[SunRay-Users] utauthd crashes?

Nishimura, Scott (IT Solutions) scott.nishimura at ngc.com
Thu Apr 10 21:22:03 EEST 2008




[SunRay-Users] utauthd crashes?
Hana Skoumalova Hana.Skoumalova at ff.cuni.cz
<mailto:sunray-users%40filibeto.org?Subject=%5BSunRay-Users%5D%20utauthd
%20crashes%3F&In-Reply-To=a4209c300803181120n36e647a7p1532dd97f8b16132%4
0mail.gmail.com> 
Wed Mar 19 09:08:25 EET 2008 
We had a similar problem:  utauthd would eventually go off into La La
land and not respond.  Sun finally traced the problem to an active but
unconfigured network port.  No, no one ever explained the logic on this
one.

Here are my notes from that occasion, scattered as they are:
the following was performed via 'rsunsu01' upon determining that its
"netstat -g" output apparently lacked an 'e1000g0   224.101.101.101   1'
string:

   - "snoop 224.101.101.101"
	[Note- yielded <NULL> -- confirming SRSS non-viability...
	[Note- same also invoked upon 'rsunsu02' / 'rsunsu03' --
indicating NO 'rsunsu01' interface activity...]

   - "/opt/SUNWut/srds/lib/utldbmcat
/var/opt/SUNWut/srds/dbm.ut/id2entry.dbb > /var/tmp/sr_store"
	[Note- same also performed on 'rsunsu02' / 'rsunsu03' --
"just-in-case"...]
	 {...recovery protocol contained in SRSS 3.1 Administrator's
Guide, dated September 2005 -- pages 206-208}

   - "reboot"
	[Note- SRSS viability restored...]

   - "snoop 224.101.101.101"
	[Note- yielded "rsunsu01 -> 224.101.101.101 IGMP v2 membership
report" -- indicating SRSS viability...]
	[Note- same also invoked upon 'rsunsu02' / 'rsunsu03' yielded  
		"rsunsu01 -> 224.101.101.101 UDP D=7009 S=7009 LEN=260"
-- further indicating SRSS viability...]

Scott

ottomeister wrote:
> On Tue, Mar 18, 2008 at 2:25 AM, Hana Skoumalova
> <Hana.Skoumalova at ff.cuni.cz
<http://www.filibeto.org/mailman/listinfo/sunray-users> > wrote:
>>  I suspect that the files 12, 13 and 16 disappeared from
>>  /tmp/SUNWut/config/displays/, because gdm complains about it:
>>
>>  $ cat /var/log/gdm/:12.log
>>
>>  Fatal server error:
>>
>> SunRay Session map does not exist (display 12)!
> 
> Does this machine have a cron job or some other mechanism
> that deletes "old" files from /tmp ?  If so, tell it to not do that in
> /tmp/SUNWut.  Under some circumstances (for instance, when
> the X session is not logged in) SRSS plays with the timestamps
> on these files in order to make them look less recent, therefore
> less attractive for hotdesking purposes.

Yes, we run tmpwatch daily. Thank you for the hint.

Hana

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.filibeto.org/pipermail/sunray-users/attachments/20080410/a3fc92fa/attachment.html


More information about the SunRay-Users mailing list