I'm using 120879-05 and windows connector 1.1 (with no patch).
I would update my CAM scipt, but we had problems originally with the CAM
script taking too look to execute and the end user would be bounced
between all servers in the FOG and never get passed the hour glass on
the DTU screen. So we trimmed our CAM script down and that made a huge
difference. In order to compensate for not having checks in the CAM
script, we created another script that runs every 15 mins via cron that
checks for a disconnected session and forces a utsessionkill on the
process. We then realized that it left orphaned uttsc-bin processes
that would eat CPU cycles and max out at .5%. So we then added the
second part of the script that searches for them and kills the orphaned
processes.
-----------------------------------------------------------------
#!/usr/bin/ksh
# /opt/SUNWut/lib/utsessionkill token $TOKEN
# /opt/SUNWut/sbin/utsession -k -t $TOKEN
/opt/SUNWut/sbin/utsession -p | grep D | while read a
do
TOKEN=`echo $a | awk '{print $1}'`
/opt/SUNWut/lib/utsessionkill token $TOKEN
done
sleep 10
/usr/bin/prstat -c 1 1 | grep uttsc-bin | grep 0.5% | while read a
do
PID=`echo $a | awk '{print $1}'`
/usr/bin/kill -9 $PID
done
------------------------------------------------------------------
Using these scripts has significantly lowered the CPU usage on all the
servers in the FOG which I routinely monitor via canary. I believe we
were getting a lot of disconnected session because the bulk of our DTU's
are coming in from a variety of locations that are behind cable modems
(VPN)that would routinely loose there connectivity for tiny amounts of
time jam up the sessions. I think adding these scripts has helped the
LDAP error, but there is always a window of 15min that the CPU could get
slammed and break the utauthd service.
Would you suggest adding a dedicated SRDS to our setup and if so what
would be the steps in doing so?
Thanks, Costa
-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Craig Bender
Sent: Friday, September 07, 2007 12:13 PM
To: SunRay-Users mailing list
Subject: Re: [SunRay-Users] SRSS 3.1 (could not bind to DS server)
LDAPerror...
What's the version of 120879, and what rev of the Sun Ray connector are
you using. Does prstat show java as the top consumer (PID will be for
authd). This sounds like a bug where uttsc tries to launch when the DTU
is disconnected. Pegs the CPU anything authd related gets horked.
The latest SR patch and latest verison of 1.1 for Sun Ray connector (or
just go to 2) has this fixed. Or you could simply change your CAM
script (if this is the issue) to only launch uttsc if the DTU is
connected. Like so:
http://blogs.sun.com/ThinkThin/entry/determining_wether_your_session_is
Constantine Morris wrote:
> I'm currently running 4 x T2000 servers in a hosted group (FOG ) with
> SRSS 3.1 configured and the windows connector running in CAM mode.
As
> the number of DTU's started to increase, we begain experiencing
issues
> with logging into the Sunray Administration console via them web
> (invalid login) and we were getting the following errors in the
> messages log:
>
>
>
>
>
> Sep 6 14:15:59 BPPMDT01 utauthd: [ID 702911 user.info]
> search_for_entries(): LDAP search failed - Can't contact LDAP server
>
> Sep 6 14:15:59 BPPMDT01 utauthd: [ID 702911 user.info]
ut_addDsEntry():
> DS operation failed - Can't contact LDAP server
>
> Sep 6 14:15:59 BPPMDT01 utauthd: [ID 702911 user.info]
desktop_event():
> Could not create desktop in LDAP (Internal system error)
>
> Sep 6 14:16:00 BPPMDT01 utauthd: [ID 702911 user.info]
> search_for_entries(): LDAP search failed - Can't contact LDAP server
>
> Sep 6 14:16:00 BPPMDT01 last message repeated 1 time
>
> Sep 6 14:16:00 BPPMDT01 utauthd: [ID 702911 user.info]
ut_addDsEntry():
> DS operation failed - Can't contact LDAP server
>
> Sep 6 14:16:00 BPPMDT01 utauthd: [ID 702911 user.info]
desktop_event():
> Could not create desktop in LDAP (Internal system error)
>
>
>
> It only seems to happen over a period of time for each of the servers
in
> the FOG. Currently, 3 out of the 4 are experiencing the same error.
> When I restart the sunray services on a particular server all is well
> again for several days and then it happens again.
>
>
>
> Thanks in advance...
>
>
>
>
>
>
>
------------------------------------------------------------------------
>
> _______________________________________________
> SunRay-Users mailing list
> SunRay-Users@filibeto.org
> http://www.filibeto.org/mailman/listinfo/sunray-users
_______________________________________________
SunRay-Users mailing list
SunRay-Users@filibeto.org
http://www.filibeto.org/mailman/listinfo/sunray-users
_______________________________________________
SunRay-Users mailing list
SunRay-Users@filibeto.org
http://www.filibeto.org/mailman/listinfo/sunray-users
