rebooted still not working.
i think the issue is the other possible reasons
would be useful if we had another ts-877 user for their input on this 
Did you try sudo chmod etc? Depends if your logged in as admin or root.
Edit: I see you did.
Pass through seems the right solution.
1 Like
sudo chmod a+x lcd_postinit.sc
sudo chmod a+x lcd_preinit.sc
sudo chmod a+x lcd_shutdown.sc
this seems to work. but, i had already passthrough so, not sure if it would have worked had it been set to itâs previous setting 
so anyway i did that, then rebooted but the script doesnât work still.
based on what elvis said, i think for this model could be it was using a different coding than the other models so you needed to adjust the baud and the other thing.
based on what i can see, the ts-653a and TVS-872XT used the script as is and it worked for them. maybe itâs different for the TS-877 so that is what iâm investigating atm xd
1 Like
Read robâs qnap to truenas conversion guide, he didnât mention anything about the fans or how to fix the lcd
but if you want a clean step by step i guess you can follow his guide instead. just keep in mind the things he omitted but i didnât 
Can add TS-451+ to another model tested working with truenas
https://www.reddit.com/r/qnap/comments/kq9dkk/repurposed_a_qnap_ts451_with_truenas_12/
1 Like
i guess i will work on the lcd on the side if i am able to get help on that cause i wouldnât know on my own
i scheduled to install the qnap 2.5gbe nic addon to my ts-653a. to check if that works.
then i can test zfs replication between ts-877 and ts-653a both running truenas scale dragonfish. thatâs the next thing i will be testing.
then i will have 2 backups with 2.5gbe multigig (usually 1 backup for local is sufficient. then the other would normally be offsite for the 3-2-1 backup golden standard. so donât copy me in this aspect necessarily i just had an extra nas with drives in it so why not xd)
more testing on the duckdns domain.
it does work, but after reboot there is a slight delay initially.
whereas lan ip is more or intantaneous and works more reliably.
with that in mind, iâm now inclined to believe timâs setup that uses a local domain name which is set on his own local dns (i assume on the router though i think his example was using pihole which i did try before so am familiar with) would be much more responsive than the duckdng org method.
but this is my mere speculation. i did attempt to try timâs method but got stuck with setting up traefik 
so for truenas with setup in this regard, those are your 3 options. if you want least amount of fuss to setup and donât care for valid certs https, then stick to the lan ip.
you can go further and do a local domain, either set on router for dns (for pfsense you modify the dns resolver custom), or using say pihole. In addition you may or may not need to also modify the filehosts on windows as a fall back in case you use an external dns e.g. 1.1.1.1 & 1.0.0.1 which may break those local domains from working.
for troubleshooting lcd
rather than simply restarting each time, i just went to shell and tried the commands to check if it works.
the command i was testing was shutting the lcd to make it sleep. cause right now the lcd is still open.
printf '\115\015' > /dev/ttyS1
sleep 1
A few suggestions i tried changing ttyS0, ttyS2, ttyS3
another suggestion was change baud but not sure how.
*update
this command sets the baud
stty -F /dev/ttyS0 1200
this was one of the commands for troubleshooting i tried but didnât work for me
root@truenas[/mnt/xxx/test/lcd]# dmesg | grep -i 16550 dmesg | grep -i 16550
[ 4.011954] Serial: 8250/16550 driver, 4 ports, IRQ sharing enabled
[ 4.012491] 00:03: ttyS0 at I/O 0x3f8 (irq = 4, base_baud = 115200) is a 16550A
[ 4.017503] 00:04: ttyS1 at I/O 0x2f8 (irq = 3, base_baud = 115200) is a 16550A
[ 0.940405] Serial: 8250/16550 driver, 4 ports, IRQ sharing enabled
[ 0.940733] 00:03: ttyS0 at I/O 0x3f8 (irq = 4, base_baud = 115200) is a 16550A
[ 0.941479] 00:04: ttyS1 at I/O 0x2f8 (irq = 3, base_baud = 115200) is a 16550A
[ 0.942497] AMDI0020:00: ttyS2 at MMIO 0xfedc9000 (irq = 3, base_baud = 3000000) is a 16550A
[ 0.943204] AMDI0020:01: ttyS3 at MMIO 0xfedca000 (irq = 4, base_baud = 3000000) is a 16550A
[: ']' expected
zsh: unknown file attribute: i
root@truenas[/mnt/xxx/test/lcd]#
For now for lcd, i think i should shift my focus to the ts-653a which has better results for others. If i get that to work, then i can come back to the ts-877. At least i would know i did the steps correctly, and that the issue for the ts-877 is that the script doesnât work for this model without changing some values. to what values exactly? i donât know its the ttyS⌠and the baud so iâm told. or maybe other stuff not sure.
