Board index Solaris

101318-64 for MP Suns under Solaris 2.3

101318-64 for MP Suns under Solaris 2.3

Post by Ian G Batt » Fri, 18 Nov 1994 23:18:11



Just in case there's any doubt, 101318-64 appears to solve the problems
with MP suns.

$ ntpq -p zippy (10/512, Generic_101318-59):
     remote           refid      st when poll reach   delay  offset    disp
===========================================================================
#kevin-the-files fulcrum.fulcrum  3   40   64   37     3.65  -56.91  899.87

$ ntpq -p doctor-seuss (690/512, Generic_101318-59)
+kevin-the-files fulcrum.fulcrum  3    9   64  137     4.50 -1282.3  299.53
*fulcrum.fulcrum woodcock.cl.cam  2   50   64  377     2.58 -1096.7  342.22
+kether.fulcrum. fulcrum.fulcrum  3   20   64  377    -8.68 -693.75  442.23

$ ntpq -p bungle (10/512, Generic_101318-64)
*kevin-the-files fulcrum.fulcrum  3   28   64  377     4.46    0.17    1.51

$ ntpq -p biff (10/512, Generic_101318-64)
*kevin-the-files fulcrum.fulcrum  3   57  128  377     3.43    0.05    2.49

$ ntpq -p george (10/514, Generic_101318-64)
*kevin-the-files fulcrum.fulcrum  3    4  128  377     4.46    0.31    0.53

 
 
 
Top

1. ypbind fixed on Solaris 2.3 by 101318-64

We have had a problem since we first installed Solaris 2.2 that has
continued through Solaris 2.3 that when the system is busy mail
will bounce because users names can not be looked up because
the connection to ypbind fails.  We have worked for months with
sun engineers on this to get it fixed.  We had lost all hope of
this ever getting fixed.  Now, out of the blue, the patch has
come to fix it.  It's a miricle!!!!!! 101318-64 is the best thing
to happen to yp since the UDP reliability patch.  I just put it in
yesterday and the test I used to generate the trusses I sent sun
that showed it was failing because it was not connecting work fine
now.  I went back I found the letter I sent them on the 20th of Apr
that showed what was wrong with ypbind.  Of cource since then they
closed my service order saying it could not be fixed.  I really
wonder who got them to fix it.

With this fix done I think today is the first day I can say
Solaris 2.3 is better then 4.1.3.  Yea!! no more bounced mail!!!

I notice from the truss of the groups command, the command I used
to prove it fails, all opens of /dev/ticotsord are gone.  I wonder
if this means that /dev/ticotsord continues to be broke in some
way?

Here is what Sun says abount the fix in the patch description:

Under a heavy system load NIS clients could fail to contact the ypbind
process and would then fail.  This would cause a variety of symptoms
that can be traced to failed calls on getpwnam(), getpwuid(),
gethostbyname(), etc.  The ypbind process has been changed to cache the
ypserv transport address in a file.  Now NIS clients can get the
address of ypserv without connecting to ypbind.  In addition, NIS
clients will now keep trying to get results as long as ypbind is
running.

--
Chad Adams              __               Educational Computing Network
Systems Programmer     /  ) /         /  University Park, IL  USA
1-708-235-2200        /    /_  __, __/   ECN, networking in IL since 1975

2. passwd expiration

3. Patch 101318-64

4. How to get the arg immediately afterwards?

5. 101318-64 warning

6. Linux Universe ?

7. Just another problem with 101318-64.....

8. network question

9. more 101318-64 conflicting views

10. Kernel Jumbo Patch 101318-64?

11. Last log problem after 101318-64 patch

12. 101318-64 problem

13. need input on kernel jumbo patch for Solaris 2.3 (101318-70)


All times are UTC
Board index