| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 522.1 | same problem | NETRIX::"robert.schmoelzer@grt.mts.dec.com" | NSIS Austria | Thu Mar 06 1997 04:37 | 12 |
| Hi Keith,
Some other guys (and i) have posted this in 333.* and 334.* but there is
still no answer to this problem.
Seems that there is no more interest in these product from engineering,
support, marketing, whoever. Another good example how we jeopardize our
customer base.
Hopefully someone will tell us that i am wrong.
a little frustrated
Robert
[Posted by WWW Notes gateway]
|
| 522.2 | | TUXEDO::BAKER | | Thu Mar 06 1997 18:37 | 14 |
| When 333 and 334 were posted there was a response that "on a
busy system it may take awhile for it to happen" and the person
making that reply did some testing locally. No problem was
found locally and nobody responded in the replys that "waiting
awhile" didn't solve the problem, hence the problem was closed
locally.
The problem has been reopened locally. If anyone can give
more detail of how the nodes were unmanaged, i.e, individually
or by a higher level submap, it would be appreciated. Any other
detail would also be useful. We have unable to reproduce the
problem here.
Bob...
|
| 522.3 | more info | NETRIX::"robert.schmoelzer@grt.mts.dec.com" | NSIS Austria | Fri Mar 07 1997 11:46 | 29 |
| Hi Bob,
I can reproduce this problem on several machines, internal and customer ones.
Unfortunately i dont have a PNV machine now for testing purposes.
The problem as described occurs if you "unmanage" a Subnet with many
nodes(>100)
or if you unmanage a bunch of "selected" nodes. It does NOT happen if unmanage
ONE node after the other(which of course is not very practicabel in the real
world.
So, IMHO it has something to do with daemons losing events if many of them
occur at the same time.
Besides this, there is another problem with the daemons which maybe has some
impact on this. After some days(5-6-7...) some daemon seems to die, since no
more events are beiing displayed. You can look at the daemon status, but
everything looks fine. If you shut down and restart PNV AND the Daemons
it works fine, at least some days.
Arent there any patches or bugfixes or is it just the fact that we dont
recieve
information about it from Tivoli?
What about the long awaited "NT 4.0 compatible" version??????????
some update would be nice to see what we can expect and what we cannot
rgds
Robert
[Posted by WWW Notes gateway]
|
| 522.4 | Next Release | WOTVAX::dhcp20.olo.dec.com::hunt_js | | Tue Mar 11 1997 14:55 | 23 |
| FYI
I talked to Jeffrey Snover of Tivoli (previously of Digital?) last week and
he also demonstrated the beta release of Netview-NT on WNT 4.0. I guess if its
in beta it can't be far from a real release (next couple of months??)
Tivoli are being very bullish about the product. They say they got an
excellent product from Digital (sigh!) and now made it even better. It appears
they have put a lot of effort into the next release.
My understanding is that they have removed the dependency on the nutcracker
kernel thus improving performance.
They have also completely restructured the way events are handled, such that
it can now handle a significant number of events a second (don't ask me how
many - Jeffrey did say, but I forgot!). When I mentioned the startup problem
of Netview polling all nodes (even unmanaged ones) during startup, I think he
said that that hadn't changed - though I wouldn't swear to that!
It also supports client/server, enabling it to be driven from a system other
than where the databases reside.
Cheers, Jim
|
| 522.5 | | TUXEDO::MINTZ | Erik Mintz | Thu Mar 13 1997 14:44 | 9 |
| RE: .-1
Yes, it is the same Jeffrey Snover who was here at Digital.
There is information on the upcoming release available in the NetView
PID (available, I believe, through the IR). A field test version
will be available internally shortly.
-- Erik
|
| 522.6 | Polling continues when "unmanaged" from upper level maps | TUXEDO::BAKER | | Fri Mar 21 1997 17:03 | 12 |
| The problem of polling continuing after the nodes have been
marked "unmanaged" has been confirmed.
This situation occurs when the nodes are unmanaged from a higher
level map. When unmanaged from maps showing the individual nodes,
polling to those nodes stops as it is suppose to.
There are no plans to fix this in v3.1, it will be fixed in v5.0.
V5.0 is the NT 4.0 compatable version.
Bob...
|