Hi,
nothing unusual in audit.log (selinux set to permissive):
----
time->Fri Nov 16 14:24:08 2012
type=SERVICE_START msg=audit(1353061448.978:98): pid=1 uid=0
auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='
comm="osad" exe="/usr/lib/systemd/systemd" hostname=? addr=?
terminal=? res=success'
----
time->Fri Nov 16 14:24:09 2012
type=SERVICE_STOP msg=audit(1353061449.608:99): pid=1 uid=0
auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='
comm="osad" exe="/usr/lib/systemd/systemd" hostname=? addr=?
terminal=? res=success'
----
time->Fri Nov 16 14:24:53 2012
type=SERVICE_START msg=audit(1353061493.899:100): pid=1 uid=0
auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='
comm="osad" exe="/usr/lib/systemd/systemd" hostname=? addr=?
terminal=? res=success'
----
time->Fri Nov 16 14:24:56 2012
type=SERVICE_STOP msg=audit(1353061496.924:101): pid=1 uid=0
auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='
comm="osad" exe="/usr/lib/systemd/systemd" hostname=? addr=?
terminal=? res=success'
Don't think it's time sync related - time is synced and from same source.
And i'm able to connect to jabberd directly via telnet.
I've done some more testes, and it appears that osad stays alive and
working *only* on first registration in jabberd (i.e when new machine
profile is registered in spacewalk or after cleanup of
/var/lib/jabberd/db/*).
For example - if i kickstart new fedora17 VM, it will register in SW and
osad works fine, till reboot. After reboot - it just dies with words
"pushing into background". After cleanup of /var/lib/jabberd/db* (in
most cases, but not in all - sometimes i need to do it twice/triple
times) - osad on fedora client starts and stays alive, again - till
reboot or restart of osad
in /var/log/messages on spacewalk-server i see this kind of picture,
first time:
Nov 16 14:44:14 spacewalk jabberd/c2s[30499]: [13] [192.168.10.52,
port=34891] connect
Nov 16 14:44:14 spacewalk jabberd/c2s[30499]: [13] created user:
user=osad-599e6b58fd; realm=
Nov 16 14:44:14 spacewalk jabberd/c2s[30499]: [13] registration
succeeded, requesting user creation:
[email protected]
Nov 16 14:44:14 spacewalk jabberd/sm[30492]: created user:
[email protected]
Nov 16 14:44:14 spacewalk jabberd/c2s[30499]: [13] legacy
authentication succeeded: host=, username=osad-599e6b58fd,
resource=osad, TLS negotiated
Nov 16 14:44:14 spacewalk jabberd/c2s[30499]: [13] requesting session:
[email protected]/osad
Nov 16 14:44:14 spacewalk jabberd/sm[30492]: session started:
[email protected]/osad
Nov 16 14:46:20 spacewalk jabberd/c2s[30499]: [13] [192.168.10.52,
port=34891] disconnect
[email protected]/osad, packets: 12
Nov 16 14:46:20 spacewalk jabberd/sm[30492]: session ended:
[email protected]/osad
and after reboot or restart of osad:
Nov 16 14:46:43 spacewalk jabberd/c2s[30499]: [13] [192.168.10.52,
port=34898] connect
Nov 16 14:46:43 spacewalk jabberd/c2s[30499]: [13] [192.168.10.52,
port=34898] disconnect jid=unbound, packets: 0
Yuriy Demchenko
On 11/13/2012 03:29 PM, Jan Pazdziora wrote:
On Sat, Nov 10, 2012 at 05:06:15PM +0400, Yuriy Demchenko wrote:
Hi,
I'm having troubles with OSA daemon on fedora 17 machines - it doesnt work.
using osad.noarch 5.11.10-1.fc17 from spacewalk-1.8-client fedora17
repository and spacewalk 1.8
when i try to start it, I'm getting this:
[root@fedtest ~]# systemctl status osad.service
osad.service - OSAD daemon
Loaded: loaded (/usr/lib/systemd/system/osad.service; enabled)
Active: failed (Result: exit-code) since Sat, 10 Nov
2012 16:46:35 +0400; 35s ago
Process: 26886 ExecStart=/usr/sbin/osad --pid-file
$PIDFile (code=exited, status=2)
CGroup: name=systemd:/system/osad.service
Nov 10 16:46:35 fedtest osad[26886]: Usage: osad [options]
Nov 10 16:46:35 fedtest osad[26886]: osad: error: --pid-file
option require
error message pretty obvious, so I've changed systemd file
osad.service and put full path instead of var $PIDFile
now systemd starts service
That change sounds correct.
but it dies in a second:
[root@fedtest ~]# systemctl status osad.service
osad.service - OSAD daemon
Loaded: loaded (/usr/lib/systemd/system/osad.service; enabled)
Active: inactive (dead) since Sat, 10 Nov 2012 16:56:05
+0400; 1min 13s ago
Process: 27155 ExecStart=/usr/sbin/osad --pid-file
/var/run/osad.pid (code=exited, status=0/SUCCESS)
CGroup: name=systemd:/system/osad.service
I've turned on debug logging, but found nothing useful in log -
process just dies after "pushing into background" and no more
messages in log:
2012-11-10 16:56:05 jabber_lib._auth_dispatch: <features><address
xmlns = 'http://affinix.com/jabber/address'
192.168.10.50</address><mechanisms xmlns =
'urn:ietf:params:xml:ns:xmpp-sasl'
><mechanism>EXTERNAL</mechanism><mechanism>PLAIN</mechanism><mechanism>DIGEST-MD5</mechanism></mechanisms><auth
xmlns = 'http://jabber.org/features/iq-auth' /><register xmlns =
'http://jabber.org/features/iq-register' /></features>
2012-11-10 16:56:05 jabber_lib.connect: connect returning
2012-11-10 16:56:05 jabber_lib.setup_connection: Connected to
jabber server spacewalk.spb.stone.local
2012-11-10 16:56:05 jabber_lib.push_to_background: Pushing process
into background
Is there anything in /var/log/audit/audit.log, by any chance?
I've tried osad-5.10.41-1.fc16.noarch from spacewalk-1.7-client repo
- it starts/stops fine via sysV, but doesnt work too - somehow it
cant connect to jabber server (firewalls and selinux disabled, both
spacewalk-server and machine have fully resolvable fqdn).
jabber_lib.main: Unable to connect to jabber servers, sleeping 105 seconds
Could it be time sync issue (if using SSL)? Are you able to connect to
that jabberd directly, with nc or telnet?
_______________________________________________
Spacewalk-list mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/spacewalk-list
