All Questions(server)

I have an rsyslog server configured to accept remote messages from around ~400 servers, however it seems once the server gets about ~256 connections it stops accepting new connections and the clients see error messages like the following

cannot connect to syslog.example.org:514: Connection timed out

The rsyslog imtcp module is configured as follows

module(load="imtcp"
          MaxSessions="1000"
          StreamDriver.Mode="1"
          StreamDriver.AuthMode="anon"
          StreamDriver.Name="gtls"

)

the systemd unit is also configured with LimitNOFILE=16384 and when i check /proc/$pid/limits i see that the number of limits looks good

$ /proc/2767537# cat limits
Limit                     Soft Limit           Hard Limit           Units
Max cpu time              unlimited            unlimited            seconds
Max file size             unlimited            unlimited            bytes
Max data size             unlimited            unlimited            bytes
Max stack size            8388608              unlimited            bytes
Max core file size        0                    unlimited            bytes
Max resident set          unlimited            unlimited            bytes
Max processes             7393                 7393                 processes
Max open files            16384                16384                files
Max locked memory         65536                65536                bytes
Max address space         unlimited            unlimited            bytes
Max file locks            unlimited            unlimited            locks
Max pending signals       7393                 7393                 signals
Max msgqueue size         819200               819200               bytes
Max nice priority         0                    0
Max realtime priority     0                    0
Max realtime timeout      unlimited            unlimited            us

however when i start seeing this error the number of files in /proc/$pid/fd is close to 256.

/proc/2767537$ ls -l fd | wc -l
253

Further to this the established connections always close to 256

$ ss -ntp4 state established sport eq 514 | wc -l
257

this suggest to me that there is some other limit of 256 i need to tweak but im unable to find what/where.

• I can't see any errors i the logs on the syslog server
• using tcpdump I can see the syn packet come in on port 514 but no sys/ack is ever sent

I'm trying to configure DNS for handling mail across multiple subdomains, where each subdomain has its own mail server. My goal is to route mail for user@%subdomain%.example.com to the server hosted under %subdomain%.example.com.

Example of desired configuration:

• Mail for [email protected] should be handled by abcde.example.com server
• Mail for [email protected] should be handled by xyz.example.com server

And so on for all possible subdomains. There are many subdomains, so manually adding an MX record for each one is not an option.

What might work

_smtp._tcp.*.example.com. IN SRV 0 0 25 %.example.com.

This configuration looks logical, but I'm not sure how well it's supported by mail servers. I don't know if something similar is possible for IMAP and POP3.

My questions

1. What DNS records need to be created to implement such a routing?
2. Is it possible to set up a wildcard MX record that will automatically route mail to the corresponding subdomain?
3. If a wildcard MX record is not possible, how can the process of creating MX records for new subdomains be automated (preferably in a way that doesn't take much time, perhaps using dynamic DNS)?

What doesn't work

• Wildcard routing of all subdomains to a single domain, like IN MX 10 central-mail.example.com.. Such a configuration is only acceptable if the email can then be routed to the correct server without needing to decrypt the packet content (like SNI in TLS/SSL).

I would appreciate a detailed explanation of the necessary DNS records and any additional recommendations.

I want to add a static IP to my ECS Fargate service. I am currently able to setup the Elastic IP with a Network Load Balancer but I don't want to add any load balancer.

• Why the AWS knowledge center suggests using a Network Load Balancer for assigning Elastic IP to Fargate service?
• What are the approaches to add a static Elastic IP to my Fargate service, without using a Load Balancer?

I am somewhat newb to Linux but this is my 5th or so Linux VM I've ever setup so I have some experience here. I have an Ubuntu 24 VM running in Oracle VirtualBox running in a Windows 11, 8 core, 32GB RAM PC. The Gala TownStar v2 Node that I am running on this PC requires a 2 core, 8GB RAM, 40GB hard drive space to run and so I setup the VM with exactly those specs. I have this VM bridged on the network so that it has its own IP because I also am running another instance of that node on that actual PC, not the VM. So the PC has the node running, then in a VM running on that same PC, I have another instance of that node running.

This seems to work well but once or twice a day I notice that the node running in the VM is down! When I go to the VM I just run sudo gala-node status and that is all it takes for the node to come back alive. The node running directly on my PC has never gone down. So daily I go to the dashboard and when it's down, I run that command... but this means it could be hours before I notice. Every so often when I go back to the VM to run that command, I see Ubuntu with an error that it wants to report, when I look at the details, the message is

systemd-logind crashed with SIGABRT

I am not certain that this error is the reason for the node going down, but I have nothing else to go on. I found this page with some discussion about it and possible solutions but none of that seemed to work for me. Someone suggested I go back to Ubuntu 20.04, but before I spend time on that, does anyone have any suggestions? BTW, I tried running an "every 5 min" crontab to run that status command, but that either doesn't work or I am not certain that it's doing the thing. sudo crontab -l shows that */5 * * * * sudo gala-node status is there, but I'd love for it to spit out the output to my console, I think I could have the output of that crontab go to a file, so that's probably my next step.

Edit: Installed Ubuntu 20.04 and am now getting much more frequent error of the same type, like every 5 minutes! The exact error message there is: "systemd-resolved" crashed with SIGBRT in epoll_wait() Also a recent output from one of the many times it crashed, Ubuntu 20.04:

I have a device that is failing. Using ddrescue, I have recovered all all but about 500 KB in 83 regions. The map file tells me the raw offset and length of each lost region. I want to find out what files are affected. How can I obtain a list of physical sector offsets associated with a file on an exFAT volume?