- 22 Feb, 2018 9 commits
-
-
Felip Moll authored
-
Felip Moll authored
Only a single io_timeout_thread should be created for each sls struct. Creating multiple, while seemingly harmless in operation, can lead to fatal() messages when srun shuts down by destroying mutex locks that are in use by threads that srun doesn't expect to still have running. Regression caused by a1185f04. Bug 4596
-
Morris Jette authored
Bug 4806.
-
Felip Moll authored
This patch fixes the situation that makes features unrecognized where a node features plugin is active and features are defined to nodes in slurm.conf. It also preserves KNL node features when slurmctld daemons are reconfigured including active and available modes. Features not belonging to node features plugin are reset to what is in slurm.conf when restarting or reconfiguring. Bug 4734
-
Alejandro Sanchez authored
-
Alejandro Sanchez authored
_resv_overlap function was only checking the flags for the updated reservation, but not for the rest of present ones. This implied that the allowed overlap derived from these flags only applied depending on the update order. Bug 4806.
-
Alejandro Sanchez authored
-
Alejandro Sanchez authored
After commit b31fa177, we do not defer slurmd node registration if HealthCheckProgram fails. So at slurmd startup, slurmd executes: run_script_health_check(); _spawn_registration_engine(); And does not keeps spinning if NHC fails. Now if there are nodes managed by the Power Save logic, when they are requested to be POWER_UP because a job is allocated resources, then at slurmd startup NHC is executed before node registers. The problem comes when this NHC execution fails, if the NHC program decides to update the node to DRAIN, since the job was already allocated before this update, then the job will attempt to start RUNNING but might fail since NHC detected there's something wrong. So this change what it does is to detect DRAIN/FAIL node update requests, then check if node is ALLOC/MIXED and POWER_[SAVE|UP] and if so then force a requeue, so that the job doesn't start on a failed node. Bug 4689.
-
Felip Moll authored
Can frequently throw scary-sounding messages on short-lived processes that disappear while the stats are collected. Bug 4759.
-
- 21 Feb, 2018 20 commits
-
-
Brian Christiansen authored
-
Chris Samuel authored
Bug 4793
-
Morris Jette authored
-
Morris Jette authored
When a pending job's state includes "UnavailableNodes" do not include the nodes in FUTURE state. bug 4612
-
Brian Christiansen authored
Bug 4504
-
Brian Christiansen authored
submission option. Bug 4548
-
Brian Christiansen authored
Bug 4548
-
Morris Jette authored
Bug 4786
-
Brian Christiansen authored
Continuation of 873871ed Bug 4758
-
Danny Auble authored
Bug 4758
-
Brian Christiansen authored
Bug 4800
-
Morris Jette authored
-
Isaac Hartung authored
Bug 4793
-
Alejandro Sanchez authored
-
Danny Auble authored
Bug 4698.
-
Morris Jette authored
-
Tim Wickberg authored
-
Dominik Bartkiewicz authored
Bug 4809.
-
Dominik Bartkiewicz authored
NEWS entry also applies to prior 31667d5d. Bug 4809.
-
Dominik Bartkiewicz authored
Fixes problems restoring: - Alternate - DefMemPerCPU - MaxCPUsPerNode - MaxMemPerNode - OverTimeLimit Also move AllowNodes, DefaultTime, and GraceTime into alphabetical order. Bug 4809.
-
- 20 Feb, 2018 9 commits
-
-
Morris Jette authored
Configuration parameters "ControlMachine", "ControlAddr", "BackupController" and "BackupAddr" replaced by an ordered list of "SlurmctldHost" records with the optional address appended to the name enclosed in parenthesis. For example: "SlurmctldHost=head(12.34.56.78)". An arbitrary number of backup servers can be configured.
-
Morris Jette authored
-
Marshall Garey authored
Bug 4636
-
Brian Christiansen authored
Bug 4716
-
Brian Christiansen authored
Bug 4716
-
Felip Moll authored
In perl tools, fix for regexp that caused extra incorrectly shown results. Bug 4766
-
Danny Auble authored
-
Danny Auble authored
requesting any tasks. Bug 4730
-
Morris Jette authored
wait to retry or not. I discovered this bug regression testing. Some similar situations will result in srun continuously issuing step create requests and the launch_common_create_job_step() function not sleeping between RPCs. Basically launch_common_create_job_step() sleeps for some error codes and srun retries the step create on some error codes. The problem is that those error codes do not match in both places, resulting in constant retries without sleeps. This situation is very likely with job preemption combined with salloc, but other conditions can trigger the same event. The following errno will all trigger this situation: EAGAIN, ESLURM_DISABLED, ESLURM_POWER_NOT_AVAIL, ESLURM_POWER_RESERVED, ESLURM_PROLOG_RUNNING, ESLURM_INTERCONNECT_BUSY. Bug 4786
-
- 16 Feb, 2018 2 commits
-
-
Brian Christiansen authored
Bug 4772
-
Brian Christiansen authored
Was checking time ranges against a bad cast'ed variable. And MaxQueryTimeRange was being stored as minutes and was being compared against the difference of two time_t's -- which produces seconds. Bug 4772
-