tasks/internal/syslog: gather all journald entries

Soemtimes a machine crashes during tests and log entries around that time are
not persisted because we're only looking at this boot.

(This choice seem odd since the lifetime of these operating system images is
the same as the teuthology run.)

Now looks like:

    2025-07-21T15:01:22.249 INFO:teuthology.task.internal.syslog:Gathering journactl...
    2025-07-21T15:01:22.250 DEBUG:teuthology.orchestra.run.smithi045:> sudo journalctl > /home/ubuntu/cephtest/archive/syslog/journalctl.log
    2025-07-21T15:01:22.254 DEBUG:teuthology.orchestra.run.smithi083:> sudo journalctl > /home/ubuntu/cephtest/archive/syslog/journalctl.log
    2025-07-21T15:01:22.256 DEBUG:teuthology.orchestra.run.smithi184:> sudo journalctl > /home/ubuntu/cephtest/archive/syslog/journalctl.log

From: /teuthology/pdonnell-2025-07-21_14:01:58-fs-main-distro-default-smithi/8399938/teuthology.log

Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>

diff --git a/teuthology/task/internal/syslog.py b/teuthology/task/internal/syslog.py
index 64032a8e7ce980173fd0dc263c18418800bc39b9..955e933f8cedb24c6bc99dbbe39541290968d8c1 100644 (file)
--- a/teuthology/task/internal/syslog.py
+++ b/teuthology/task/internal/syslog.py
@@ -181,11 +181,11 @@ def syslog(ctx, config):
             )
         )
 
-        log.info('Gathering journactl -b0...')
+        log.info('Gathering journactl ...')
         run.wait(
             cluster.run(
                 args=[
-                    'sudo', 'journalctl', '-b0',
+                    'sudo', 'journalctl',
                     run.Raw('|'),
                     'gzip', '-9',
                     run.Raw('>'),