iâll work on that tomorrow because i want to install the multigig nic as well.
*update
noticed jailmaker went down. caution when using these commands, i sus maybe that caused it to break jailmaker though iâm not really sure.
Different qnap model but very detailed explanation for the qnap lcd
for some hints to troubleshoot
1 Like
thkfully i had a backup of my truenas config. so i loaded that, it rebooted the nas, and now my jailmaker is working as normal.
the only thing that broke was before when on vpn urls worked. but it doesnât now instead i had to go windows 11 edit hostfiles with entries e.g.
192.168.0.24 authentik.xxxxxxx.duckdns.org
after that it worked. if i disable vpn it reverts to my pfsense router dns which has this custom rule
(note: theline breaks between each entry is important or it wonât work)
server:
local-zone: "xxxxxxxxxx.duckdns.org" redirect
local-data: "xxxxxxxx.duckdns.org A 192.168.0.24"
server:include: /var/unbound/pfb_dnsbl.*conf
the only reason why i have that duckdns is simply because itâs part of wolfgangs guide to get lets encrypt to work.
other wise i would much prefer say e.g. dashy.local.lan
or some variation of that. then i can have a shorter url which is what tim uses for his lan only setup.
for url management i have dockge link to 192.168⌠service port, so when i click those service ports it opens up using the lan ip.
but for dashy i set it to open using the duckdns.org urls.
so if my main urls dont work or work slow, i have my fallback. in dockge however if ur app has multiple ports, u have to recall which of those points to the apps web ui
someone recommended this to me
did a brief look, seemed ok though i havenât tried it yet. too many malware on github so bit wary where i download from but from what i can see supposedly they added more options. iâd just be happy if we can turn off the screen is more than enough xd
ok just checked, the ts-653a doesnât support addon card installs.
desktop already has 2.5gbe inbuilt to motherboard (modern desktop motherboards not trending with it. you would need to get the cheapest mobo possible just to get 1gbe only ports these days)
the ts-877 already has all the pcie slots populated so no room to put it there not to mention itâs already using sfp+ 10g fiber optic so no need for a 2.5gbe nic (it has a 1gbe for fallback if i need quick access in case the sfp+ has issues)
and the ts-253d which supports addons already natively has 2.5gbe not to mention itâs already installed with the tl-d400s sff-8088 card so no room for anything else.
so i donât have a home for this card. it didnât cost that much, will just keep it for a future project
will proceed with the ts-653a truenas scale dragonfish and the ts-877 truenas replication test. iâm just checking how that works since iâm new to this.
but from what i understand, you can select specific datasets and have that zfs replicate sync to the other truenas which makes it one alternative to rsync for truenas for backup and restore. I find that very intriguing.
It syncs so it only transfers data that changes only, so itâs quick and efficient.
ran into a roadblock trying to get ssh connection to work to the other truenas.
But when i tested rsync ssh connection between ts-253D (QNAP HBS) and TS-877, that worked fine when using root credentials.
But why doesnât truenas rsync tasks work between 2 truenas? still figuring that one out
Hi, I Had the same issue after upgrading. I managed to create the replication task by doing the following.
- On the source server go to Credentials > Backup Credentials and add a ssh Keypairs.
- Copy the public key.
- On the destination servers go to Credentials > Local Users and edit the root user. Paste the public key in the authorized keys section.
