Can I somehow run the plugin on 6.12.14? I don't want to update to V7 or any Beta at the moment.

24 minutes ago, hYtas said:

Can I somehow run the plugin on 6.12.14? I don't want to update to V7 or any Beta at the moment.

Nope, only possible on Unraid 7.0.0 since keepalived is not compatible with 6.x.x any more.

 

The better question is why don't you upgrade to Unraid 7.0.0 since stable was released or are you also such a user who waits a few weeks for whatever reason...

I fucked up many times for being an early adopter. I’ll wait some time.

 

Even if the beta was stable, the stable version always has bugs. 
I will only use my second new server as a data backup until a few versions have been released. 

I like your plugin but I don't want to break everything

Edited January 12, 20251 yr by hYtas

1 hour ago, hYtas said:

I fucked up many times for being an early adopter. I’ll wait some time.

Sorry but I don‘t understand this? What should a version update destroy?

Just look at the release post, almost no complaints/error reports, most of the times it‘s some weird configuration that causes issues…

I dared to carry out an update... does the plugin only back up the docker or can it also back up the shares?

10 minutes ago, hYtas said:

I dared to carry out an update... does the plugin only back up the docker or can it also back up the shares?

It can back up all mapped paths in containers, it is mostly designed to backup Docker and LXC containers (including their paths), I didn't got a lot of feedback so I have moved it down in my list of things that I have to do.

 

You could create a "dummy" container which is always stopped but has /mnt (or all the shares individually) mounted and back it up like that as a workaround.

Hi @ich777, thank you for making this! I’m planning to build a second, mirrored unraid server as an offsite fallback in case of long term power outages. Will this work?

 

As in, can I have my primary in one state (California) and backup in another state (Texas), and have the backup kick in if the primary goes offline?

2 minutes ago, AstroBoy said:

Hi @ich777, thank you for making this! I’m planning to build a second, mirrored unraid server as an offsite fallback in case of long term power outages. Will this work?

Yeah it should, however keepalived is mostly designed to be used in a LAN environment. Maybe this could be set up to go through Tailscale or similar but I've never played around with that.

Could be a bit complicated to set up but it should be possible.

 

6 minutes ago, AstroBoy said:

As in, can I have my primary in one state (California) and backup in another state (Texas), and have the backup kick in if the primary goes offline?

Maybe but that depends how your ISP and of course you can handle Failover from your Internet connection.

As said above this is maybe a bit difficult to set up but maybe I'm overthinking it currently.

 

Maybe Wireguard or Tailscale can help.

Unfortunately, I only got around to testing the plugin today. Everything worked without any problems. First there were problems because I had not deleted a <, but it was my own stupidity.
Tomorrow evening I will simulate the failure of my Unraid1.
 

Wondering if I could get some clarity on this plugin (looks amazing btw).

 

I want to deploy this to an offsite backup.

So can I point this at a public IP or an unraid connect address at all?

Alternatively what would be the best way to accomplish this if not that?

 

Secondly can i set this somehow to take over serving things like bitwarden etc using the same webaddress (i would setup the cloudflare load balancing thing)

But can NPM just kick on remotely and take over in that case then?

 

Or am I wanting something completely different than what this plugin offers?

 

Thanks

47 minutes ago, Arcau said:

I want to deploy this to an offsite backup.

So can I point this at a public IP or an unraid connect address at all?

Sort of, I wouldn't recommend to expose Unraid SSH service to the Internet but maybe you could accomplish this through Tailscale to connect the two servers together and keepalived is also meant for your LAN <- possibly works over Tailscale too.

 

47 minutes ago, Arcau said:

Or am I wanting something completely different than what this plugin offers?

Depends, this plugin is meant for replication and not a failover/load balancing, so to speak the data is synchronized on a schedule and not in sync all the time.

However you can, with keepalived, configure the second server to take over for applications which are no critical like Media services or static web services where you don't need the most recent data.

Yeh the services I am wanting to "failover" are mostly just going to be bitwarden, immich and keeping nextcloud alive - I am not in need of super up to date data - if its a few hours old or a day old it makes little difference but not having the services live when needed.

