weber@brand.usc.edu (Allan G. Weber) (08/24/87)
I have a MicroVAX-II running Ultrix 1.2A that seems to die when the ethernet traffic gets very heavy. Whenever the ethernet gets jammed up with traffic because of other nodes sending out broadcasts, the probability of the machine crashing goes way up. Our SUN systems just print out a message about the ethernet being jammed but the MicroVAX often crashes when this happens. All the core dumps I've looked at seem to point to some problem in the DEQNA driver. Has anybody else seen this happen? If so, does DEC know about it and is it fixed in Ultrix 2.0? Is there a fix for it that can be applied to Ultrix 1.2A? DEC field service replaced the DEQNA board with no improvement so I don't think its a hardware problem with the board. However, we do have two device drivers of our own in the system running lab devices so I can't rule out it being due to our software. Any help will be appreciated. Thanks. Allan Weber USC Signal and Images Processing Institute Arpa: weber%brand.usc.edu@oberon.usc.edu UUCP: ...sdcrdcf!oberon!brand!weber
pdb@sei.cmu.edu (Patrick Barron) (08/24/87)
In article <4195@oberon.USC.EDU> weber@brand.usc.edu (Allan G. Weber) writes:
#I have a MicroVAX-II running Ultrix 1.2A that seems to die when the
#ethernet traffic gets very heavy. Whenever the ethernet gets jammed
#up with traffic because of other nodes sending out broadcasts, the
#probability of the machine crashing goes way up. Our SUN systems just
#print out a message about the ethernet being jammed but the MicroVAX
#often crashes when this happens. All the core dumps I've looked at
#seem to point to some problem in the DEQNA driver. Has anybody else
#seen this happen? If so, does DEC know about it and is it fixed in
#Ultrix 2.0? Is there a fix for it that can be applied to Ultrix 1.2A?
#DEC field service replaced the DEQNA board with no improvement so I
#don't think its a hardware problem with the board. However, we do
#have two device drivers of our own in the system running lab devices
#so I can't rule out it being due to our software. Any help will be
#appreciated. Thanks.
We had this problem too, with machines panicing with (I think) segmentation
violation in qeput(). DEC has a patch for Ultrix 1.2, which you were
supposed to get if you got the DEQNA FCO to rev. E (if you didn't get the
patch - you're not alone. We didn't get it with the FCO either, even though
Software Support told me we were supposed to). The patch fixes the crash
problem, and also fixes the "qerestart" problem (only if you have a rev. E
or later DEQNA). The distribution channel for the patch is (or was, when I
got it) your field service office.
--Pat.sullivan@marge.math.binghamton.edu (fred sullivan) (08/25/87)
>We had this problem too, with machines panicing with (I think) segmentation >violation in qeput(). DEC has a patch for Ultrix 1.2, which you were >supposed to get if you got the DEQNA FCO to rev. E (if you didn't get the >patch - you're not alone. We didn't get it with the FCO either, even though >Software Support told me we were supposed to). The patch fixes the crash >problem, and also fixes the "qerestart" problem (only if you have a rev. E >or later DEQNA). The distribution channel for the patch is (or was, when I >got it) your field service office. > Is this fixed in Ultrix 2.0, which I am about to install, or should I ask for the patch? Fred Sullivan Department of Mathematical Sciences State University of New York at Binghamton Binghamton, New York 13903 Email: sullivan@marge.math.binghamton.edu
pdb@sei.cmu.edu (Patrick Barron) (08/25/87)
In article <659@bingvaxu.cc.binghamton.edu> sullivan@marge.math.binghamton.edu (fred sullivan) writes: >> [in reference to my description of the Ultrix 1.2 DEQNA driver patch] >Is this fixed in Ultrix 2.0, which I am about to install, or should I ask >for the patch? This patch (or something like it) was incorporated into Ultrix 2.0. We have a few 2.0 machines, and none of them have had any qeput() crashes, and they don't seem to have "qerestart" troubles, either. --Pat.
jch@omnigate.clarkson.edu (Jeffrey C Honig) (08/31/87)
>In article <659@bingvaxu.cc.binghamton.edu> sullivan@marge.math.binghamton.edu (fred sullivan) writes: >>> [in reference to my description of the Ultrix 1.2 DEQNA driver patch] >>Is this fixed in Ultrix 2.0, which I am about to install, or should I ask >>for the patch? > >This patch (or something like it) was incorporated into Ultrix 2.0. We >have a few 2.0 machines, and none of them have had any qeput() crashes, >and they don't seem to have "qerestart" troubles, either. I've been having "qerestart" problems with BSD 4.3 on a MicroVAX. Has anyone incorporated this fix into the BSD 4.3 driver? Thanks Jeff