Posts on XGA

MikroTik LTE modem configuration generator

Mon, 06 Jul 2026 00:00:00 +0000

AI Disclaimer Link to heading

This tool was generated using Claude. The outputs from the tool were manually verified against existing tooling, then tested on a physical device before deployment.

Where this project started Link to heading

The remote monitoring stack I manage utilises LTE modems which are connected to the device to be monitored.

Each modem must be manually configured and tested before being shipped to site. Having a standardised configuration provides consistent troubleshooting steps if a device doesn’t work as planned.

Testing PGP email signing and encryption

Fri, 03 Jul 2026 00:00:00 +0000

What is this post? Link to heading

This post contains a few hints to prove PGP encryption is working on Thunderbird or your PGP mail client of choice. This post is not a guide for configuring PGP or Thunderbird.

Background Link to heading

Email isn’t secure and we shouldn’t trust a thing we receive. We should know that by now. If you don’t, see this article from Jim Salter.

Monitor Restic backups with Healthchecks.io

Sat, 27 Jun 2026 00:00:00 +0000

This script is used on Linux hosts to complete additional offsite file backups as part of our larger DR plan. It has several features:

File backup and prune jobs
Integrated monitoring on Healthchecks.io, including start, success and failure messages.
Multiple attempts to complete both backups and prunes in case of failure.

#!/bin/bash
eval `ssh-agent -s`
ssh-add /root/.ssh/rem-bkups
export RESTIC_REPOSITORY="sftp:username@host:/data/directory"
export RESTIC_PASSWORD="12345"
HC="https://hc-ping.com/12345"

MAX_ATTEMPTS=3
RETRY_DELAY=60 # seconds between attempts

# Generic retry wrapper: retries the given command up to MAX_ATTEMPTS times
run_with_retries() {
 local label="$1"; shift
 local attempt=1
 while [ "$attempt" -le "$MAX_ATTEMPTS" ]; do
 echo "$label: attempt $attempt of $MAX_ATTEMPTS at $(date)"
 if "$@"; then
 echo "$label succeeded on attempt $attempt"
 return 0
 fi
 echo "$label failed on attempt $attempt"
 attempt=$((attempt + 1))
 [ "$attempt" -le "$MAX_ATTEMPTS" ] && sleep "$RETRY_DELAY"
 done
 echo "$label failed after $MAX_ATTEMPTS attempts"
 return 1
}

do_backup() {
 restic backup \
 /home \
 /etc \
 /opt \
 /var/www \
 --verbose
}

do_prune() {
 restic forget \
 --keep-last 1 \
 --keep-daily 2 \
 --keep-weekly 3 \
 --keep-monthly 4 \
 --prune
}

echo "Starting system backup at $(date)"
curl -fsS "$HC/start" >/dev/null 2>&1

# Only signal HC success if BOTH backup and prune succeed
if run_with_retries "Backup" do_backup && run_with_retries "Prune" do_prune; then
 echo "System backup completed at $(date)"
 curl -fsS "$HC" >/dev/null 2>&1
else
 echo "System backup or prune failed at $(date)"
 curl -fsS "$HC/fail" >/dev/null 2>&1
 exit 1
fi

Incorrect timezones on new Teams and Outlook accounts

Mon, 13 Apr 2026 00:00:00 +0000

When adding new users to a Microsoft tenant, it seems to be creating the users with Pacific time, giving incorrect working hours on Teams and causing issues with Outlook calendar events

Connect from a non-admin PS prompt:

Import-Module -Name ExchangeOnlineManagement
Connect-ExchangeOnline

Check time-zones for all users:

Get-Mailbox | Where {$_.RecipientTypeDetails -eq "UserMailbox"} | get-MailboxCalendarConfiguration

Check timezone for single user:

Get-MailboxCalendarConfiguration -Identity $emailAddr

Set the timezone for a user:

Set-MailboxCalendarConfiguration -Identity $user -WorkingHoursTimeZone "GMT Standard Time"

Changes take effect fairly quickly.

External Link: A320 Virtual Cockpit ✈

Wed, 01 Jan 2025 00:00:00 +0000

External Link: A330 Virtual Cockpit ✈

Wed, 01 Jan 2025 00:00:00 +0000

Zabbix DNS Caching

Mon, 30 Dec 2024 00:00:00 +0000

DNS resolution started to feel a bit laggy recently, most notable on iOS devices. When I went down the troubleshooting rabbit-hole, I found a large number of DNS queries from the Zabbix monitoring infrastructure.

Of the 37 hosts that are monitored by Zabbix, only 16 have DNS lookups for the hostnames. This caused multiple lookups per second for each host - thousands of entries per 10 minutes.

Effect of DNS caching

Missing snapshot on Hik-Connect

Tue, 10 Dec 2024 00:00:00 +0000

A recent project had us reset a Hikvision video intercom system after an upgrade and created a new issue that we hadn’t seen before.

The client encountered an issue where snapshots were no longer sent to the Hik-Connect application when a caller pressed the button on the video intercom.

This can be resolved by changing a setting on the master indoor station:

Go to Settings -> Preference -> Shortcut Setting -> Enable Snap Shot.

High BGP CPU usage on Mikrotik CHR

Thu, 04 May 2023 00:00:00 +0000

While experimenting with BGP using Mikrotik Cloud Hosted Routers (CHR) on low-power, shared X86 hardware, it became apparent that CPU usage was an issue with full feeds.

Before Link to heading

The router was handling approximately 420-thousand IPv6 routes on a single CPU. When I tried to view the routes in WinBox or SSH, the CPU would go to 100% and remain there, leaving the router unresponsive until rebooted.