEVloYYGz0TXiVeIzCiOXdFQGcwZp61yncoOS3oBYMzA65KuEs-juuaf=w1199-h897-no?authuser=0


    AM-JKLVOgbaeGbK9xEr0iuqnlOEXk-CaiJncrGdzFLlmR1IvZ6eIJY2Tzz7Gtb50bm_sAsp98RlSk46QOsp_DRnnUkNljzQCoe2TV1F4txvNk3S1BP4Xp3OVcbyWoupY_5laJJmE8fobP0ShyMpDEiGGPhV2=w1199-h897-no?authuser=0


    Best regards

  • Well, before continuing to install furter resources , I must fix the main reason for building this NAS: somehow my media player computers are locked out from the very shares they must tap.


    At first, I thinking that some debian updates (both machines were running Strecth) were preventing samba clients to connect via unsecure protocols (both former NAS use NT1 version). But, after struggle to build this OMV box AND install/use openmediavault-resetperms plugin, mpd still can't access my media share!


    Yesterday, after install the plugin and first run it, I spent almost 4 hours trying to make one of the media computers to update the mpd database and play music, with no success. The only way I could produce some effect was by using BubbleUPnP as a proxy between the NAS and the media computer. That way, I know the mpd is functioning and that there is something wrong with Samba (server or clients). Today, as a last effort, I upgraded one of the computers to Debian Bullseye. The only thing thar I accomplished was that now, mpd database is totally empty. By sshing as the user I created to play music into the machine, I can use smbclient and connect and browse through the media share, but the mpd still can't do a thing.


    Fact is, I think I reached my incompetence point. There is not a thing that I can think of to diagnose and solve the problem. I need help, please.


    Here are some pics:


    AM-JKLVjVjxzZ4mQdWmCvLzuzfBFo0kazgvqZG37_vU6oNRBLLem77GHi1lELKh7nxMhM0GMITdFmaAmeiLF3WzAdH3WFK1dOKHnZIXwEV9ax2JFWBq8Xiut6TT6vYXAlU9C5ju1OULb4ufo0ZhBWwjk6Nmk=w1157-h897-no?authuser=0

    The windows MPD client from one of the media computers showing its former database.


    AM-JKLWr-gQBCPX6qhF51rJ5tGj024gSINak2Srj3cUDrmer3LuMbgN6kecAPQntiKLOhk167AgOfyWq0D4RPzcUPf5S7Dp89iLbn4oAB0jQNNZAUikwS8R6cy-XKcONyyyWiZVgpx9URCvzuuILHuhBp9_9=w1151-h897-no?authuser=0

    And the permission error.


    AM-JKLUYs3jhtbrZd4f46NDpNC4WqUB4cJol3kJwbno3FJ5D6JCDVrT8yxjt-rTCsPhdT0qlt2tscViPqC1LTfYQhQFktXNgN6E95vI8kfdQQTubLAgjOWlEiMsmSl1Wg1594ceVC_AaX1CPhiL-AO61GiOH=w565-h897-no?authuser=0

    Some info about the media computer. I was able to navigate though the share until the point a .dsd file must be shown.


    AM-JKLUStPoe1nk3ehLXWNHiFr-i4GD6qPeAEwbcks_8XhIzUyBwk-rPkXDJ6xChkLimPeJvL4W0BEmDXplyyhP44UWUKEN0IYos6JjhJyPzAIj7q939qw7BCKUxPOXNSNzynqKC-jGBN3RZPglCabN3FU7H=w1156-h897-no?authuser=0

    The same picture of the MPD client from the Bullseye updated computer. The database is blank.


    AM-JKLV-RcpXZZk1Q9yOyU78u42ewr6-V28q-eFehAPeXcEWik72B4R_FPuTcJTX_QWIlu6xJ83YsRc_-8GXe1r8uQFsn1fmIZCQVB90FGGCS0MshKTOp2SfC4oIJ5oWcR9I33jiXxkRPARAXTNwE9SgPUnO=w1157-h897-no?authuser=0

    Despite the void database, the list of recent files is available, but the files unreacheable.


