| Title: | DEChub/HUBwatch/PROBEwatch CONFERENCE |
| Notice: | Firmware -2, Doc -3, Power -4, HW kits -5, firm load -6&7 |
| Moderator: | NETCAD::COLELLA DT |
| Created: | Wed Nov 13 1991 |
| Last Modified: | Fri Jun 06 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 4455 |
| Total number of notes: | 16761 |
I have a Customer with 10 DEChub 90s. Most Hubs have an agent installed
in the backplane. Total of 8 Agents. The network has agents both on
the local LAN and over a bridged WAN.
The problem is that intermittently the Agents appear to go to sleep.
That is they can't be pinged and Hubwatch reports no response from
agent. According to the customer all agents have had this problem and
there is no detectable pattern.
According to the customer the agents 'look normal' but the only way
that they can get them to work again is to pull them out of the hub and
re-install them. Then things work fine until at some random point the
agent stops responding.... (Today they customer pinged 3 hubs ok then
the 4th hub failed). He is espically concerned about the remote sites
where he has no on site technical resources.
Hardware Rev. d01 ro bl05.00
Current Firmware 2.1.3
This has been a problem since day 1 and firmware upgrades have not
changed the symptom. customer has Hubwatch 3.1 and new firmware kit on
site. He has no confidence that this will 'resolve his problem'.
Any Thoughts?
Thanks
Jim
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 2824.1 | Forgot one point | CHOWDA::FAHEY | Are we having 'FUN' yet? | Mon Oct 02 1995 19:49 | 6 |
I forgot one thing. It's not the Hubwatch workstation. The agent can't
be pinged from a VMS system running UCX. The VMS system can ping the
HUBwatch workstation and vice versa.
Thanks
Jim
| |||||
| 2824.2 | Not a known problem | NETCAD::WALTER | Tue Oct 03 1995 16:42 | 30 | |
Jim,
The symptoms you describe don't match any known problems I'm aware of,
at least with that revision of firmware. V2.1.3 has been in the field
for about 1.5 years with no similar reports. There are hardware
problems that can cause such behavior, but typically when the hardware
fails, it stays dead, regardless of resets. There was a firmware bug
in V3.0 code which caused a similar symptom, but you aren't running
that revision (by the way, that bug was fixed in V3.0.1, which is the
latest version).
I recommend that you get as much information from the customer as you
can regarding his configuration: the hardware and firmware rev of the
Agents, what other modules are in the hubs, is there a bridge between
the Agent and the management system. Also more detail on the symptoms:
how often does it happen, do all 8 Agents exhibit the problem, can
anything else in the hub be pinged when the Agent is asleep, does
resetting the Agent always fix the problem immediately. Finally,
the Agent maintains an errorlog of the last 8 events. That errorlog
can be dumped to the console screen, or it can be read via the SNMP
objects in the Dechub90 mib, by doing a getNext on the da90ErrorlogTable.
The information in the errorlog can be instrumental in discovering the
source of the problem.
You can forward this information to me at NETCAD::WALTER, or call me
at DTN 226-6799.
Regards,
Dave
| |||||