We have been having a problem for months with system crashes.

Users will be working along with no problems and then all of a sudden
some one will get the message:

Remote Operation Failed


O=Logoff / Q=Quit / R=Retry / C=Continue ? r
*** Remote File System Error ***

This problem will then cascade quickly through the system often causing
the system to crash without the benefit of a shutdown.

We reboot and will work fine for a 1-3 days and it happens again.

Here is our version information:

System Release Information
==========================
D3 Release Version 7.4.6.NT
Implementation. . . . . . NT
Software Serial Number. . 11031251
System ID Number . . . .
Release . . . . . . . . . D3
Windows Information . . . WindowsNT;pick0;5.2.3790;;5.2

Boot Monitor. . . . . . . 7.4.6.0390; 12/23/05
Boot ABS. . . . . . . . . 7.4.6.A15; 05 Dec 2005
Boot ABS Data File. . . . 7.4.6.A15; 05 Dec 2005
System Data Files . . . . 7.4.6.D15; 05 Dec 2005
FlashBASIC Revision . . . 7.4.5.F13; 28 Apr 2005
SQL Revision. . . . . . . 7.4.5.S8; 28 Apr 2005

Current ABS . . . . . . . 7.4.6.A15; 05 Dec 2005
Current ABS Data File . . ;
ABS Patch Level . . . . . no patches loaded

Any help on this would be appreciated.

Thanks
Cliff Ponce
cponce (AT) easternmetal (DOT) com

"Cliff" wrote:
Holy cow Cliff - It's great seeing you again! Been a few years!

Your working on a relatively new release and certainly under a support
agreement. I don't think there's much we can do in the field to
diagnose this one. I recommend taking it through RD Support, or at
least posting a note to the RD forum and asking in general if there is
a known problem in 7.4.x which might cause this specific error.

Take care!
Tony
TG@ removethisNebula-RnD.com

Cliff,

As Tony suggested, you NEED to report this to RD. We had problems with
OSFI drivers (DOS on this release (actually impacts back to at least
7.3.x, and every earlier release - one more "plug" in the VME-creep
memory leaks) that was causing monitor halts & was repeatable (our
NASTY burn in program) - but it is being fixed (in the meantime, we
have modified to use c%function I/O routines)

Likewise for anyone on D3/Linux, we uncovered some issues with
FlashBasic with the latest F16/A210 combo this week (in the process of
reporting - dropped back to F14 to "fix" with A210)

Cliff,

As Tony suggested, you NEED to report this to RD. We had problems with
OSFI drivers (DOS on this release (actually impacts back to at least
7.3.x, and every earlier release - one more "plug" in the VME-creep
memory leaks) that was causing monitor halts & was repeatable (our
NASTY burn in program) - but it is being fixed (in the meantime, we
have modified to use c%function I/O routines)

Likewise for anyone on D3/Linux, we uncovered some issues with
FlashBasic with the latest F16/A210 combo this week (in the process of
reporting - dropped back to F14 to "fix" with A210)

We had these errors since day 1 on D3NT. RD support has been notified
many times but had us running in circles for a while and then we gave
up on them. The best thing you can do is to restart VME. And you should
restart it asap otherwise it will get worse and worse.

One more thing I forgot to mention. It seems to happen after select
statements on files with b-tree indexes.

Cliff,

The only time I noticed a similar problem was when I converted a system
from D3-AIX (some UNIX form) to D3-NT and forgot to issue the
UPDATE-MD command for every converted account. There was
enough subtle differences in the D3-AIX verb structure and the D3-NT
structure (probably some version changes also) that these a similar
error would crop up from time to time.

Larry Hazel

"Cliff" <cponce (AT) easternmetal (DOT) com> wrote

yes.. Remote File System Error is the "normal" and "known" issue from
the day 1 we used D3/NT. We had also reported this to RD but they need
us to provide a way for them to replicate the problem.. >.<


We find that this error may be related to the D3 Virtual Memory
mechanism and Flash Basic.

If you are using Win2k3 server, you can activate the /3G option during
bootup which it will help.

Rgds,

I am not sure which files have indexes created for them here but I do
know that they do exist. I will look to see which ones and see if the
processes using those file indexes are more likely to get the first
issue. Thanks for the idea.

Yeah, we try to get a proper shutdown as soon as we see these occuring
but some times it cascades so fast we can't get it shutdown before it
crashes on us.