Bug 1714373
| Summary: | systemd-cat output not associated with the unit that generated it | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Robin A. Meade <robin.a.meade> |
| Component: | systemd | Assignee: | systemd-maint |
| Status: | CLOSED DUPLICATE | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | unspecified | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 30 | CC: | lnykryn, msekleta, ssahani, s, systemd-maint, zbyszek |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2019-07-21 09:39:53 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
Possible explanation: https://bbs.archlinux.org/viewtopic.php?pid=1326954#p1326954 "The reason for this is a race condition, because the logger/systemd-cat process dies before journald can get the information it needs." -- September 2013 I integrated the workaround offered in the above archlinux.org forum and it worked:
File: /usr/bin/test_service.sh
```
# The script: /usr/bin/test_service.sh
sudo tee /usr/bin/test_service.sh > /dev/null << "EOF"
# For info messages:
info=/tmp/myservice-systemd-cat-pipe-info
mkfifo "$info"
trap "exec 3>&-; rm $info" EXIT
systemd-cat -p info < "$info" &
exec 3>"$info"
# For warning messages:
warning=/tmp/myservice-systemd-cat-pipe-warning
mkfifo "$warning"
trap "exec 4>&-; rm $warning" EXIT
systemd-cat -p warning < "$warning" &
exec 4>"$warning"
# For err messages:
err=/tmp/myservice-systemd-cat-pipe-err
mkfifo "$err"
trap "exec 5>&-; rm $err" EXIT
systemd-cat -p err < "$err" &
exec 5>"$err"
echo "This is an information message" > "$info"
echo "This is an warning message" > "$warning"
echo "This is an error message" > "$err"
DATE=`date '+%Y-%m-%d %H:%M:%S'`
echo "Example service started at ${DATE}" > "$info"
while :
do
echo "Looping...";
sleep 30;
done
EOF
```
Yep, not much we can do unless the kernel helps us. *** This bug has been marked as a duplicate of bug 963620 *** |
Description of problem: I tried a simple tutorial [1] and didn't get the expected behavior from systemd-cat: File: /lib/systemd/system/myservice.service ``` [Unit] Description=Example systemd service. [Service] Type=simple ExecStart=/bin/bash /usr/bin/test_service.sh [Install] WantedBy=multi-user.target ``` File: /usr/bin/test_service.sh ``` DATE=`date '+%Y-%m-%d %H:%M:%S'` echo "Example service started at ${DATE}" | systemd-cat -p info while : do echo "Looping..."; sleep 30; done ``` Then I started and got status: ``` sudo systemctl start myservice sudo systemctl status myservice ``` EXPECTED: I expcted to see the "Example service started at ..." line in the output of the status command, but did not. I did see it in the output of `sudo journalctl -f` but not `sudo journalctl -f myservice` Thus it seemms the output of systemd-cat is not getting associated with the unit. Version-Release number of selected component (if applicable): $ systemd-cat --version systemd 241 (v241-8.git9ef65cb.fc30) +PAM +AUDIT +SELINUX +IMA -APPARMOR +SMACK +SYSVINIT +UTMP +LIBCRYPTSETUP +GCRYPT +GNUTLS +ACL +XZ +LZ4 +SECCOMP +BLKID +ELFUTILS +KMOD +IDN2 -IDN +PCRE2 default-hierarchy=hybrid [1] https://www.linode.com/docs/quick-answers/linux/start-service-at-boot/