Bug 1967962
| Summary: | Annoying error message "Failed to create session: Start job for unit user-0.slice failed with 'canceled'" when sudoing rapidly | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 8 | Reporter: | Renaud Métrich <rmetrich> |
| Component: | systemd | Assignee: | David Tardon <dtardon> |
| Status: | CLOSED CURRENTRELEASE | QA Contact: | Frantisek Sumsal <fsumsal> |
| Severity: | high | Docs Contact: | |
| Priority: | high | ||
| Version: | 8.4 | CC: | broose, chorn, dtardon, msekleta, qguo, systemd-maint-list |
| Target Milestone: | rc | Keywords: | Triaged |
| Target Release: | --- | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2023-05-18 14:42:10 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: | |||
git bisect shows this is fixed by https://github.com/systemd/systemd/pull/9824/commits/25a1ab4ed48b72e974f77a68dcbe3521014787bb . If it still fails with http://people.redhat.com/dtardon/systemd/bz1642460-backport-UserStopDelaySec=/, then I've either made a mistake in the backport or missed an additional PR it relies upon. - I reproduced https://bugzilla.redhat.com/show_bug.cgi?id=1967962#c0 on rhel8.2GA - reproduces on rhel8.3 systemd, but not on rhel8.4 systemd and later. 8.2 systemd-239-29 issue 8.3 systemd-239-40 issue 8.4 systemd-239-45 fixed 8.5 systemd-239-51 fixed I see that commit https://github.com/systemd/systemd/pull/9824/commits/25a1ab4ed48b72e974f77a68dcbe3521014787bb has quite some code changes, and also 239-51 does not have all the changes. I vote for declaring that systemd-239-45 have the fixes, and closing this bz as CURRENT. Opinions? Concluding this is fixed in rhel8.5/systemd-239-51.el8. bz1892833 is also around this issue. (In reply to Christian Horn from comment #3) > Concluding this is fixed in rhel8.5/systemd-239-51.el8. Let's close it then. |
Description of problem: When performing a lot of quick sudo sessions in a row, the following message is seen in the journal at Error level: -------- 8< ---------------- 8< ---------------- 8< ---------------- 8< -------- [...] sudo[42836]: pam_systemd(sudo:session): Failed to create session: Start job for unit user-0.slice failed with 'canceled' -------- 8< ---------------- 8< ---------------- 8< ---------------- 8< -------- This message should be logged at Debug level only (or even removed) because actually there is **no issue at all**. To do this, systemd-logind needs to be smarter: -------- 8< ---------------- 8< ---------------- 8< ---------------- 8< -------- 2744 static int session_jobs_reply(Session *s, const char *unit, const char *result) { : 2753 if (streq(result, "done")) 2754 r = session_send_create_reply(s, NULL); 2755 else { 2756 _cleanup_(sd_bus_error_free) sd_bus_error e = SD_BUS_ERROR_NULL; 2757 2758 sd_bus_error_setf(&e, BUS_ERROR_JOB_FAILED, "Start job for unit %s failed with '%s'", unit, result); 2759 r = session_send_create_reply(s, &e); 2760 } -------- 8< ---------------- 8< ---------------- 8< ---------------- 8< -------- On line 2758, it should test if "result" is "canceled" or not and ignore the latter. Version-Release number of selected component (if applicable): systemd-239 How reproducible: Very easily Steps to Reproduce: 1. Create a script that sudo'es repeatedly /usr/local/bin/sudo_loop: -------- 8< ---------------- 8< ---------------- 8< ---------------- 8< -------- #!/bin/sh while :; do sudo touch /tmp/foo sudo rm /tmp/foo sleep 0.1 done -------- 8< ---------------- 8< ---------------- 8< ---------------- 8< -------- 2. Make sure there is no root session and start the script as a service from an admin account -------- 8< ---------------- 8< ---------------- 8< ---------------- 8< -------- $ sudo systemctl stop user-0.slice $ sudo systemd-run --uid=user --unit=sudo_loop.service /usr/local/bin/sudo_loop -------- 8< ---------------- 8< ---------------- 8< ---------------- 8< -------- 3. Monitor the journal from the admin account -------- 8< ---------------- 8< ---------------- 8< ---------------- 8< -------- $ journalctl --follow | tee journal.txt | grep "Failed to create session" -------- 8< ---------------- 8< ---------------- 8< ---------------- 8< -------- Actual results: -------- 8< ---------------- 8< ---------------- 8< ---------------- 8< -------- Jun 04 16:24:02 vm-rhel8 sudo[42836]: pam_systemd(sudo:session): Failed to create session: Start job for unit user-0.slice failed with 'canceled' Jun 04 16:24:03 vm-rhel8 sudo[42861]: pam_systemd(sudo:session): Failed to create session: Start job for unit user-0.slice failed with 'canceled' Jun 04 16:24:04 vm-rhel8 sudo[42945]: pam_systemd(sudo:session): Failed to create session: Start job for unit user-0.slice failed with 'canceled' Jun 04 16:24:05 vm-rhel8 sudo[42999]: pam_systemd(sudo:session): Failed to create session: Start job for unit user-0.slice failed with 'canceled' Jun 04 16:24:06 vm-rhel8 sudo[43025]: pam_systemd(sudo:session): Failed to create session: Start job for unit user-0.slice failed with 'canceled' Jun 04 16:24:06 vm-rhel8 sudo[43049]: pam_systemd(sudo:session): Failed to create session: Start job for unit user-0.slice failed with 'canceled' : -------- 8< ---------------- 8< ---------------- 8< ---------------- 8< -------- Expected results: No error message at all (there is no issue at all ...). Additional info: I tested the suggestion in BZ #1892833 (build http://people.redhat.com/dtardon/systemd/bz1642460-backport-UserStopDelaySec=/) but this doesn't help.