    AM-JKLVTFNRM8YexhN8qVzzprBvNmgbpMzoQKrbrzf3dJQ0UburOp8SyfAN5saWdqlY9Qbzsh86PwpIX1LUt1BIdbbBE-FvGB1VEIoYjqUqozibiVHgXodWrXqFzPmSW87Tr32sQsrbwr4146BcXxcCBQlUq=w791-h878-no?authuser=0

    Info from the Bullseye computer. Same behaviour tho.



    Best regards.

  • Today, with all data transferred to OMV, it was time to gut the old NAS for its drives.

    AM-JKLXk8UVv6ym_gmC8Qja8-p4hL1Ry811yabk7I9iT26sdSVaP6WdRlWw77QN1MKZuolUQthA1PRkqAhDDZzt1hHqh9SSaC2A65jo6E2PKLSp-eHcImllnIPn_LpxQGt-cNp3-135BOP6eCAraMxpFOwVi=w1199-h897-no?authuser=0


    AM-JKLUQFqaBs6NGuQ1ZK9q8kXyO4D4cDAbxnld7EkpUE_0MZoozT0XSFKoaRuh7AyTFTPHN3SHmozIS-YkP9b1URG_WJ4iGlxvZp86UmyJzn8beIq6rDvs_3EIopzvuITQK3QmCoQ_EBUPv_1yCho3LK52l=w1199-h897-no?authuser=0


    The one on the right will go to the OMV server and the other were supposed to replace the external HDD that suffered the needle noise thing.

    AM-JKLWDninJxWczeYkTha0hgEtfAmYrViiFKriZPzYX1Op9fmjVMh9CIFTPatp2UkP7avAQI-BzWtRlONY2ecqcMFDVzrmA_h35esxLbJOkx9LtZF-MssSD-hNBriN0U3yoSWsFVz_p2CKuEWgFRsUdKr_B=w1199-h897-no?authuser=0


    This choice was made for they are from the same model.

    AM-JKLXorFOmrRTcZbRw_PYgkNPRAYX0Uuk93AjRlhBPQPXLbNoPmyoDspok1iCMOLB5YXPenA7HVfyQkFRU_ihGbuk8BsCvbmDXXOBPqeGYo7kXbJdbvyTirj4WbmJlDSMnwbK-RJx0nBiU1iHS6I7qADwi=w1199-h897-no?authuser=0


    I plugged both, one at a time, in a Debian computer to wipe they partitions clean.

    AM-JKLXaUv_xNkB22qSo3_n58mWWoAZJpWBqXN2IjPR68TrAVEh3C--bJsWjm3__HHq_8X8Z0om2gyZB-nQpdVE5VkzbuOfnc1-ejFYWX_AKBbNeF-bKb568_gHJ4gFIg_kyS4FZrd5m2ehgSPZDqZrRB1Eo=w1199-h897-no?authuser=0


    AM-JKLW--Av5DiuVOoANE0-tNSrKRKzkp8RBwE6rr0BpLTy3aJlTTUrruWQrUlRw-Q997dLkTYzWkNeHJDupwbGnRjXXUAmHU54CJ71pKOUoTBlntEp7tYkwuIGHmQLGkAW5Y1YafgxPZTmi7ZhvheF95Kdp=w1199-h897-no?authuser=0


    After installing the drive in its place, I added a case lock and powered it up. At this moment, it is being wiped (again) in OMV.

    AM-JKLUpkkyuJTPWNbVcksZsvrJR0UHGxWEJxDuTVg0Vwf1EHxt1qU8JrjlWdSRvLoakmvIepdnDZXU6Pi7lso-J6aR730S4qHZxr4vwzD6xu7esnQDvMWz0gDkVSoZACN2Mu6o4P6vZtBhI2LuLNcIOt703=w1199-h897-no?authuser=0


    AM-JKLUhKrxzGmayHJ4I1apR_8KCK7x_U7dsHiwoUBgCYndLBkoMlumsanBAvs5UTT9sT3tlrxnnwNjqACT3-cGB8UTEIbodI2xFYI4qUFSQIe9klJkuWlwtGi93MPtiLeLuUznt_EXEwVKe6EPgvRz1tlEF=w1199-h897-no?authuser=0


    AM-JKLW7glqUqGF2BEfJzk-CyAE6ZhnbsK0SNrPmCuATK2uXgzT4EYeT0lXZ0uS-_9rHXp7Ol_HLCr0SjZrssfSro7E7B6D_hIj_B_80-dE3gxazK1MEkxSuzTWaF0xuly3GeCrhW-uLHDjjeOl39Z82G4_y=w1199-h897-no?authuser=0


    Now, the oddest thing: I was about to throw the old external 4TB drive in the bin, the one with the needle noise… But the Debian computer was laying there, already opened, just waiting for a last effort try. So, I plugged the drive and it worked! I deleted its partition, created a new one and putted it in its case. I don’t know how reliable it is (I think not much), but I may have another drive to escalate the NAS size.

