Commits · eb6306160762803d8221bba37d4fdfb58ad4bbb2 · hpc-team / HPCasCode

Nov 13, 2017
- added check_nvidia_uvm.nhc · eb630616
  Jafar Lie authored 7 years ago
  
  Former-commit-id: fed3035e
  eb630616
Aug 30, 2017
- Adding new NHC for xorg.conf so that vglrun works on vis nodes · 785cca36
  Kerri Wait authored 7 years ago
  
  Former-commit-id: 6ef94fd1
  785cca36
Aug 11, 2017
- finally corrected the cuda collectd script · 700719e2
  Chris Hines authored 7 years ago
  
  Former-commit-id: 5a682b06
  700719e2
- Update cuda_collectd.py.j2 · 3aed1f4f
  Chris Hines authored 7 years ago
  
  Former-commit-id: 359f414a
  3aed1f4f
- Remove unused fan_speed monitoring, add monitoring of memory and gpu utilization · b1c0843d
  Chris Hines authored 7 years ago
  
  Former-commit-id: 008241b7
  b1c0843d
Aug 08, 2017
- forgot to include the buddyinfo script for collectd · 1e77381a
  Chris Hines authored 7 years ago
  
  Former-commit-id: 35af92cb
  1e77381a
Aug 07, 2017
- Added "set quorum ratio" command · 5c66255a
  Simon Michnowicz authored 7 years ago and Chris Hines committed 7 years ago
  
  Former-commit-id: 6d320631
  5c66255a
- changed a role to run on "+ delegate_to: "{{ gluster_servers[0] }}"" so it is... · 6dd91122
  Simon Michnowicz authored 7 years ago and Chris Hines committed 7 years ago
  
  changed a role to run on "+ delegate_to: "{{ gluster_servers[0] }}"" so it is consistent within playbook added a ping value and restarted the daemon. This was to try to fix startup bugs we see in Monarch. Former-commit-id: 7ef98227
  6dd91122
- collectd monitoring changes ... monitor additional things in proc, don't... · 0d2c7a53
  Chris Hines authored 7 years ago
  
  collectd monitoring changes ... monitor additional things in proc, don't monitor slabinfo as its noisy, monitor gpus Former-commit-id: 71b1c79d
  0d2c7a53
Jul 31, 2017
- nagios script for slurmctld · e6b0e826
  Chris Hines authored 7 years ago
  
  Former-commit-id: bf7146ab
  e6b0e826
Jul 13, 2017
- Update main.yml · 5c6e37d0
  Chris Hines authored 7 years ago
  
  Former-commit-id: 059f0f24
  5c6e37d0
Jul 11, 2017
- Remove adding cluster on slurm · d96df6d3
  Gin Tan (Monash University) authored 7 years ago
  
  Former-commit-id: 811cbc72
  d96df6d3
- Removed the /tmp clean-up until the script is sorted · 8986babd
  Gin Tan (Monash University) authored 7 years ago
  
  Former-commit-id: 1cfb95f8
  8986babd
- Removed the /tmp clean-up until the script is sorted · 661f1f9d
  Gin Tan (Monash University) authored 7 years ago
  
  Former-commit-id: 92795eb7
  661f1f9d
Jul 10, 2017
- Dropping cache on epilog & prolog · 48e3df57
  Gin Tan (Monash University) authored 7 years ago
  
  Former-commit-id: 1474dab1
  48e3df57
Jul 05, 2017
- Update the mofed version · 16d4e966
  Gin Tan (Monash University) authored 7 years ago
  
  Former-commit-id: 325432a2
  16d4e966
Jun 19, 2017
- add a collectd role that can be deploed to compute and login nodes · e6ff52f0
  Chris Hines authored 7 years ago
  
  Former-commit-id: b89bb713
  e6ff52f0
May 26, 2017
- update the hpcsystems role to include the naggy_quota cronjob and update the tools via pip · 1336e777
  Chris Hines authored 7 years ago
  
  Former-commit-id: 75d04efa
  1336e777
- fix the volcreate to be more sane · 27f880ec
  Chris Hines authored 7 years ago
  
  Former-commit-id: 55e2e3c7
  27f880ec
May 15, 2017
- Removed id_rsa and .pub . This is moved to clusterbuild files dir for security reasons · 7c696618
  Simon Michnowicz (Monash University) authored 7 years ago
  
  Former-commit-id: 63ee9bab
  7c696618
May 11, 2017

Added new role to enable Slurm DB SQL to be backed up and then ssh to a node. · 39ce2637

Simon Michnowicz (Monash University) authored 7 years ago

Location of node, backup dir, and dummy user account are contained in defaults/main.yml
Both SQL nodes and Management node need to have this role applied, with 'server' parameter determining the different

