#232: GetComponents has a zero return code even when all components were not
checked out
---------------------------+------------------------------------------------
Reporter: hinder | Owner: eric9
Type: defect | Status: new
Priority: major | Milestone:
Component: GetComponents | Version:
Keywords: |
---------------------------+------------------------------------------------
When I run GetComponents, I frequently run into what I can only assume are
transient network errors of the following form:
Checking out module: CactusNumerical/Cartoon2D
from repository:
http://svn.cactuscode.org/arrangements/CactusNumerical/Cartoon2D/trunk
into: ./arrangements
svn: REPORT of '/arrangements/CactusNumerical/Cartoon2D/!svn/vcc/default':
Could not read response body: connection was closed by server.
(http://svn.cactuscode.org)
It used to be the case that GetComponents would return a nonzero exit code
in this case so that my automated script can detect this and try again.
Now GetComponents says:
163 components checked out.
0 components updated.
Unable to process CactusNumerical/Cartoon2D
Unable to process EinsteinInitialData/Exact
Unable to process EinsteinInitialData/TwoPunctures
Time Elapsed: 270 minutes, 51 seconds
and returns a zero exit code. This is preventing nightly build and test
from running.
I am using the version of GetComponents from
http://svn.cactuscode.org/Utilities/trunk/Scripts/GetComponents
--
Ticket URL: <https://trac.einsteintoolkit.org/ticket/232>
Einstein Toolkit <http://einsteintoolkit.org>
The Einstein Toolkit
#371: Repeat warnings at end
---------------------------+------------------------------------------------
Reporter: eschnett | Owner: eric9
Type: defect | Status: new
Priority: critical | Milestone:
Component: GetComponents | Version:
Keywords: |
---------------------------+------------------------------------------------
When there are conflicts or other problems, the warnings are output as
GetComponents proceeds. The warnings should be repeated at the end. At the
very least, the fact that there were warnings should be made clear.
Otherwise, conflicts or other problems remain undetected, and the next
--update operation will then lead to serious problems because the
repositories are in an unclean state.
--
Ticket URL: <https://trac.einsteintoolkit.org/ticket/371>
Einstein Toolkit <http://einsteintoolkit.org>
The Einstein Toolkit
#353: Remove use-nodes handling in simfactory
-------------------------+--------------------------------------------------
Reporter: eschnett | Owner:
Type: enhancement | Status: new
Priority: minor | Milestone:
Component: Other | Version:
Keywords: |
-------------------------+--------------------------------------------------
Simfactory currently tries to determine which nodes the queuing system
provides to a run. This list of nodes is not used anywhere, and the
mechanism is specific to (probably) two flavours of PBS.
This mechanism is not necessary, since this functionality is contained in
the individual run scripts. The run scripts also save the list of nodes
into a file for possible later use.
The enclosed patch removes this functionality, simplifying the code.
--
Ticket URL: <https://trac.einsteintoolkit.org/ticket/353>
Einstein Toolkit <http://einsteintoolkit.org>
The Einstein Toolkit
#355: Flush stdout and stderr
-------------------------+--------------------------------------------------
Reporter: eschnett | Owner: mthomas
Type: enhancement | Status: new
Priority: major | Milestone:
Component: SimFactory | Version:
Keywords: |
-------------------------+--------------------------------------------------
The attached patch flushes stdout and stderr when warnings are output.
This ensures that they are printed in the correct order.
The patch also adds some TODO statements where the current code could be
improved.
--
Ticket URL: <https://trac.einsteintoolkit.org/ticket/355>
Einstein Toolkit <http://einsteintoolkit.org>
The Einstein Toolkit
#352: Patch: Support presubmitting on several systems
-------------------------+--------------------------------------------------
Reporter: eschnett | Owner: mthomas
Type: enhancement | Status: new
Priority: major | Milestone:
Component: SimFactory | Version:
Keywords: |
-------------------------+--------------------------------------------------
The enclosed patch enables presubmission in the MDB entries on several
systems.
