-
Notifications
You must be signed in to change notification settings - Fork 5
Closed
Description
-
MODULEPATH*variables get initialized as expected ; for both centos6 & centos7, with or w/out Tmod -
ml --gitversion;module --versionreport reasonable output, consistent to RPM package name - modules behave persistently and coherently across parent-children shell invocations (excl:
shared?)- invoking
bashdoes NOT reset the environment, and letsloads anduseoperations be consistent - invoking
tcshinstead also preserves the modules environment as it should, without any surprises - also,
module unloads should function properly (fi.-sgevs empty$LD_LIBRARY_PATH) ref. issuemodule unloadshould work correctly with Lmod #45
- invoking
-
SitePackage.luagets honored and syslog messages get injected aftermodule load -
module save,module savelistandmodule restorebehave correctly -
bashtab completion should be OK too, also on successive operands; maybe alsozsh -
$USER_IS_ROOThas correct value, as well as$LMOD_SYSTEM_DEFAULT_MODULES- functions work correctly for both
rootand non-rootusers -
sgeandclustermodules, and any dependencies (fi.$SITENAME, get autoloaded)
- functions work correctly for both
-
tcshandbashboth initialize/behave correctly- also under
qsubandqrsh - also under
crontab(at least withbash -lsupport) &at
- also under
- bash script tracing should work consistently at all times (Expand work on commit 4c92836, because it breaks bash tracing, as it has now #64)
- and also pass: https://github.com/rtmclay/Lmod_test_suite (ok, modulo the infamous bash bug)
Reactions are currently unavailable