Linux Service Operations
65 incident problems about Linux Service Operations. Start with the reviewed ones.
Read first
How to split Linux failures by systemd, permission, and filesystem signalsAn InfraTree guide that lays out the first signals to check, the CLI verification order, common misdiagnoses, and a safe recovery path when a service fails but you need to isolate whether the cause is a systemd unit, permission, inode, mount, or process.Linux3 min readWhen sudo works in the shell but fails under automationAn InfraTree guide that lays out the first signals to check, the CLI verification order, common misdiagnoses, and a safe recovery path when only automation jobs fail because of TTY, sudoers, environment reset, or service user differences.Linux3 min readWhen permissions are fixed but only new directories are created with the wrong groupAn InfraTree guide that lays out the first signals to check, the CLI verification order, common misdiagnoses, and a safe recovery path when existing file permissions are correct but group inheritance breaks for newly created files and directories.Linux3 min read
Recommended problems
Reviewed problems first, then problems with detailed scenarios.
LINUX-005A missing WorkingDirectory problem that puts a systemd service into a restart loopA situation where the application itself is fine but a wrong unit-file path setting kills the service.ReviewedLinuxIntermediate20 minFreeLINUX-010A missing environment variable that fails under cron but works when run manuallyA problem for checking which defaults are dropped when the same script runs in a different execution environment.ReviewedLinuxBeginner15 minFreeLINUX-096NTP is synchronized but the hardware clock drifts back after every reboot and breaks certificatesThe running host time looks correct, but restarts reintroduce skew because the persistent hardware clock was never aligned.ReviewedLinuxIntermediate14 minFreeEADDRINUSE (address already in use): the new version will not startEADDRINUSE (address already in use): the new version will not start is a hands-on troubleshooting drill. Find which process already holds the port. Linux Service Operations needs to be checked by narrowing scope, recent change, and the current live signal before rollback. 실무에서...ReviewedLinuxBeginner3 minFreeLINUX-008It looked like a memory leak, but it was really an orphan process that never terminatedAnalyzes a situation where a side worker outlives the main process and holds on to resources.LinuxIntermediate21 minFreeLINUX-011A bind-address problem where the service port is open but only reachable from localhostA bind-address problem where the service port is open but only reachable from... is a hands-on troubleshooting drill. A situation where the application is up but the listen address is wrong, so external connections fail. Linux Service Operations needs to be checked by narrowin...LinuxBeginner15 minFree
All problems (65)
EADDRINUSE (address already in use): the new version will not startEADDRINUSE (address already in use): the new version will not start is a hands-on troubleshooting drill. Find which process already holds the port. Linux Service Operations needs to be checked by narrowing scope, recent change, and the current live signal before rollback. 실무에서...ReviewedLinuxBeginner3 minFreeLINUX-096NTP is synchronized but the hardware clock drifts back after every reboot and breaks certificatesThe running host time looks correct, but restarts reintroduce skew because the persistent hardware clock was never aligned.ReviewedLinuxIntermediate14 minFreeLINUX-005A missing WorkingDirectory problem that puts a systemd service into a restart loopA situation where the application itself is fine but a wrong unit-file path setting kills the service.ReviewedLinuxIntermediate20 minFreeLINUX-010A missing environment variable that fails under cron but works when run manuallyA problem for checking which defaults are dropped when the same script runs in a different execution environment.ReviewedLinuxBeginner15 minFreeLINUX-011A bind-address problem where the service port is open but only reachable from localhostA bind-address problem where the service port is open but only reachable from... is a hands-on troubleshooting drill. A situation where the application is up but the listen address is wrong, so external connections fail. Linux Service Operations needs to be checked by narrowin...LinuxBeginner15 minFreeLINUX-016A shell script where set -e skips even the normal cleanup stepA situation where the failure-handling intent was good, but an overly aggressive exit condition prevents even the recovery routine from running.LinuxBeginner16 minFreeLINUX-014A backup script that looks successful but actually keeps only an empty directoryCovers an incident where the backup output is empty because of a difference between the relative path and the current working directory.LinuxIntermediate18 minFreeLINUX-008It looked like a memory leak, but it was really an orphan process that never terminatedAnalyzes a situation where a side worker outlives the main process and holds on to resources.LinuxIntermediate21 minFreeLINUX-089rsync backup over SSH stallsNetwork reachability is fine, but the batch backup never starts because the negotiated SSH crypto overlap disappeared during a hardening change.LinuxIntermediate16 minFreeLINUX-091Tmpfs mount survives the deploy but vanishes after rebootThe application worked during the change window, then failed after restart because the critical mount existed only as a temporary runtime action.LinuxIntermediate14 minFreeLINUX-071logrotate copytruncate races with a high-write process and drops critical linesThe rotation completes, but incident evidence is missing because copytruncate cut the file while the application was still writing heavily.LinuxIntermediate15 minFreeLINUX-076tmpfiles.d recreates the runtime directory with the wrong owner after rebootPermissions look correct before restart, but the service fails after boot because tmpfiles recreated the path with ownership that no longer matches the daemon user.LinuxIntermediate15 minFreeLINUX-061journald rate limiting drops repeated kernel alerts during a packet stormThe host is noisy during the incident, but the most useful log lines never reach disk because journald suppression hides repeated bursts.LinuxIntermediate16 minFreeLINUX-099Emergency reboot script runs from cron but loses access to the package manager lock namespace inside systemd-runThe fallback script exists, yet repair automation still fails because the execution environment differs from the shell where the operator tested it.LinuxIntermediate18 minFreeLINUX-051systemd-tmpfiles cleanup removes the application socket directory after rebootThe service worked before restart, but after boot its runtime socket path is missing because the tmpfiles policy cleaned the directory more aggressively than expected.LinuxIntermediate18 minFreeLINUX-030Process limit reached by short-lived workers instead of main appThe service owner blames the main process, but the true limit is exhausted by a burst of helper processes spawned underneath.LinuxIntermediate19 minFreeLINUX-032Systemd service starts manually but fails on boot without network-online targetThe unit runs when started interactively, yet boot-time startup fails because dependency ordering assumes the network is ready earlier than it actually is.LinuxIntermediate19 minFreeLINUX-074Boot stalls on a dead NFS mountThe host itself is healthy, but boot takes far too long because a remote mount is treated like a local dependency and blocks the sequence.LinuxBeginner13 minFreeLINUX-024File exists but service user cannot follow parent directoryFile exists but service user cannot follow parent directory is a hands-on troubleshooting drill. The file permission looks fine, but one directory in the path denies execute permission and breaks access. Azure Linux Service Operations needs to be checked by narrowing scope, re...LinuxBeginner14 minFreeLINUX-060Package manager lock remains after a dead process and blocks patch automationThe scheduled patch job says another package process is running, but the real blocker is an abandoned lock or stale transaction state from an older failure.LinuxBeginner14 minFree