Dealing With Flakey CI Commands With a Retry Loop in Bash
One of the most frustrating things to deal with in Continuous Integration is flakey commands. Whether it’s flakey tests, or intermittent networking issues, when your build fails for issues outside of your control not only does it cause frustration, it reduces the trust in your CI process.
One strategy for dealing with this type of issue is to introduce some retry logic into your commands. This can easily be accomplished with good old bash.
For example, pretend that I have $FLAKEY_COMMAND and I want to retry it three times before finally failing my build. I could wrap the whole thing up in a bash loop like this.
counter=1
max=3
$FLAKEY_COMMAND
while [[ $? -ne 0 && $counter -lt $max ]]; do
counter=$((counter+1))
$FLAKEY_COMMAND
done
This script will run my command, if the exit code (the output of $?) is non zero (i.e something went wrong) and my counter is less than three, then it will retry the command. You can increase or decrease the number of attempts by adjusting the max variable.
This is not a foolproof strategy, but is one approach to handle flakey commands in your CI pipeline.
Thank you for reading! Share your thoughts with me on bluesky, mastodon, or via email.
Check out some more stuff to read down below.
Most popular posts this month
- 2024
- Reinstalling Windows at 1am
- SQLite DB Migrations with PRAGMA user_version
- My Custom Miniflux CSS Theme
- How to Disable Wayland in Debian Testing
Recent Favorite Blog Posts
This is a collection of the last 8 posts that I bookmarked.
- workflows for ai coding from /* 🤖🛠️ */
- Reinvent the Wheel from Matthias Endler
- CAPTCHAs are over (in ticketing) from pretix – behind the scenes
- You Can Choose Tools That Make You Happy from Fernando Borretti
- Using Git as S3 from Kris Tun
- Future Fonts from Blog – Brad Frost
- 21st Century C++ from Communications of the ACM
- Submarines DevCon 2025 Keynote Speech from JoshHaines.com
Articles from blogs I follow around the net
Artificial Unintelligence – We are so inured to using AI that we believe we need it
Before people around the world were connected by the internet the proliferation of new ideas and modern technologies was slow. This was bad for the originators who had no means of quickly gauging…
via V.H. Belvadi June 1, 2025How often do LLMs snitch? Recreating Theo's SnitchBench with LLM
A fun new benchmark just dropped! Inspired by the Claude 4 system card - which showed that Claude 4 might just rat you out to the authorities if you told it to "take initiative" in enforcing its morals values while exposing it to evidence of malfe…
via Simon Willison's Weblog: Entries May 31, 2025OpenAI featured chatbot is pushing extreme surgeries to “subhuman” men
OpenAI's featured chatbot recommends $200,000 in surgeries while promoting incel ideology
via Citation Needed May 31, 2025Generated by openring