 

I will play around with tailscale see if I can get things to work.

 

The failover part in theory i'm hoping can just be handled at the cloudflare level with their load balancing - and I can just leave NPM and the containers running on that box with my primary location set to the default and then if unreachable cloudflare will failover to that site wan IP

 

Looks super great though and I appreciate all the work you put into the containers you put onto the community app store @ich777❤️

 

If my use case works out I will try to do some documentation for any other people wanting to attempt it

49 minutes ago, Arcau said:

If my use case works out I will try to do some documentation for any other people wanting to attempt it

Thanks! This is much appreciated!

I did have a question about keepalived - does it need to loose network connection to the master totally? It doesnt check if docker container is running then fire up the replicated one to take over does it?

 

So far I have setup the secondary box locally and replication is working and everything seems great - I am yet to get the box out to the offsite location to test remote.

I have setup an SD-WAN between the sites so I am hoping that it can just function much the same just on a different subnet, as long as it can talk both ways it should be ok right??

1 minute ago, Arcau said:

I did have a question about keepalived - does it need to loose network connection to the master totally? It doesnt check if docker container is running then fire up the replicated one to take over does it?

Yes, when keepalived on the master is not reachable (may it be a network loss or keepalived itself not online).

No it doesn't check if a container is not online and then takes over.

 

5 minutes ago, Arcau said:

I have setup an SD-WAN between the sites so I am hoping that it can just function much the same just on a different subnet, as long as it can talk both ways it should be ok right??

I hope so, I've never tried that because I don't have a offsite server.

18 hours ago, ich777 said:

I hope so, I've never tried that because I don't have a offsite server.

I have tested on a vlan and it is working so considering SD-WAN acts basically the same this should work assuming the firewall rules allow the cross talk.
Redirect the tunnels from cloudflare to use the keepalived IP for the services and tested while on same subnet - about to test it on the new vlan to see if it will still kick in or if I need to make some changes to the backup server to get it to work like that - will let you know how it goes - and then hopefully this weekend I will be ready to deploy offsite and test.

Hrmmm yeh it doesnt seem to work nicely when on different subnet - will have to try figure out how to make that work over sd-wan then.

It doesnt acknowledge that the master goes down - the update and send from master to backup works fine - just the backup always sees the master as up for some reason.

 

Going to setup a /26 subnet on both sides and see if it can negotiate that with vlan tagging. works locally now so thats something.

The other issue I am now having is that the Nextcloud AIO doesnt actually start properly - the master container starts but the rest of them don't auto as they arent recognised by keepalived it seems

 

OK so the SD-WAN will work for replication just fine - but because of the way I have it setup using unifi site magic we can't share subnets which basically kills the failover / docker auto start.

 

If I have that up the backup server can't find the host - because of weird vlan shenanigans but will be available for the failover - so I can still have it set to manual / failover with unraid connect basically - if my services go down I just need to click that on.

 

Until I can figure out how to make the SD-WAN play nicely or somehow setup a site-to-site that can be on the same subnet - at least the remote backup and file transfer is working nicely.

 

Down another rabbit hole of networking I go!

1 minute ago, Arcau said:

Until I can figure out how to make the SD-WAN play nicely or somehow setup a site-to-site that can be on the same subnet - at least the remote backup and file transfer is working nicely.

Tailscale also doesn't work when you specify the Tailscale subnet for keepalived?

It immediately went down even though the server was still up and able to talk to each other - and unable to point the cloudflare tunnel to the tailscale IP it seems - just wont resolve for me

I may just be doing this very wrong I have not really worked with VPN / SD-WAN  or anything like this before all new territory for me

 

Unraid 7.0.1 seems to have broken this plug-in.

Templates are created, paths are copied successfully but images aren't pulled and containers aren't created.

Known issue?

Edited March 7, 20251 yr by glenn.miller

15 minutes ago, glenn.miller said:

Unraid 7.0.1 seems to have broken this plug-in.

I look into that, this will definitely take me a few days but I'll let you know.

 

@alturismo does it work for you on 7.0.1?

