- 12 Aug, 2015 1 commit
-
-
Morris Jette authored
Remove redundant persist_rem field from job record Fix logic for buffer state related to failure handling Move pre_run logic into a separate thread for improved system performance Disable changes to job burst buffer field burst_buffer contents are validated at job submit time and data is possibly being staged at later times. It can not be changed. Refactor bb_p_job_cancel() logic Fix some job staging problems Eliminate logging of unchanged pool information Document that job arrays don't support burst buffers Refactor job buffer allocation to eliminate race condition with stage out. Needs more work after prolog_running functionality is in place (merged from v14.11). Improve cold-start state clean up
-
- 11 Aug, 2015 1 commit
-
-
Morris Jette authored
Clean-up of of job stage-out and teardown logic Improve restart state clean-up
-
- 10 Aug, 2015 4 commits
-
-
Morris Jette authored
Fix some job state logic Add logic to preserve persistent buffer account, QOS and partition for limit enforcement
-
Morris Jette authored
Fix slurm.spec file for recent layouts change Refactor bb_pack_bufs() and bb_alloc_job_rec() functions
-
Morris Jette authored
Add partition information to buffer record for limit enforcement
-
jette authored
Add account and QOS to burst buffer data structures for TRES limit work sview: Correct size units sview: Fix free of uninitialized iterator logic burst_buffer.conf: Fix parsing of DefaultPool parameter
-
- 09 Aug, 2015 2 commits
- 08 Aug, 2015 2 commits
-
-
jette authored
-
Morris Jette authored
Move limits checks to burst_buffer/common Disable persistent burst buffer with numeric value Refactor bb_p_job_get_est_start() logic Split out persistent buffer create/delete into a pthread Fix job buffer size parsing logic Get through stage-in and support for non-Cray systems (e.g. VM)
-
- 07 Aug, 2015 1 commit
-
-
Morris Jette authored
Add new buffer state of "BB_STATE_PENDING" Refactor job buffer state structure Refactor job buffer creation environment Rename "bb_hash" to "bb_ahash" Set up job state hash table Add magic cookies to data structures Rename bb_find_job_rec() to bb_find_alloc_rec() Rename bb_free_rec() to bb_free_alloc_buf()
-
- 06 Aug, 2015 4 commits
-
-
David Bigagli authored
add option --enable-really-no-cray to allow vanilla linux installatio…
-
Doug Jacobsen authored
add option --enable-really-no-cray to allow vanilla linux installations on Cray external login nodes
-
Yiannis Georgiou authored
Remove the "power_cpufreq" layout-type and transfer the logic on the "power" layout-type, by creating a new layout named "cpufreq" under the already existing layout-type "power". Based on this modification adjust the logic of powercapping.
-
Yiannis Georgiou authored
-
- 05 Aug, 2015 2 commits
-
-
Morris Jette authored
This reverts part of commit c2fbf88f see bug 1846
-
Morris Jette authored
-
- 04 Aug, 2015 4 commits
-
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
Nathan Yee authored
-
- 03 Aug, 2015 5 commits
-
-
jette authored
Renames some sview variables for better clarity and put some node fields into alphabetic order
-
Dorian Krause authored
-
Dorian Krause authored
This commit adds support for free_mem as part of the sinfo command using the key 'e' ('f' and 'F' are already taken).
-
Dorian Krause authored
This commit adds information about the free memory to the output created by slurm_sprint_node_table(). The unit of FreeMem is MiB like RealMemory.
-
Dorian Krause authored
This commit adds a new field free_mem to the node_info struct and augments the 15.08 RPC protocol to exchange this information as part of the ping and registration status messages. The free memory is taken from the sysinfo() information on systems that support this function. Overall, the implementation sticks to the way cpu_load (the 5 min load average) is handled.
-
- 31 Jul, 2015 5 commits
-
-
Danny Auble authored
note the change and remove the new time limit if the job is requeued.
-
Danny Auble authored
-
Danny Auble authored
and give different error codes for different limits violated.
-
Morris Jette authored
A job being considered for backfill scheduling can have its time limit altered to the partition limit, job minimum time limit, etc. This fixes two places where the backfill scheduler could leave the job in a pending state with a time limit different from it's original value.
-
Morris Jette authored
add more startup/initialization logic fix some locking issues fix some issues related to persistent buffers
-
- 30 Jul, 2015 4 commits
-
-
Danny Auble authored
-
Morris Jette authored
argument "size" to _print_topo_record() could be zero, resulting in invalid array size specification of zero. Dead assignment to "len" in second call to _print_topo_record().
-
Morris Jette authored
Add accounting for persistent buffer space Fixes for data structure removal
-
David Bigagli authored
-
- 29 Jul, 2015 1 commit
-
-
Morris Jette authored
Add "job_process" function. Minor clean-up from code review.
-
- 28 Jul, 2015 3 commits
-
-
Morris Jette authored
Add logic to purge persistent burst buffers. Create dummy job script as needed for buffer teardown Multiple persistent buffer create/destroy calls in single job working
-
Thomas Cadeau authored
-
Thomas Cadeau authored
-
- 27 Jul, 2015 1 commit
-
-
Morris Jette authored
-