On Wed, 9 Sep 2020 at 13:33, Ranjan Maitra <maitra@xxxxxxxxx> wrote:
Thanks, I will file the bug report: should it be upstream or on Fedora? I know about Fedora's BZ, but what is upstream?
Fedora/RedHat bugzilla. If the automated bug reporting system was working the report would be seen by Fedora
people. Since the failure is on Fedora and not other platforms, linux expertise may be needed to understand the issue.
I have VM's for other linux distros so I plan to check to see if the issue occurs on other linux distros.
Ranjan
On Wed, 9 Sep 2020 13:15:57 -0300 "George N. White III" <gnwiii@xxxxxxxxx> wrote:
> On Wed, 9 Sep 2020 at 10:57, Ranjan Maitra <maitra@xxxxxxxxx> wrote:
>
> > On Wed, 9 Sep 2020 08:08:23 -0300 "George N. White III" <gnwiii@xxxxxxxxx>
> > wrote:
> >
> > > On Tue, 8 Sep 2020 at 18:34, Ranjan Maitra <maitra@xxxxxxxxx> wrote:
> > >
> > > > Dear friends,
> > > >
> > > > I have a strange situation that shows up only on F32 (only linux
> > > > distributions tried) but does not show up on Windoze or Muck. The way
> > to
> > > > illustrate this is rather complicated, so I will first summarize.
> > > >
> > > > For my introductory students, I use a java-based front-end to R called
> > > > Deducer. I am able to install without a hitch and even bring it up and
> > run
> > > > some functions and graphical displays. However, when I ask it to do a
> > few
> > > > specific functions, the windows all come crashing down. So, my
> > question is:
> > > > how do I figure out the source of the problem? I have used this
> > program for
> > > > years, and including last spring under F31, and linux has never been an
> > > > issue. This has been tried on three F32 machines by me and a former
> > student
> > > > and all have resulted in crashes. Embarrasingly, these continue to
> > work as
> > > > expected with Windows and MacOS, though installation, especially for
> > the
> > > > latter is a pain.
> > > >
> > >
> > > It would be good to mention enough detail so others can reproduce your
> > > configuration:
> > >
> > > graphics hardware and driver: did the 3 machines use the same graphics
> > > hardware and drivers?
> >
> > Hi,
> >
> > Thanks! The 3 machines are very different (all laptops, though) so likely
> > have different graphics HW and drivers.
> >
> > > graphics display server: wayland +xwayland or X11
> >
> > Sorry, how do I figure this out? I think one of the machines uses gnome so
> > whatever comes with that (likely wayland +xwayland?) and mine I think uses
> > X11 (these use openbox).
> >
> >
> I tried both wayland and weston in Fedroa 32 as a VM. It is the JRE that
> crashes. My system generated a problem report (which fails to upload -- I
> think the server is down) for java-1.8.0-openjdk-headless, which is odd
> considering the application uses graphics.
>
>
> > > JRE -- some Java apps require Oracle's proprietary JRE
> >
> > I am using the default Fedora version, and also tried with fedora's
> > java-latest. Importantly, upto F31, this has never been an issue.
> >
> > >
> > > > How can I get error messages out of this mishap to file a meaningful
> > and
> > > > targeted bug report?
> > > >
> > >
> > > https://www.linuxjournal.com/content/oops-debugging-kernel-panics-0
> > >
> > >
> > > >
> > > > Any suggestions?
> > > > Ranjan
> > > >
> > > >
> > > > If you are interested, here is how I go about installing Deducer. I
> > need
> > > > R-core installed, as well as curl-devel, and of course java. Other
> > fedora
> > > > packages may also be needed.
> > > >
> > > > sudo dnf install R-devel curl-devel java-1.8.0-openjdk-devel
> > > >
> > > > then:
> > > >
> > > > sudo R
> > > > install.packages(c('JGR','Deducer'))
> > > > q()
> > > >
> > > > Note that the second line will ask for which repo to use, and any are
> > > > fine, the first one is from rstudio.com. Also, the installation
> > function
> > > > may ask for more libraries to be installed.
> > > >
> > > > Once installed, go to R and inside R, type:
> > > >
> > > > library(JGR)
> > > > library(Deducer)
> > > > JGR()
> > > >
> > > > The last function call will open a graphical interface with a jaguar
> > logo
> > > > (pretty cool). All is well. In the bottom window type:
> > > > data(cars)
> > > >
> > > > Then go to the graphical interface above, and under "Plots" -> "Quick"
> > ->
> > > > "simple boxplot"
> > > > This opens a window, and put speed as x variable and run.
> > > > A window begins to open and the entire window system running JGR comes
> > > > crashing down.
> > >
> > >
> > > > Btw, since I know the commands to give in R, on the commandline, when I
> > > > call the function, things work as expected. It is only on the JGR()
> > window
> > > > that it crashes.
> > > >
> > >
> > > Have you checked rJava and JGR bug reports?
> >
> > I looked around but was unable to find this specific crash detailed
> > anywhere.
> >
> >
> > > >
> > > > The worst part? I get no error messages. So, how can I get error
> > messages
> > > > out of this mishap to file a meaningful and targeted bug report?
> > > >
> > > > Can you connect to the system with ssh or use a console <Alt Fn>? If so
> > > the kernel is still alive and the graphics system has crashed. If not,
> > > you can first try adjusting your configuration. The last resort is
> > kernel
> > > debugging.
> >
> > Yes, the system and the graphics itself is up and running. It is only the
> > JGR window that crashes (and brings down that part of the program). What
> > surprises me is that only certain parts of the program crash. For instance,
> > the histograph and density plots or scatterplot (both also using ggplot2
> > software) get displayed fine.
> >
> >
> >
> > So, your message set me thinking of looking at
> >
> > tail -f /var/log/messages
> >
> > while there is a successful operation, such as using the histogram
> > function, and an operation (boxplot) which crashes. Here is the output for
> > when it succeeds (displays a histogram):
> >
> >
> > [...]
> >
> >
> >
> > Here is the output for when it crashes (displays a boxplot):
> >
> >
> > Sep 9 08:51:12 localhost audit[53948]: ANOM_ABEND auid=1000 uid=1000
> > gid=1000 ses=2 subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023
> > pid=53948 comm="Thread-5"
> > exe="/usr/lib/jvm/java-1.8.0-openjdk-1.8.0.265.b01-1.fc32.x86_64/jre/bin/java"
> > sig=11 res=1
> > >
> > > This suggests a graphics issue. If you have been using wayland you should
> > > try the X11 server. if both graphics systems are failing you could try
> > an
> > > Oracle JRE
> > >
> > https://www.oracle.com/java/technologies/javase/javase8-archive-downloads.html
> >
> > I could do that, I guess, but does the above help?
> >
>
> I can reproduce the problem. It merits a bug report against openJDK (best
> if you report it as
> I may be otherwise busy and not able to keep up with requests for more info
> or testing
> potential fixes).
>
> --
> George N. White III
--
Important Notice: This mailbox is ignored: e-mails are set to be deleted on receipt. Please respond to the mailing list if appropriate. For those needing to send personal or professional e-mail, please use appropriate addresses.
_______________________________________________
users mailing list -- users@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe send an email to users-leave@xxxxxxxxxxxxxxxxxxxxxxx
Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: https://lists.fedoraproject.org/archives/list/users@xxxxxxxxxxxxxxxxxxxxxxx
George N. White III
_______________________________________________ users mailing list -- users@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to users-leave@xxxxxxxxxxxxxxxxxxxxxxx Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@xxxxxxxxxxxxxxxxxxxxxxx