--
Ticket URL: <https://trac.einsteintoolkit.org/ticket/352>
Einstein Toolkit <http://einsteintoolkit.org>
The Einstein Toolkit
#338: GetComponents on Cactus website is out of date
----------------------------+-----------------------------------------------
Reporter: hinder | Owner:
Type: defect | Status: new
Priority: major | Milestone:
Component: Cactus website | Version:
Keywords: |
----------------------------+-----------------------------------------------
When I download GetComponents using the instructions at
http://cactuscode.org/documentation/GetComponents.php
I get the following warning when I run it:
You are using an outdated version of GetComponents. Please use the version
from GitHub (https://github.com/gridaphobe/CRL) (if you are using the
Einstein Toolkit, the new version should already be in
Cactus/bin/GetComponents
The version of GetComponents on the Cactus website should be automatically
updated from the GitHub repository when the repository is changed. It
should not be necessary for users to have Git to check out Cactus.
--
Ticket URL: <https://trac.einsteintoolkit.org/ticket/338>
Einstein Toolkit <http://einsteintoolkit.org>
The Einstein Toolkit
#216: Thorns in ExternalLibraries should abort if the old "extras" mechanism has
been selected
-------------------------------+--------------------------------------------
Reporter: hinder | Owner:
Type: enhancement | Status: new
Priority: minor | Milestone:
Component: Cactus | Version:
Keywords: ExternalLibraries |
-------------------------------+--------------------------------------------
The old way of specifying the location of external libraries, from
Cactus/lib/make/extras, for example setting HDF5 = yes, is not compatible
with the new mechanism where you must simply include the appropriate thorn
from ExternalThorns. Using both can be confusing.
I propose modifying the thorns in ExternalLibraries to check if the old
mechanism has been selected, and to abort with an explanatory error
message telling the user not to set, e.g. HDF5 = yes, if they are using
ExternalLibraries/HDF5.
I am attaching a completely untested patch for the HDF5 thorn which
implements something like this. The code to determine if the user
selected HDF5 = yes was taken from the extras directory.
Comments? If this is appropriate, it can be adapted to all the thorns in
ExternalLibraries.
--
Ticket URL: <https://trac.einsteintoolkit.org/ticket/216>
Einstein Toolkit <http://einsteintoolkit.org>
The Einstein Toolkit
#318: add "Report bug" link to Einstein Toolkit sidebar item
-------------------------------------+--------------------------------------
Reporter: rhaas | Owner: rhaas
Type: enhancement | Status: new
Priority: optional | Milestone:
Component: EinsteinToolkit website | Version:
Keywords: |
-------------------------------------+--------------------------------------
Is there a link somehwere on the ET website to link to the Issue tracker?
If not I'd like to add it undernead the Contribute link, ie:
{{{
Index: global/navigation.php
===================================================================
--- global/navigation.php (revision 463)
+++ global/navigation.php (working copy)
@@ -16,6 +16,7 @@
<li><a href="/toolkit/capabilities/">Capabilities</a></li>
<li><a href="/toolkit/community/">Mail Lists</a></li>
<li><a href="/info/contribute/">Contribute</a></li>
+ <li><a href="https://trac.einsteintoolkit.org">Report
issues</a></li>
<li><a href="/info/citing/">Cite</a></li>
</ul></li>
}}}
--
Ticket URL: <https://trac.einsteintoolkit.org/ticket/318>
Einstein Toolkit <http://einsteintoolkit.org>
The Einstein Toolkit
#302: Multipole HDF5 file truncation with multiple variables
-----------------------+----------------------------------------------------
Reporter: hinder | Owner:
Type: defect | Status: new
Priority: minor | Milestone:
Component: Other | Version:
Keywords: Multipole |
-----------------------+----------------------------------------------------
Previously a "first_time" variable was used to determine if truncation
should happen. When decomposing multiple variables, this logic is
incorrect. We now store the first_time information per output file.
I have tested the patch.
--
Ticket URL: <https://trac.einsteintoolkit.org/ticket/302>
Einstein Toolkit <http://einsteintoolkit.org>
The Einstein Toolkit
#225: Patch: Create prototypes for all scheduled functions
-------------------------+--------------------------------------------------
Reporter: eschnett | Owner:
Type: enhancement | Status: new
Priority: major | Milestone:
Component: Cactus | Version:
Keywords: |
-------------------------+--------------------------------------------------
The attached patch makes the CST stage create prototypes for all scheduled
functions into a new file cctk_ScheduleFunctions.h, which is included into
cctk.h.
This is done only for C (and C++) since Fortran prototypes cannot be
declared at file scope.
--
Ticket URL: <https://trac.einsteintoolkit.org/ticket/225>
Einstein Toolkit <http://einsteintoolkit.org>
The Einstein Toolkit