Hello everyone,

I am still trying to run the arXiv-1111.3344 TOV collapse simulation and having continued issues with updating the thornlist, specifically CoreCollapseControl.

Specific questions:

1. After updating my thornlist with CoreCollapseControl, my Cactus rebuild fails while compiling CoreCollapseControl, citing a few undeclared variables warnings before a final "Error 2." Am I missing a step in the update process?

2. Per Roland's previous email, I attempted to find CartesianCoordinates in the Llama repository but it does not exist. Has it been absorbed/replaced by the Coordinates thorn? And, if so, should I simply replace "CartesianCoordinates" with "Coordinates" in the par file?

3. With the known issues with the TOV collapse simulation, are there any recommendations for another collapse simulation? I am just looking for a working code shell and am not particularly tied to TOV.

Please see below for the steps I took to update my thornlist and build a new experimental Cactus configuration. I have also attached a log file with the output from my build.

1. For CoreCollapseControl:

a. I made a new repository called corecollapse and checked out the thorn into Cactus/repos/corecollapse using:

svn checkout https://svn.einsteintoolkit.org/incoming/CoreCollapseControl/

b. I created Cactus/arrangements/corecollapse then inside of it created a soft link to Cactus/repos/corecollapse/CoreCollapseControl

c. I added the following lines to my thornlist:

!TARGET = $ARR

!TYPE = git

!URL = https://bitbucket.org/einsteintoolkit/tickets/issues/1358/import-corecollapsecontrol-into-et

!REPOPATH = $2

!Checkout =

corecollapse/CoreCollapseControl

e. Built a second experimental configuration called sim2 from within Cactus directory with:

sim build sim2 --thornlist ./thornlists/einsteintoolkit.th

Once again, thank you ahead of time for any help!

Kindest Regards,

Justin Verde

On Fri, Jan 24, 2020 at 2:17 PM Haas, Roland <rhaas@illinois.edu> wrote:

Hello Justin,

those were some thorns that were private at the time.

CoreCollapseControl has been proposed for inclusion in this ticket:

https://bitbucket.org/einsteintoolkit/tickets/issues/1358/import-corecollapsecontrol-into-et

and you can obtain a copy using svn:

svn checkout
https://svn.einsteintoolkit.org/incoming/CoreCollapseControl/

CartesianCoordinates is part of Llama and in the llama repository that
has since become part of the Toolkit.

So you can use it by adding a line like this:

Llama/CartesianCoordinates

to your ThornList. You may have to create the appropriate symbolic link
first via (something like):

cd arrangements/Llama
ln -s ../../repos/llama/CartesianCoordinates CartesianCoordinates

TriggerTerminationManual has been superseeded my the TerminationTrigger
thorn (in CactusUtils) in the toolkit, though you can also just comment
it out since it does not affect physics.

Even with all those in place, the TOV collapse is "temperamental" and
sometimes does not work well.

Please see here:

http://lists.einsteintoolkit.org/pipermail/users/2019-March/006802.html

for someone else's experience.

Basically Tony had to increase resolution by a factor of 2 (to 1.6
from 3.2) to avoid failures.

Yours,
Roland

> Hi everyone,
>
> I'm a new user to the toolkit and have been trying to run the TOV collapse
> example from arXiv-1111.3344 on my home system with 4 cores. I keep getting
> the following errors before the job crashes:
>
> ===============================================
> Error: Thorn CartesianCoordinates not found
> Error: Thorn corecollapsecontrol not found
> Error: Thorn TriggerTerminationManual not found
> Activation failed - 3 errors in activation sequence
> WARNING level 0 from host euler process 0 in thorn Cactus, file
> /home/verde029/ETK/Cactus/configs/sim/build/Cactus/main/SetParams.c:93:
> -> CCTKi_SetParameter: Error at line 66 in parameter file
> /home/verde029/simulations/collapse_crash_test/output-0000/tov_collapse_vc_l3.par
> while activating thorns
> --------------------------------------------------------------------------
> MPI_ABORT was invoked on rank 0 in communicator MPI_COMM_WORLD
> with errorcode 1.
>
> NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
> You may or may not see output from other processes, depending on
> exactly when Open MPI kills them.
>
> ============================================
>
> I can't find anything on these thorns through the thorn documentation
> guide. Have the thorns or par file been updated, or is there a quick
> workaround to update the file?
>
> I appreciate any help!
>

--
My email is as private as my paper mail. I therefore support encrypting
and signing email messages. Get my PGP key from http://pgp.mit.edu .

Kindest Regards,

Justin Verde