Error 502 Bad Gateway after migrating from NPM to NPMplus

[Joly0]

Hey, i swaped out my NPM container for NPMplus in Unraid, afterwards i only get 502 errors for all my websites.
I have my domain on cloudflare and i am using cloudflare tunnel to directly connect to the cloudflared container running on my unraid server aswell.
This worked no problem for NPM, but something is buggy with NPMplus here. Not sure were to pinpoint this, so i am opening an issue here now.

[Zoey2936]

Please send the docker logs

[Joly0]

Nvm, i found the issue. My cloudflare tunnel was configured wrong and had the old hostname of NPM still as the configured proxy instead of npmplus.

[Joly0]

Btw, as there is none, i am creating a template for npmplus for unraid, so others can use your container aswell there easily

[joaoferreira-git]

Hi, im trying to migrate to this but when I do it doesn't have the proxys or my cert from cloudflare, can you help me?
I basically made a copy of my current nginx config and then pointed the npmplus container to this folder only changing the ports and path mapping to the correct ones.
Can you explain how you managed to migrate to it?
Edit: Thanks for making the template

[Joly0]

Hi, im trying to migrate to this but when I do it doesn't have the proxys or my cert from cloudflare, can you help me?
I basically made a copy of my current nginx config and then pointed the npmplus container to this folder only changing the ports and path mapping to the correct ones.
Can you explain how you managed to migrate to it?
Edit: Thanks for making the template

Hm, it worked for me. Are the permissions right? Might be an issue. For certs, if they are custom, afaik you have to delete them

[Zoey2936]

Sounds good

[Joly0]

Btw, while we are here talking about this. I found out, that changing some settings of a proxy host reloads nginx. This is known and i am aware of this. But somehow this kills the complete container. I am not sure this is intended?

[Zoey2936]

please provide the full log

[scampower3]

From my syslog

Jan 23 19:38:59 HomeNAS  avahi-daemon[21933]: Joining mDNS multicast group on interface veth454fc89.IPv6 with address fe80::1490:d6ff:feb9:f3cc.
Jan 23 19:38:59 HomeNAS  avahi-daemon[21933]: New relevant interface veth454fc89.IPv6 for mDNS.
Jan 23 19:38:59 HomeNAS  avahi-daemon[21933]: Registering new address record for fe80::1490:d6ff:feb9:f3cc on veth454fc89.*.
**Jan 23 19:39:22 HomeNAS kernel: nginx[20862]: segfault at 151a0be4f060 ip 0000151a11a6ed44 sp 00007ffcf7160950 error 4 in libperl.so[151a11a47000+1a3000]**
Jan 23 19:39:22 HomeNAS kernel: Code: 1a 0f 1f 44 00 00 48 83 c0 01 80 38 00 75 f7 48 83 c1 08 49 89 c7 49 39 c8 75 c1 48 8b 05 04 57 34 00 4c 8b 28 4d 85 ed 74 1f <49> 8b 45 00 49 8d 4f 01 48 39 c8 0f 84 c7 23 00 00 4d 85 e4 75 09
Jan 23 19:39:22 HomeNAS kernel: docker0: port 1(veth454fc89) entered disabled state
Jan 23 19:39:22 HomeNAS kernel: vethb439d56: renamed from eth0
Jan 23 19:39:22 HomeNAS  avahi-daemon[21933]: Interface veth454fc89.IPv6 no longer relevant for mDNS.
Jan 23 19:39:22 HomeNAS  avahi-daemon[21933]: Leaving mDNS multicast group on interface veth454fc89.IPv6 with address fe80::1490:d6ff:feb9:f3cc.
Jan 23 19:39:22 HomeNAS kernel: docker0: port 1(veth454fc89) entered disabled state
Jan 23 19:39:22 HomeNAS kernel: device veth454fc89 left promiscuous mode
Jan 23 19:39:22 HomeNAS kernel: docker0: port 1(veth454fc89) entered disabled state
Jan 23 19:39:22 HomeNAS  avahi-daemon[21933]: Withdrawing address record for fe80::1490:d6ff:feb9:f3cc on veth454fc89.
Jan 23 19:39:23 HomeNAS kernel: docker0: port 1(veth3990869) entered blocking state
Jan 23 19:39:23 HomeNAS kernel: docker0: port 1(veth3990869) entered disabled state
Jan 23 19:39:23 HomeNAS kernel: device veth3990869 entered promiscuous mode
Jan 23 19:39:23 HomeNAS kernel: docker0: port 1(veth3990869) entered blocking state
Jan 23 19:39:23 HomeNAS kernel: docker0: port 1(veth3990869) entered forwarding state
Jan 23 19:39:23 HomeNAS kernel: eth0: renamed from veth63e25d4
Jan 23 19:39:23 HomeNAS kernel: IPv6: ADDRCONF(NETDEV_CHANGE): veth3990869: link becomes ready
Jan 23 19:39:25 HomeNAS  avahi-daemon[21933]: Joining mDNS multicast group on interface veth3990869.IPv6 with address fe80::a:eff:fe24:b23.
Jan 23 19:39:25 HomeNAS  avahi-daemon[21933]: New relevant interface veth3990869.IPv6 for mDNS.
Jan 23 19:39:25 HomeNAS  avahi-daemon[21933]: Registering new address record for fe80::a:eff:fe24:b23 on veth3990869.*.
Jan 23 20:00:54 HomeNAS  crond[1259]: exit status 1 from user root /usr/local/sbin/mover &> /dev/null

[Zoey2936]

I will remove perl

[Joly0]

I think adding "--restart unless-stoppe" might also help, so the container can restart, if the nginx process cant restart on its own

Or maybe "--restart on-failure"

[scampower3]

Using the develop docker tag rn. Haven't had any crashes for the past few hours. Much more stable without the perl module.