[Solved] Fresh 6.9.1 install failing when starting Docker swag

[tech_rkn]

First, apologies for my English and for localizing my post in general setting, if not the good location.

 

Will try to be as throughout as possible.

 

As my original DIY nas/nextcloud server ( deb10,apache2,php7,demaria,lvm2,mdadm,nftables,fail2ban) is dying of old age, i would like to simplify the administration using Unraid.

 

I am using as a test bench my futur hardware, a MB Gigabytes B450M-DS3H rev 0.1 with latest uefi bios F61a with a AMD Zen3 2200G.

Also, i have a PCI-e to SATA card based on a Marvell 88SE9215.

 

I was able to install, boot, create an array with 3 disks ( test disks ).

 

Important info, i want to use IPV6 if possible.

 

I want to use 2 CA called SWAG and NEXTCLOUD. And maybe more, but later. And i would like them to "communicate as a pool" ...

 

I did find some info, YouTube stuff to understand how to add plug-ins and CA.

 

So, i hope the community might be able to help me.

 

First:

Network context. I am using dhcp from my FTTH box. My "Unraid Tower" is saying:

unRAID Server OS version: 6.9.1
Ipv4 address: not set
ipv6 address: 2a01...truncated, of course.....

Seem strange for me, as network is configure for Ipv4+Ipv6. And Ip is 192.168.1.137 ... not very important, except if it is the main bug !

 

Following tutorial from user superboki ( in french les copains, in french ) i did create a "custom" network using console for this docker and command line:

docker create network dockernet

witch ended with an lonnnnng number as a response.... seem to be ok, so far.

 

After that, i install CA SWAG, used mostly basic option for config, except network type using "custom:dockernet" from dropdown list.

Used my domain, emails  ... 

Install went with a successful message.

 

TROUBLES START HERE !

 

Unfortunately, when starting the CA docker for SWAG, i got an EXECUTION ERROR.

Here some logs and stuff... I believe this is link to the network but i am not so good at managing IP's and routes...

FYI: able to reproduce bug when:

- using new test hardware.

- using ipv4 only

- not using custom network within unraid set-up.

 

Command:
root@localhost:# /usr/local/emhttp/plugins/dynamix.docker.manager/scripts/docker create --name='swag' --net='dockernet' -e TZ="America/Los_Angeles" -e HOST_OS="Unraid" -e 'EMAIL'='[email protected]' -e 'URL'='xxx.ovh' -e 'SUBDOMAINS'='www,cloudcmd' -e 'ONLY_SUBDOMAINS'='false' -e 'VALIDATION'='http' -e 'DNSPLUGIN'='' -e 'EXTRA_DOMAINS'='' -e 'STAGING'='false' -e 'DUCKDNSTOKEN'='' -e 'PROPAGATION'='' -e 'PUID'='99' -e 'PGID'='100' -p '80:80/tcp' -p '443:443/tcp' -v '/mnt/user/appdata/swag':'/config':'rw' --cap-add=NET_ADMIN 'linuxserver/swag'
b05c98af741c9402e340256e33e80d8e71eddd5254ac62166f238bf1d821fdd4

The command finished successfully!

 

nmbd[1932]:   NOTE: NetBIOS name resolution is not supported for Internet Protocol Version 6 (IPv6).  ( pas grave )

webGUI: Successful login user root from 2a01:e0a:225:4e50::  ( cool )

kernel: mdcmd (38): set md_num_stripes 1280    ( bla bla array start correctement )
kernel: mdcmd (39): set md_queue_limit 80
kernel: mdcmd (40): set md_sync_limit 5
kernel: mdcmd (41): set md_write_method
kernel: mdcmd (42): start STOPPED
kernel: unraid: allocating 20870K for 1280 stripes (4 disks)
kernel: md1: running, size: 976762552 blocks
kernel: md2: running, size: 468850520 blocks 

je relance docker
root: starting dockerd ...
avahi-daemon[2326]: Joining mDNS multicast group on interface docker0.IPv4 with address 172.17.0.1.
avahi-daemon[2326]: New relevant interface docker0.IPv4 for mDNS.
avahi-daemon[2326]: Registering new address record for 172.17.0.1 on docker0.IPv4.

SWAG ERROR
driver failed programming external connectivity on endpoint swag (8reduit8): Error starting userland proxy: listen tcp 0.0.0.0:80: bind: address already in use"

 

 

I hope is just a very simple error from a very beginner....

 

Thank you for any help.

 

 

 

 

  

tower-diagnostics-20210407-0004.zip

[tech_rkn]

I found the problem. As I said, a beginner mistake, or a lack of warnings in the CA/unraid.

 

Under Setting/Management Access the ports need to be change from 80/443 to anything else...

 

Simple ports conflicts between main system and CA docker.

 

 

 

 

[tech_rkn]

Solved 

 

A button anywhere for the creator of the post to closed is own question ??

[ChatNoir]

Good news that you found the issue.

You can simply edit the initial post and add Solved to it in the future.

 

I did it for this topic.

[trurl]

4 hours ago, tech_rkn said:

I found the problem. As I said, a beginner mistake, or a lack of warnings in the CA/unraid.

 

Under Setting/Management Access the ports need to be change from 80/443 to anything else...

 

Simple ports conflicts between main system and CA docker.

The usual way to deal with this would be to remap the docker ports and leave the Unraid ports alone.

 

You are going to find a lot of dockers have port conflicts unless you map them.

[JonathanM]

20 hours ago, tech_rkn said:

Also, i have a PCI-e to SATA card based on a Marvell 88SE9215.

Marvell based cards are not recommended.

[tech_rkn]

3 hours ago, jonathanm said:

Marvell based cards are not recommended.

HO ! Thanx for the warning. and a JMB575 ?? 

[ChatNoir]

 

[tech_rkn]

Thank you both. I read some threads about this controller. But I believe ( maybe wrongly) that I may not be concerned as I will not have VM. Also, I will keept the parity disk link on the motherboard sata, and the data on the expender.

[trurl]

3 minutes ago, tech_rkn said:

I will keept the parity disk link on the motherboard sata, and the data on the expender

Not putting parity on a port multiplier makes sense for efficiency. Just thought I would comment though in case you think parity is somehow more important than other disks. Parity requires all other disks to rebuild a missing disk, and parity actually contains none of your data. So parity is arguably the least important disk.

[tech_rkn]

On my old DIY ubuntu server in softraid 60, I never experienced troubles when the disk aggregated in a specific pool array were on the same controler pool. Heavy shit fall on it when the disk were scattered accross multiple controler.

I had 2 pools of pci extenders, and 2 array ... In 5 years, had only 1 disk going mad on 2 pools made of 6 disk each. In the last 6 month, lost 4 disk. All the same brand, bought over 3 month... So I guess I have about 1 to 3 month for a massive hardware failure.

 

 

I will test my new rig as the primary end point for my Nextcloud. And rsync in real time the data on my old rig... If thing go smoothly, I will then discontinued my old stuff. 

 

Thanks for all comments, advice and help.