- Go back the source server and add an SSH Connection, select manual method. Select the private key that was created and then click on âDiscover Remote Host Keyâ
- Now you can go and create your replication task to your remote server.
Just wanted to add some feedback. Going from core to scale @kev89 's post works just fine.
Generate keys under: System â SSH Keypairs
Create connection under: System â SSH ConnectionsThanks @kev89
was hoping to test if connection worked that way before moving on to testing the replication. cause that would need a working ssh wouldnât it.
and another weird thing was, truenas semi automatic setup worked previously but it didnât this time. hm 
I can generate the SSH key pair and connection just fine for the two systems (Bluefin sys1 and Cobia sys2) and in fact I am using the ssh connection actively right now between both Truenas servers with Rsync just fine, though it is glacial slow. Thatâs the only way connecting the two machines will work. The SSH connection does not work at all if the replication task is created and attempted to be used instead of a Rsync task This is where the errors are being generated from; the replication task.
The convoluted special workaround settings for the replication task I found in the documentation only under advanced replication, say the issue is because of upgrades from systems that used root (earlier-Bluefin) and the eventual forced change to the use of admin and removal of root. I have tried following those special workaround instructions, but they also donât seem to work at this point. Changing from root to admin appears to have caused quite a number of issues for lots of things. I also could not find a way to not use SSH between the two systems.
When I tried core a year or so ago everything was fine and when I tried older Scale versions up until June everything was fine.
now i am beginning to wonder, could the reason be because of this?
because the ts-653a had upgraded from old truenas to dragonfish, maybe something got messed up along the way so maybe thatâs why it doesnât work exactly correctly for this?
but ssh works fines to ts-253d because no issue for that, but might be an issue for the 653a going through the few truenas upgrades over time? thatâs all i can think about.
Iâm more than willing to redo that from scratch if i have too. itâs a non critical backup (i have 2 backups).
taking a break to work with other stuff. but will return back later to update on progress with this at a later date ^^;
just have this running 24/7 but will report back on stability whether there were any crashed or anything like that if it happens. so far so good.
1 Like
Previously in regards to networking for jailmaker i recommended these 2 sources
they are still important and relevant. However, Stux came out with a tutorial which is more specific and detailed for jailmaker networking setup process. I suggest watching that as well (i did)
In my own journal in this thread i also covered how you would go about adding Nginx Proxy Manager for use with jailmaker docker containers, as well as setting up a domain name url with working https lets encrypt certs that are valid even for a lan only homelab setup.
For NPM use this guide
For Traefik use this guide
If you want to go further, here is some guides on creating and managing docker networking
so after i created a docker network called proxy, i then deploy a compose like this in dockge as an example for how i use networking in docker compose
services:
librespeed:
image: lscr.io/linuxserver/librespeed:latest
container_name: librespeed
environment:
- PUID=1000
- PGID=996
#note: change puid and pgid to a non root user ideally. Mine is called docker and part of a docker group.
- TZ=newyork
- PASSWORD=xxxxxxxxxx
- CUSTOM_RESULTS=false #optional
- DB_TYPE=sqlite #optional
- DB_NAME=xxxxxxxxxxx #optional
- DB_HOSTNAME=xxxxxxxxxxx #optional
- DB_USERNAME=xxxxxxxxxxxxx #optional
- DB_PASSWORD=xxxxxxxxxxxxxx #optional
- DB_PORT=xxxxxxxx #optional
- IPINFO_APIKEY=xxxxxxxxxxxx #optional
volumes:
- /mnt/docker/data/librespeed/appdata/config:/config
ports:
- 3480:80
networks:
- proxy
restart: unless-stopped
#networks: {}
networks:
proxy:
external: true
Iâm by no means an expert in this, so maybe there is more to get into this. I merely explained what i do for my own docker setup. But basically if the docker containers you deploy are not part of this network, then they wonât be able to communicate with it. So if they want to, then you merely have to add the proxy network (or whatever you name yours) to their docker compose and redeploy. hope this helps in covering an overall summary of the networking setup for truenas using jailmaker and for docker.