16 minutes ago, glenn.miller said:

Unraid 7.0.1 seems to have broken this plug-in.

 

 

Just now, ich777 said:

@alturismo does it work for you on 7.0.1?

 

here it looks ok

 

from source Server log

 

.
..
...
sent 690,225 bytes  received 25,049 bytes  1,430,548.00 bytes/sec
total size is 178,563,913  speedup is 249.64
Info: Starting sync from path: /mnt/user/appdata/homeassistant/custom-cont-init.d/ for container: homeassistant
sending incremental file list

sent 104 bytes  received 12 bytes  232.00 bytes/sec
total size is 1,027  speedup is 8.85
Info: Remote host replication task for Docker container: homeassistant starting
homeassistant
Untagged: lscr.io/linuxserver/homeassistant:latest
Deleted: sha256:dbbfb5b83cfcb858cb6d97c1f6e8d70f92412a13f2acac68d337fcd119b1be8b
Deleted: sha256:cdd951fb171ac1ddf09e9f16892dd0506fa5cb4c2ceaa0f3e6405e07bffac3bf
Deleted: sha256:5def9803cb27d85184144edae4584c6bf422cdb99a285c84d29b4c13e2c656ef
Deleted: sha256:2c27cffeec9b34a10c1abac483440d94b5d02b2c1d7cf199ba3cd6732817c5b2
Deleted: sha256:713725e2d98aa25450de57a6844edbf9f02e40554bc8d73b6a546570da3d8d91
Loaded image: ur_replication_lscr.io/linuxserver/homeassistant:latest
Untagged: ur_replication_lscr.io/linuxserver/homeassistant:latest
Info: Creating new container: homeassistant
7720ab672429758afdf26c8cf665c4dd090114a86bcdf01c9c8eb1ef375a0779
Info: Remote host replication task for Docker container: homeassistant finished
Info: Replication from container: homeassistant finished
Info: Docker Replication finished
Info: Unraid-Replication finished 2025-03-06 04:42:30
root@AlsServerII:~# 

 

any more Infos needed ?

12 hours ago, alturismo said:

 

 

here it looks ok

 

from source Server log

 

.
..
...
sent 690,225 bytes  received 25,049 bytes  1,430,548.00 bytes/sec
total size is 178,563,913  speedup is 249.64
Info: Starting sync from path: /mnt/user/appdata/homeassistant/custom-cont-init.d/ for container: homeassistant
sending incremental file list

sent 104 bytes  received 12 bytes  232.00 bytes/sec
total size is 1,027  speedup is 8.85
Info: Remote host replication task for Docker container: homeassistant starting
homeassistant
Untagged: lscr.io/linuxserver/homeassistant:latest
Deleted: sha256:dbbfb5b83cfcb858cb6d97c1f6e8d70f92412a13f2acac68d337fcd119b1be8b
Deleted: sha256:cdd951fb171ac1ddf09e9f16892dd0506fa5cb4c2ceaa0f3e6405e07bffac3bf
Deleted: sha256:5def9803cb27d85184144edae4584c6bf422cdb99a285c84d29b4c13e2c656ef
Deleted: sha256:2c27cffeec9b34a10c1abac483440d94b5d02b2c1d7cf199ba3cd6732817c5b2
Deleted: sha256:713725e2d98aa25450de57a6844edbf9f02e40554bc8d73b6a546570da3d8d91
Loaded image: ur_replication_lscr.io/linuxserver/homeassistant:latest
Untagged: ur_replication_lscr.io/linuxserver/homeassistant:latest
Info: Creating new container: homeassistant
7720ab672429758afdf26c8cf665c4dd090114a86bcdf01c9c8eb1ef375a0779
Info: Remote host replication task for Docker container: homeassistant finished
Info: Replication from container: homeassistant finished
Info: Docker Replication finished
Info: Unraid-Replication finished 2025-03-06 04:42:30
root@AlsServerII:~# 

 

any more Infos needed ?

Logs from my host plus container list on Client (Backup)

 

Mar 8 09:57:05 unraid root: unraid-replication: Info: SSH connection working properly!

