set -e (Exit on Error) and Its Gotchas

#!/bin/bash
(
    set -e
    echo "step 1"
    false
    echo "step 2 - never reached"
) || echo "the subshell stopped early because of set -e"

#!/bin/bash
(
    set -e
    if false; then
        echo "unreachable"
    else
        echo "set -e did NOT trigger here, because false was an if-condition"
    fi
    echo "the subshell kept running normally"
)

#!/bin/bash
(
    set -e
    false | true
    echo "reached because only the last command (true) mattered to set -e"
)

(
    set -e -o pipefail
    false | true
    echo "unreachable with pipefail active"
) || echo "with pipefail, the earlier failure was noticed and stopped the subshell"

#!/bin/bash
(set -e; false; echo "unreached") || echo "caught: the previous command failed under set -e"
echo "the main script keeps running normally"

1. Assuming set -e catches every possible failure; it does NOT trigger for a failing command inside an if/while condition, before a &&/||, or as any-but-the-last part of a pipeline.
2. Being surprised a script keeps running after a failing command inside a function call used in a condition, since set -e's exemptions apply there too.
3. Relying on set -e alone for robust error handling instead of combining it with explicit checks for the specific commands that matter most.

• set -e makes the script exit immediately if a command exits non-zero, instead of continuing to the next line.
• set -e does NOT apply to a command being tested by if, while, until, or before &&/|| — those are expected to potentially fail.
• In a pipeline, set -e only reacts to the LAST command's exit status, unless pipefail is also set.
• set -e is best combined with set -u and set -o pipefail for more predictable strict-mode behavior.