amazon web services – SSHtunnelForwarder to Linux Box, then connect to Postgres RDS (AWS)

I am trying to write a python script to update DDL in an Aurora RDS (postgres) on AWS. To do this, I need SSHTunnel to a Linux box which is on the edge of the VPC, then to make another jump to the RDS server. When I connect to RDS via the psql client on the Linux Box, I can connect correctly. But, when I try to sshtunnel via SSHTunnelForwarder in python, I get an error saying: sshtunnel.BaseSSHTunnelForwarderError: Could not establish session to SSH gateway

Here is my code:

with SSHTunnelForwarder(
    (LinuxBoxHost,22),
    ssh_username='ubuntu',
    ssh_pkey="/path/to/.pem file",
    remote_bind_address=(RDSHostName,5432)) as server:

    server.start()
    print(server.local_bind_port)

I can not even test my connection to the RDS database because I can not touch the Linux Box via SSH.

I do not know what I'm doing wrong here. I have looked at various examples online, and it appears that the parameters I am entering appear to be correct.

More details:
–Can SSH -i via the command line to the linux box very well
–Can connect to RDS via DBeaver, which also uses a .pem file for ssh tunnel
–Check syslog and auth.log, but there is no useful error information in both.

I wonder if I'm just missing something conceptual here? Is it only possible? Maybe the way I imagine our network infrastructure is not correct, but I spoke to our IT specialist and it seems to me that I have the right idea. Any advice would be appreciated, thanks in advance!

postgresql – SSHtunnelForwarder to Linux Box, and then connect to Postgres RDS (AWS)

I am trying to write a python script to update DDL in an Aurora RDS (postgres) on AWS. To do this, I need SSHTunnel to a Linux box which is on the edge of the VPC, then to make another jump to the RDS server. When I connect to RDS via the psql client on the Linux Box, I can connect correctly. But, when I try to sshtunnel via SSHTunnelForwarder in python, I get an error saying: sshtunnel.BaseSSHTunnelForwarderError: Could not establish session to SSH gateway

Here is my code:

with SSHTunnelForwarder(
    (LinuxBoxHost,22),
    ssh_username='ubuntu',
    ssh_pkey="/path/to/.pem file",
    remote_bind_address=(RDSHostName,5432)) as server:

    server.start()
    print(server.local_bind_port)

I can not even test my connection to the RDS database because I can not touch the Linux Box via SSH.

I do not know what I'm doing wrong here. I have looked at various examples online, and it appears that the parameters I am entering appear to be correct.

I wonder if I'm just missing something conceptual here? Is it only possible? Maybe the way I imagine our network infrastructure is not correct, but I spoke to our IT specialist and it seems to me that I have the right idea. Any advice would be appreciated, thanks in advance!

linux – Samba: replace only illegal characters, instead of using a completely gibberish file name?

I understand that Linux and Windows have a different set of illegal characters, but it seems like a filename contains even a single special character like : on a Samba shared directory (Linux server), the entire file name becomes gibberish.

For example, "a: b" becomes "A4WTMA ~ A". It's not so useful to know what the file is … can I make the name more predictable? As,

  • Replace only special characters with "_".
  • If this name already exists, add a number at the end.

So if

exists under Linux, it would be displayed under Windows as

on Windows, or something like that instead of,

as they are currently displayed?

linux – A "source" PPA lists only a subset of the available packages

I added this PPA a long time ago, but when I try to install the polo-file-manager Tony George's package, it was not available when making a apt install, so I went to withdraw the app, and then I added it, but the result is still the same.

(sudo add-apt-repository --remove ppa:teejee2008/ppa)

When using a GUI tool such as Synaptic or Muon, the PPA only displays selene and timeshift, not the complete list of packages or, for that, polo-file-manager.

apt update does not display any error / warning. the associated gpg key is present in the apt folder.

What is going on here and how to fix it?

Thank you so much.

node.js – How to update versions of programs like Nodejs, Python, Git, Julia … etc in linux mint via a terminal?

Thanks for contributing to Stack Overflow in Portuguese!

  • Make sure you answer the question. Enter the details of your solution and share what you have found.

But to avoid

  • Ask for clarification or details on other responses.
  • Make statements based solely on your opinion; report references or previous experiences.

To learn more, read our tips for writing good answers.

systemd – Can I disable the GSSAPI proxy daemon without consequences? (Linux desktop)

I'm using Arch Linux on a desktop. I do not use Kerberos or Microsoft NTLM. sshd_config has this option defined:

GSSAPIAuthentication no

I do not know why the gssproxy.service is active or why it was started, but it is started and running since the last startup as shown here:

systemctl status gssproxy.service
gssproxy.service - GSSAPI Proxy Daemon
Loaded: loaded (/usr/lib/systemd/system/gssproxy.service; disabled; vendor preset: disabled)
Active: active (running) since Fri 2019-11-22 00:05:07 EST; 2 weeks 5 days ago
Main PID: 810 (gssproxy)
    Tasks: 6 (limit: 76919)
Memory: 3.2M
CGroup: /system.slice/gssproxy.service
        └─810 /usr/bin/gssproxy -D

The log does not display any results for gss other than:

systemd(1): /usr/lib/systemd/system/gssproxy.service:9: PIDFile= references a path below legacy directory /var/run/, updating /var/run/gssproxy.pid → /run/gssproxy.pid;

For a system without explicit GSSAPI authentication, is this unit necessary? If it is not necessary, what is the recommended method to prevent it from starting at startup?

The only authentication on my system is the connection to the standard Linux user account on the console and SSH via the RSA key pair. The system uses NFS v4 (also without Kerberos).

linux – How to read 3Ware disks without the controller?

I have an old 3Ware 9650SE which does not work with my current motherboard. It is recognized, but gets tons of CRC errors every time Linux tries to do something other than ask what kind of card it is.

I have a bunch of disks that contain old file systems that were connected to this card.

Is it possible to extract information from these disks without the card? I only have 4 drives and lots of free SATA ports to connect them all.

I was doing RAID5.

linux – Re-access the GUI from the terminal / command line in Trisquel

I am looking for a way to access the GUI again in Trisquel 8 after entering the terminal in full screen. I access the terminal with CTRL + ALT + f3. Once I get there, I don't know how to get back to the graphics environment. I have to restart with the help of the power button to return to the GUI.

I have tried startx and xclock (some suggestions that I found elsewhere looking for an answer to this question), but to no avail. These options seem to put me in a "no man's land". (which is another problem entirely that I could highlight in another question)

Chrome 79 released for Windows, Mac, Linux, Chrome OS, Android and iOS

Chrome 79 has just come out and it now has a phishing protection in real time.

Multiple Linux GUIs Hang – Super User

I'm working on a Lenovo Thinkpad X250, which I've upgraded to 8GB RAM and a 500GB SSD. I've been installing Kubuntu for a while, I'm currently running Manjaro, but there is also a Windows 7 partition on the device.

With Kubuntu as well as with the new Manjaro, I continue to have GUI freezes several times a day. As far as I know, it has nothing to do with the actual system load at that time, but since htop also freezes, I can not be sure. The error appears more often, when I have Firefox running.

I showed my laptop to a group of local Linux users, who tried to find the source using dmesgbut did not find any problem. Is there a way to reduce the possible source of this problem?