Mar 8 09:57:06 unraid root: unraid-replication: Info: Starting Docker Replication

Mar 8 09:57:07 unraid root: unraid-replication: Info: Replication from container nzbget starting

Mar 8 09:57:11 unraid root: unraid-replication: nzbget

Mar 8 09:57:12 unraid root: unraid-replication: sha256:d62bfcc3d7cce350f0a2831084891f0d1901af0472c040c6eb8bca5c4898e2fa

Mar 8 09:57:13 unraid root: unraid-replication: Untagged: ur_replication_linuxserver/nzbget:latest

Mar 8 09:57:13 unraid root: unraid-replication: Deleted: sha256:d62bfcc3d7cce350f0a2831084891f0d1901af0472c040c6eb8bca5c4898e2fa

Mar 8 09:57:13 unraid root: unraid-replication: Deleted: sha256:b25064b5fbc649f6aca569c9a7c90f07be034941fff0d075b558535d09d5e240

Mar 8 09:57:15 unraid root: unraid-replication: Info: Starting sync from path: /mnt/user/appdata/nzbget/config/ for container: nzbget

Mar 8 09:57:15 unraid root: unraid-replication: sending incremental file list

Mar 8 09:57:15 unraid root: unraid-replication:

Mar 8 09:57:15 unraid root: unraid-replication: sent 82 bytes received 12 bytes 188.00 bytes/sec

Mar 8 09:57:15 unraid root: unraid-replication: total size is 68,910 speedup is 733.09

Mar 8 09:57:16 unraid root: unraid-replication: nzbget

Mar 8 09:57:16 unraid root: unraid-replication: Info: Remote host replication task for Docker container: nzbget starting

Mar 8 09:57:17 unraid root: unraid-replication: Untagged: linuxserver/nzbget:latest

Mar 8 09:57:17 unraid root: unraid-replication: Deleted: sha256:6c1b5c0ae5cdd254795550118bbca08a211f7dafd1b7f5c0b513677354131fea

Mar 8 09:57:17 unraid root: unraid-replication: Deleted: sha256:3ee6b56e4bd42bebac4eec680397e0dcb039d78bff07e9e1b5ad200a8f5cd17b

Mar 8 09:57:17 unraid root: unraid-replication: Deleted: sha256:a75d93de9b7898ef3d62088faf36f4e7c51654ce2d668ff136630dc2264b277a

Mar 8 09:57:17 unraid root: unraid-replication: Deleted: sha256:e728b54561fdee4cd891ab8ec63bb0c162c2b2a5a1d5dcd5bc8166c330c7d4b0

Mar 8 09:57:17 unraid root: unraid-replication: Deleted: sha256:aeb98cb6333ed2f4cae98452e65e8cf0feb3ab5f19dc9c38a3831241f85c4a3d

Mar 8 09:57:17 unraid root: unraid-replication: Deleted: sha256:c3b6680bc997c32cb8441e3073c7a2021d467b4c276b608b903f0fe73dbab9ee

Mar 8 09:57:19 unraid root: unraid-replication: Loaded image: ur_replication_linuxserver/nzbget:latest

Mar 8 09:57:19 unraid root: unraid-replication: Untagged: ur_replication_linuxserver/nzbget:latest

Mar 8 09:57:19 unraid root: unraid-replication: Info: Creating new container: nzbget

Mar 8 09:57:19 unraid root: unraid-replication: Info: Remote host replication task for Docker container: nzbget finished

Mar 8 09:57:19 unraid root: unraid-replication: Info: Replication from container: nzbget finished

 

root@tower:~# docker ps -a

CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES

491cf10d4ff8 adguard/adguardhome "/opt/adguardhome/Ad…" 2 weeks ago Up 2 days adguard

root@tower:~#

 

8 hours ago, glenn.miller said:

Logs from my host plus container list on Client (Backup)

 

here from MASTER

 

 

here from BACKUP (scheduled every thursday to update, 2 days ago ...)

 

here is all fine, both running 7.01

 

only issue i have, every now and then, BACKUP is registring as MASTER even MASTER is still online

(BACKUP coming from sleep ..) but thats a different story