If anyone has suggestions or corrections, you are more than welcome.
1 Like
too bad i canât add to first post, o well this will have to do
not 100% sure but i think this was what happened to me. when setup bridge, the networking performance drops.
reason why i say this is because, last time i used docker container librespeed to test, the speeds were 7000 Mbps. but later it was only maxing at close to 1900 Mbps.
So what changed? i think it happened after i had finished setting the bridge networking for jailmaker i suspect. Not sure.
Also i was reading previous precedents mentioning something of this nature
iperf results are fine. Itâs only docker container speed apps like librespeed and openspeedtest where the speed is slower by comparison.
C:\iperf3.15_64>iperf3 -c 192.168.0.24
Connecting to host 192.168.0.24, port 5201
[ 5] local 192.168.0.152 port 56884 connected to 192.168.0.24 port 5201
[ ID] Interval Transfer Bitrate
[ 5] 0.00-1.00 sec 796 MBytes 6.68 Gbits/sec
[ 5] 1.00-2.00 sec 826 MBytes 6.92 Gbits/sec
[ 5] 2.00-3.00 sec 813 MBytes 6.82 Gbits/sec
[ 5] 3.00-4.00 sec 946 MBytes 7.94 Gbits/sec
[ 5] 4.00-5.00 sec 861 MBytes 7.22 Gbits/sec
[ 5] 5.00-6.00 sec 924 MBytes 7.75 Gbits/sec
[ 5] 6.00-7.00 sec 888 MBytes 7.45 Gbits/sec
[ 5] 7.00-8.00 sec 915 MBytes 7.68 Gbits/sec
[ 5] 8.00-9.00 sec 899 MBytes 7.54 Gbits/sec
[ 5] 9.00-10.00 sec 870 MBytes 7.30 Gbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval Transfer Bitrate
[ 5] 0.00-10.00 sec 8.53 GBytes 7.33 Gbits/sec sender
[ 5] 0.00-10.00 sec 8.53 GBytes 7.33 Gbits/sec receiver
iperf Done.
C:\iperf3.15_64>
Just oredred a PDU with a meter to keep track of power usage. It uses this which seems pretty accurate enough (not the exact same model as in the youtube but itâs the same brand and looks just like it)
note: about these pdus, some people they didnât check if they can populate all the sockets then find out you canât. this one can fit uk adapters side by side thkfully. I also checked that it can fit in my network rack horizontally. Another thing to watch out for, some cheap china manufactured pdus they use wires instead of solid copper bars, avoid those
Thought it would be interesting to share what kind of power usage for my truenas setup running
-
20 active docker containers (idle most of the time) using jailmaker
-
4x 4tb hdds in raidz1
-
2x 500gb m.2 sata ssds mirror
-
1 m.2 nvme ssd single boot drive
-
16gb ddr4 ram
-
fans max rpm
-
Nvidia 1050 GTX 2gb vram MSI Aero graphics card
-
SFP+ 10g pcie (with sfp+ 10g fiber transceiver)
once my package arrives, stay tuned. Wonât be able to do a before and after qts to truenas comparison for this, sorry
1 Like
Image
Meanwhile your wallet is screaming âLESS!â
1 Like
yeah lel. new nas owners are more than happy to add on more and more stuff. little do they realize how much power these things consume. the youtubers i follow like christian lempa and wolfgang they are really mindful of this.
heck christian had a bunch of hard drives, but he downsized and stuck with 4 hdds only, pretty much what i use myself. i also draw a line what iâm okay spending for and what i think is excessive
based on netdata observations for cpu, i think itâs low enough for me not to turn off some of these containers. when comparing it to my previous qts (i used grafana, prometheus and cadvisor for monitoring), the truenas switch seemed to consume way less. from 20% down to 1-5% with the occassional 20% blips.
It winds up to 50% + cpu when itâs doing something intensively e.g. when i first run jellyfin to scan the entire library of videos to index. which happens very rarely.
just waiting for the new hardware so i can get a more specific number rather than just guesswork
1 Like