    AM-JKLWrV28dYIEQ4bBgjSx6zNSD0hgR-4K_sVKHBRbhAYQkfJ4GKgaE3FvLaRSOHrafj3PKkUHKyl66zBn_bucjwY8iKhfWIem2cu8jOUI5lNAzsTer7F5trEK4ZVUy0FV3QM6kQl5yyDZfIYkFe-g1_NXh=w1199-h897-no?authuser=0


    Best regards

  • Well, since I got no help on post #88, I will probably open a new tread in Samba subforum.


    But I have an issue in another matter: yesterday I installed Nextcloud again following macom's tutorial. I edited docker-compose.yml as required, pointing the same paths for the Nextcloud, MariaDB and Swag volumes that I had used in the previous OMV setup (the one that had the motherboard RMAed). Therefore, these paths and the according files already existed in the file system. I paid no second thought on that, assuming the files was about to be overwrited.


    When I started the containers, I saw several lines in the CLI warning about preexistence of files ("<filename>: Already exists"). After that, I runned "docker logs -f swag" and got the same error I got before, as reported in post #71 when I assumed the error was due an ISP malfunction.


    The thing is: I don't know if this error is from the already present old files from the previous install, OR decorrent from a recurrent ISP issue, OR related to something new (or not perceived yet).


    Anyone have any clue on this matter? Here is the result from "docker logs -f swag":



    Best Regards.

  • The message is, letsencrypt can't find your server to validate the certificate.

    What is on http://www.#######.tplinkdns.com



    Use ping www.#######.tplinkdns.com

    or nslookup www.#######.tplinkdns.com


    and compare to the result of http://checkip.dyndns.org this page

    If you got help in the forum and want to give something back to the project click here (omv) or here (scroll down) (plugins) and write up your solution for others.

  • That is odd. I have a Home Assistant running (in a diferent port, of course) witch is accessed by the same address, and it is working all rigth! And both the ports 443 and 80 are forwarded to the internal OMV IP.


    Best regards.

  • Ok, I was able to remotely access my PC at home. I SSHed the OMV machine and pinged www.#######.tplink.com and the host could not be found. BUT, when I droped the "www", voilà! It pinged. Then, shame on me, I tried to nslookup the same link, but this particular command is not available in this setup. But exiting the SSH connection, my windows machine could acomplish it.


    But when I do the same from my Windows machine at work, I got diferent results:


    What can I do to solve this?


    Best regards.

  • You have to fix dns entries with tplinkdns.com. you subdomains are not resolved. Does tplink provide that service?

    If you got help in the forum and want to give something back to the project click here (omv) or here (scroll down) (plugins) and write up your solution for others.

  • It seems #######.tplinkdns.com is already a subdomain.

    Quote

    name "#######.tplinkdns.com" is subdomain, public suffix is ".com", top-level-domain is ".com", top-level-domain-type is "generic", tld-manager is "VeriSign Global Registry Services", num .com-domains preloaded: 63591 (complete: 175327)


    Best regards.

  • But you are trying to have sub-subdomains www.#######.tplinkdns.com. Each name needs a mapping to the IP address.

    If you can not have an dns entry like www.#####.tplinkdns.com or *.######.tplinkdns.com you will have to use a different dns provider or use urls

    like #########.tplinkdns.com/service

    If you got help in the forum and want to give something back to the project click here (omv) or here (scroll down) (plugins) and write up your solution for others.

  • No. Tplinkdns is a DDNS provider that comes with TP-Link routers. It is pretty decent but, it provides no more functionality than DuckDNS or NO-IP free plan.

Participate now!

Don’t have an account yet? Register yourself now and be a part of our community!