Former-commit-id: ca6e0cd5

39ce2637

Apr 21, 2017
- add a flag to nvidia-xconfig because it is not inputing busids for the m3f class nodes · 1ed59a1d
  Chris Hines authored 7 years ago
  
  Former-commit-id: 6eedf852
  1ed59a1d
Mar 27, 2017

Update provision_slurm.py.j2 to import defaultdict to silence this error · de59bd73

Kerri Wait authored 7 years ago

Traceback (most recent call last):
 File "/usr/local/sbin/provision_slurm.py", line 100, in <module>
   mk_slurmuser_batch(usergroup,"default")
 File "/usr/local/sbin/provision_slurm.py", line 67, in mk_slurmuser_batch
   userdict = defaultdict(list)
NameError: global name 'defaultdict' is not defined
Former-commit-id: 0cccb855

de59bd73

Mar 21, 2017
- changed zz_modulecmd to modulecmd, so the list of custom modules load properly · b7192849
  Jafar Lie authored 8 years ago
  
  Former-commit-id: 6efcad21
  b7192849
Mar 20, 2017
- Update provision_slurm.py.j2 with indents again · ca34179b
  Kerri Wait authored 8 years ago
  
  Former-commit-id: 03f9de20
  ca34179b
- Update provision_slurm.py.j2 because of indents · 7fd857bf
  Kerri Wait authored 8 years ago
  
  Former-commit-id: 9b0886d0
  7fd857bf
- Update provision_slurm.py.j2 to fix logic in mk_slurmuser_batch. This... · 7a65e094
  Kerri Wait authored 8 years ago
  
  Update provision_slurm.py.j2 to fix logic in mk_slurmuser_batch. This previously did not add any users if one user in the list was a member of 'default' Former-commit-id: 94ccd782
  7a65e094
Mar 10, 2017
- Cron file for Lustre Backup · 3a8816bc
  Gin Tan (Monash University) authored 8 years ago
  
  Former-commit-id: 51ace756
  3a8816bc
Feb 23, 2017
- tempalte xorg.conf when running on a node with 1GPU (this template works for... · 77a47e83
  Chris Hines authored 8 years ago
  
  tempalte xorg.conf when running on a node with 1GPU (this template works for M3 K1 nodes, will need to do something smarter for other clusters with GPUS) Former-commit-id: 285020e9
  77a47e83
- make sure we can install the hpcsystems stack · a8a79331
  Chris Hines authored 8 years ago
  
  Former-commit-id: a279e781
  a8a79331
- make things quieter in check mode. Also improve GPU upgrade by killing X servers if necessary · 661d64fd
  Chris Hines authored 8 years ago
  
  Former-commit-id: cf9d4321
  661d64fd
- remove excess debug statements from gluster_volcreate · 29344060
  Chris Hines authored 8 years ago
  
  Former-commit-id: 83e6900b
  29344060
- better filters to only execute shell commands when necessary · b580ddb9
  Chris Hines authored 8 years ago
  
  Former-commit-id: 9dc1444c
  b580ddb9
- further tweaks to allow /usr/local/bin but only as a last choice · d60f9bf9
  Chris Hines authored 8 years ago
  
  Former-commit-id: 4953330a
  d60f9bf9
- better template for gres.conf allowing for Grid K1 GPUS and different CPU counts · a20f3660
  Chris Hines authored 8 years ago
  
  Former-commit-id: 2aee70cb
  a20f3660
Feb 07, 2017
- missing pip from the list of dependencies · d4b02a3b
  Chris Hines authored 8 years ago
  
  Former-commit-id: 61df8cf6
  d4b02a3b
Feb 01, 2017
- executable bit should not be set on systemd service files · 2ad797fe
  Chris Hines authored 8 years ago
  
  Former-commit-id: 1af0bccc
  2ad797fe
- start ntpd by default · 0667a461
  Chris Hines authored 8 years ago
  
  Former-commit-id: 264d1610
  0667a461
Jan 06, 2017

Extend the GPU role · eca82477

Chris Hines authored 8 years ago

The GPU role used to just detect if any driver was installed and skip installation if one was found
it will now execute nvidia-smi to determine the current driver version and compare it against the desired
driver version, installing in the case of mismatch

Note that if an existing driver is installed, the nvidia persistence daemon must be stopped before installation can proceed

Former-commit-id: 2b1ad616

eca82477

Jan 04, 2017

Alter the way gluster peer probe works · 822eba0f

Chris Hines authored 8 years ago

I had the syntax incorrect when the glsuter cluster consists of more than two nodes. We're not attempting to use three node clusters so we can't get split brain


Former-commit-id: c94759f4

822eba0f