Gerard Henry <[EMAIL PROTECTED]> wrote:
> i have two sun ray servers configured in a failover group (sol9, SRS2.0)
> They have been rebooted, but the slave first the master, and now, there is no 
> session on slave!

That's a little unusual.  Are you sure that Sun Ray services on the slave are running 
correctly
and that the master agrees that the slave is a member of the failover group?  What does
'utgstatus' say on the master?

> Here is error message:
> 
> Jun 18 20:23:18 mombasa utpulld[527]: [ID 224068 daemon.error] Error: ldap_sasl_
> bind (host serengheti, DN cn=admin,o=utdata) returned: Can't contact LDAP server
> Jun 18 20:23:18 mombasa utpulld[527]: [ID 428662 daemon.error] utpulld: terminat
> ing normally
> 
> serengheti is master, mombasa is slave

That's OK, it just means that when the slave booted it was unable to contact the master
(because the master was still down) and it therefore couldn't pull any datastore 
updates
from the master.  When the master comes up it should push the updates to the slave.
The "pull on reboot" attempt above is just trying to get the slave into sync with the 
master
as quickly as possible.

> What 's the best way to do :
> - reboot slave mombas
> or do something like /etc/init.d/ut... restart?

Rebooting won't cause any sessions to move to the slave.  If you want to restart Sun
Ray services on the slave (which should be OK because there are no sessions there
anyway) then use '/opt/SUNWut/sbin/utrestart'.  However, the sessions that are already
on the master will stay there unless some event happens to cause the master and slave
to negotiate the placement of a new session.  That will happen whenever a new session
is requested, for instance when someone inserts a smartcard, someone attempts a
non-smartcard login, someone logs out, someone hits the 'Start Over' button on the 
login
greeter, or someone power-cycles a Sun Ray

> And i don't understand why slave needs master to work, i thought "failover" did all 
> the job?!

The slave only needs the master when there are datastore updates that need to be
performed.  Updates are always performed on the master first and then copied to the
slaves.  After a simple reboot it shouldn't matter.  Whatever the reason why the slave
didn't get any sessions, that reason has nothing to do with data store replication.  
It's
more likely the result of a misconfiguration or a network fault.

OttoM.
__
ottomeister

Disclaimer: These are my opinions.  I do not speak for my employer.

-- 
___________________________________________________________
Sign-up for Ads Free at Mail.com
http://promo.mail.com/adsfreejump.htm


_______________________________________________
SunRay-Users mailing list
[EMAIL PROTECTED]
http://www.filibeto.org/mailman/listinfo/sunray-users

Reply via email to