| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 2431.1 | The error log may have some useful info | STAR::BMATTHEWS | | Mon Mar 12 1990 14:11 | 1 |
| reply to this note with the contents of sys$manager:decw$server*.log
|
| 2431.2 | | STAR::MCLEMAN | Jeff McLeman, VMS Development | Mon Mar 12 1990 15:23 | 2 |
| The customer did put the proper cable from the system box
to the monitor, correct?
|
| 2431.3 | SPX SYSGEN PARAMETERS | ZSAZSA::READINGS | Richard Readings | Tue Mar 13 1990 10:03 | 19 |
| When we installed an SPX graphics option we experienced a stack dump whenever
GEBDRIVER was reloaded by DECW$DEVICE_GE.COM (called by DECW$STARTUP.COM).
We solved the problem by defining the following system parameters in
MODPARAMS.DAT, then running AUTOGEN:
MIN_SRPCOUNT=2100
MIN_IRPCOUNT=1280
MIN_RESHASHTBL=850
MIN_LOCKIDTBL=1800
MIN_GBLPAGES=15000
These were determined by running VPA while attempting to start DECwindows.
The symptoms were different from those described in .0 but it might be worth
checking the SYSGEN parameters.
Richard
|
| 2431.4 | SPX SYSGEN parameters for 8 Mbyte VS3100 | ZSAZSA::READINGS | Richard Readings | Tue Mar 13 1990 10:18 | 4 |
| The system parameters in .3 relate to an 8 Mbyte VS3100. Upgrading to 16Mbyte
solved the problem without modifying the system parameters.
Richard
|
| 2431.5 | | DECWIN::FISHER | Prune Juice: A Warrior's Drink! | Tue Mar 13 1990 12:31 | 8 |
| re .0: Is this still open? Or was .-1 responding for the author of .0? If
it is still open, what device are you seeing? For that matter, what device
did you see BEFORE the upgrade. (I.e. did you upgrade from the default
1-bit frame buffer system to the spx or from older GPX-style 8-bit device
to the SPX? The device of interest at this point is GE. Before, it would
have been GA for a GPX-type and GC for a 1-bit monochrome (MFB).
Burns
|
| 2431.6 | latest news | SMOGGY::JORGENSEN | | Tue Mar 13 1990 16:56 | 26 |
| Hi there.
I am the field engineer that did the upgrade.
Original config. was monochrome with no graphics board. The system
is a satellite node with a MVII as a boot node. I am not sure how
much memory is there, but it has a foriegn memory piggybacked on
a dec expansion memory.
The device the system sees is GEA0. When booting the message GEBDRIVER
BEING RELOADED is displayed. No errors are displayed. The monitor
(VRT19) works in text mode and I am able to log in. GEA0 shows on-line.
Latest news: The customer had a delqua to replace his deqna on the
boot node. He did so over the weekend and rebooted the boot node.
He didn't notice right away, but eventually he checked the 3100
in question and decwindows was up and running.
All is not well yet though. When he logged out of a window the 3100
crashed. After it rebooted windows came right up. The customer logged
in and out several times with no problem. This was as of 12-mar.
Any input would be greatly appreciated
--thx, Paul Jorgensen
|
| 2431.7 | Autogen since the upgrade? | DECWIN::FISHER | Prune Juice: A Warrior's Drink! | Tue Mar 13 1990 17:27 | 8 |
| Has anyone tried to analyze the crash dump (you do mean a SYSTEM crash, and
not just DECwindows not starting again?)
Has anyone done an AUTOGEN since the upgrade, BTW? It would probably be
especially good after running for a few days in SPX mode since it would
get better feedback.
Burns
|
| 2431.8 | latest latest news | SMOGGY::JORGENSEN | | Tue Mar 13 1990 18:04 | 11 |
| I just talked to the customer and found out it wasn't a system crash,
It just went back to the original problem w/decwindows not working.
He rebooted the 3100 and everything came up fine.
We ran autogen when trying to fix the original problem, but not
since it started working. I suggested to the customer that he run
autogen again after a couple of days.
It hasn't had any problems this week since the initial one.
thx for the response--P.J.
|