Pure-ftpd Connection Error on Google Cloud Computer

I can't connect FTP, tried everything:

1. Forwarded ports on Google Cloud
2. Stopped CSF
3. Created new user with : pure-pw useradd newftpuser -u nginx -g nginx -D /home/nginx/domains/mydomain.com/public/
3. Read other threads regarding connection problem

Yet, no way. I'm stuck.

What else can I try?

Thank you so much for your help in advance.

• CentOS Version: CentOS 7 64bit
• Centmin Mod Version Installed: 123.09beta01
• Nginx Version Installed: 1.15.3
• PHP Version Installed: 7.2.9
• MariaDB MySQL Version Installed: 10.1.13
• When was last time updated Centmin Mod code base ? : Clean install, Ran centmin.sh menu option 22 submenu, Got Letsencrypt SSL certificate Nginx vhost, Selected #4 to issue live cert with HTTPS default (trusted).
• Persistent Config:
  NGXDYNAMIC_NGXPAGESPEED='y'
  NGINX_PAGESPEED='y'
  EMAIL='[email protected]'
  PUSHOVER_EMAIL='[email protected]'
  LETSENCRYPT_DETECT='y'

 

Thank you for your reply honestly.

I opened even all the ports and tried in that way too: https://goo.gl/FfMtWJ

It might be something else, that's why I wanted to dig it more.

After the installation, my ftp login was provided as:

FTP hostname : 10.164.0.3
FTP port : 21
FTP mode : FTP (explicit SSL)
FTP Passive (PASV) : ensure is checked/enabled
FTP username created for mydomain.com : username
FTP password created for mydomain.com : password

The hostname is the internal IP of my Google Cloud instance, I guess it should be external.

I searched Google and could not find information regarding how to change it after doing CMM option 22. Although I had deleted it initially from /etc/hosts file and locked it with chattr +i - because no matter what when you restart network google adds its own configuration as well to your hosts file -, added my xternal IP and domain before adding the website with option 22.

It's OK for now, my using FTP as STFP with root, but for security reasons I want to disable it in the future.

Any further ideas?

 

I know Google Compute costs a lot, I just want to stick to it a couple of months since they gave free credits 5 months ago. After that time I want to transfer to DO.

In the meantime I can use root with SFTP for file transfers and a plugin like Sendgrid for SMTP emails (yes my CMM has a Postfix problem as well - I guess because of this hostname with internal IP issue), because apart from these two all seems fast and fine, very fine with good bechmarks indeed. Previously I used EE for a client website but no proper documentation mainly regarding webp support made me to choose CMM this time.

 

Thank you for your reply.

Here is the output:

Code (Text):

eth0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1460
        inet 10.164.0.3  netmask 255.255.255.255  broadcast 10.164.0.3
        inet6 fe80::4001:aff:fea4:3  prefixlen 64  scopeid 0x20<link>
        ether 42:01:0a:a4:00:03  txqueuelen 1000  (Ethernet)
        RX packets 556812  bytes 438886510 (418.5 MiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 998622  bytes 2396376274 (2.2 GiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

lo: flags=73<UP,LOOPBACK,RUNNING>  mtu 65536
        inet 127.0.0.1  netmask 255.0.0.0
        inet6 ::1  prefixlen 128  scopeid 0x10<host>
        loop  txqueuelen 1000  (Local Loopback)
        RX packets 9567  bytes 67164521 (64.0 MiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 9567  bytes 67164521 (64.0 MiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

Reconfiguring it with that change then reinstalling to make everything work smoothless is what I want actually, as a newbie. Thank you!

P.S. What's the shortcode for codes, [ICODE] doesntt work? Edit: Found it.

 

I had assigned it after 3 failed installs, before my last install. I assigned it then installed CMM and created the website with option 22.

Now my last instance/droplet/server has a special static IP but the problem exists.

This is what I have from as I read from another post.

Code (Text):

nano /etc/sysconfig/network-scripts/ifcfg-eth0

Code (Text):

# Generated by parse-kickstart
DHCP_HOSTNAME="localhost"
BOOTPROTO="dhcp"
DEVICE="eth0"
ONBOOT="yes"
UUID="d215aefd-9127-4949-8449-cd367355a866"
MTU=1460
PERSISTENT_DHCLIENT="y"

Should I add IP address here, what do you think?

 

I have read on other posts that you don't have an experience with Google Cloud and I totally understand George but maybe I thought this could a reference for many GCE users like me, since Google provides a credit of 300$ for 12 months.

 

It was a firewall issue, I sorted it out by adding rules via SSH:

Code (Text):

gcloud compute firewall-rules create <custom-rule-name-up-to-you> --allow tcp:20,tcp:21,tcp:22,tcp:25,tcp:53,tcp:80,tcp:110,tcp:143,tcp:161,tcp:443,tcp:465,tcp:587,tcp:993,tcp:995,tcp:1110,tcp:1186,tcp:1194,tcp:81,tcp:9418,tcp:30001-50011 --source-tags=<list-of-your-instance-names> --source-ranges=0.0.0.0/0 --description="<any-description-here>" --direction=INGRESS


gcloud compute firewall-rules create <custom-rule-name-up-to-you> --allow udp:67,udp:68,udp:1110,udp:33434-33534,udp:20,udp:21,udp:53,udp:113,udp:123 --source-tags=<list-of-instance-names> --source-ranges=0.0.0.0/0 --description="<any-description-here>" --direction=INGRESS

Thank you so much for enlightening me!

 

Thank you once again! Sorry for the headache