- 11 Mar, 2019 10 commits
-
-
Tim Wickberg authored
Plugins do not and should not keep pointers in to buf.
-
Tim Wickberg authored
-
Tim Wickberg authored
-
Tim Wickberg authored
And update all calling locations. Note that this removes a lot of unnecessary calls to slurm_get_auth_info() which should serve as a slight performance speedup.
-
Tim Wickberg authored
Now unused. This info should only be needed at verification time, which must happen before any slurm_auth_get_*() calls.
-
Tim Wickberg authored
-
Tim Wickberg authored
-
Morris Jette authored
bug 6646
-
Morris Jette authored
-
Morris Jette authored
Insure that each MPS server has a single/common percentage of a GPU
-
- 09 Mar, 2019 3 commits
-
-
Morris Jette authored
Add module to the Makefiles Add test description to README (used for logging to stderr) If "date" fails and an error (FAILURE) is logged, then make the test have an exit 1 If the test can not run due to configuration then do not log as a failure and exit 0
-
Morris Jette authored
Coverity CID 195594
-
Morris Jette authored
Coverity CID 195725
-
- 08 Mar, 2019 27 commits
-
-
Tim Wickberg authored
-
Broderick Gardner authored
Bug 6633.
-
Morris Jette authored
Coverity CID 195599
-
Morris Jette authored
Coverity CID 195608
-
Morris Jette authored
Coverity CID 195595
-
Morris Jette authored
Coverity CID 195595
-
Morris Jette authored
Coverity CID 195594
-
Morris Jette authored
Coverity CID 195603 and 195605
-
Morris Jette authored
Coverity CID 195610
-
Morris Jette authored
Coverity CID 195592
-
Morris Jette authored
Coverity CID 195593
-
Morris Jette authored
Coverity CID 195604
-
Morris Jette authored
Coverity CID 195609
-
Morris Jette authored
Coverity CID 195606
-
Morris Jette authored
Coverity CID 195598
-
Morris Jette authored
Coverity CID 187785
-
Tim Wickberg authored
Use slurm_seterrno(), rather than maintaining a plugin-local and credential local version of that errno. Remove g_slurm_auth_errno() and g_slurm_auth_errstr(). Replace all calling locations with %m format identifiers which should pick up the errno in that thread, which is now directly set by the auth plugin. The one place that this represents a slight regression in functionality is that we loose the specific EMUNGE errno values. However, the places that will set this in the auth/munge plugin are already printing it through an error() with munge_ctx_strerror(). Replace those two locations with ESLURM_AUTH_CRED_INVALID instead.
-
Tim Wickberg authored
While currently safe, this is blocking further refactoring, and much less conceptually clean than testing the cred properly with g_slurm_auth_verify().
-
Tim Wickberg authored
Stop fetching a lot of (now) unused fields, and print the two relevant lines. Drop errors to warnings - we know the decode failed, so failure to retrieve the encode and decode time is quite likely.
-
Tim Wickberg authored
-
Tim Wickberg authored
-
Tim Wickberg authored
Unused, and until earlier in this commit series the auth/munge plugin was printing to stdout directly rather than the fp argument.
-
Tim Wickberg authored
Move the message for ESLURM_AUTH_UNPACK into slurm_auth.c, which removes all error message handling except the complicated MUNGE handing from the plugins. Adjust MUNGE_ERRNO_OFFSET far enough to avoid the range these have just been renumbered into.
-
Tim Wickberg authored
Use a new cred_wrapper_t struct designed to overlap the opaque cred structures used within the individual plugins. Add the index value to the start of each of these structures. The index is set when a given credential is unpacked, and used to lookup the appropriate plugin for use with the remaining auth plugin API calls.
-
Tim Wickberg authored
The zeroth element is always the AuthType, and will be used by default. This is currently only supported in the slurmctld and slurmdbd; ignore this in all other calling locations.
-
Tim Wickberg authored
-
Tim Wickberg authored
-