From ecf681642b108bae39f797eb4b44e2df357a7b08 Mon Sep 17 00:00:00 2001
From: Danny Auble <da@llnl.gov>
Date: Fri, 14 Apr 2006 18:44:34 +0000
Subject: [PATCH] finished documentation for jobacct plugin
---
NEWS | 8 +-
doc/html/big_sys.shtml | 4 +-
doc/html/configurator.html | 8 +-
doc/html/jobacctplugins.shtml | 273 ++++++++++++++++--
doc/html/quickstart_admin.shtml | 2 +
doc/man/man5/slurm.conf.5 | 33 +--
etc/slurm.conf.example | 7 +-
src/common/slurm_jobacct.h | 2 +-
.../jobacct/common/common_slurmstepd.c | 2 +-
src/plugins/jobacct/common/jobacct_common.c | 19 +-
src/sacct/print.c | 77 +++--
11 files changed, 318 insertions(+), 117 deletions(-)
diff --git a/NEWS b/NEWS
index 57f428c3a71..72a3cb81510 100644
--- a/NEWS
+++ b/NEWS
@@ -4,15 +4,17 @@ documents those changes that are of interest to users and admins.
* Changes in SLURM 1.1.0-pre5
=============================
-- Added step completion RPC logic
- -- Vastly changed sacct and the jobacct plugin. Needs more documentation
- still.
+ -- Vastly changed sacct and the jobacct plugin. Read documentation for full
+ details.
-- Added jobacct plugin for AIX and BlueGene, they currently don't work,
- but infastructure is in place.
+ but infrastructure is in place.
-- Add support for srun option --ctrl-comm-ifhn to set PMI communications
address (Hongjia Cao, National University of Defense Technology).
-- Moved safe_read/write to slurm_protocol_defs.h removing multiple copies.
-- Remove vestigial functions slurm_allocate_resources_and_run() and
slurm_free_resource_allocation_and_run_response_msg().
+ -- moved the way forward logic waited for fanout logic mostly eliminating
+ problems with scalability issues.
* Changes in SLURM 1.1.0-pre4
=============================
diff --git a/doc/html/big_sys.shtml b/doc/html/big_sys.shtml
index ef9949cbea1..0aa2e6e0673 100644
--- a/doc/html/big_sys.shtml
+++ b/doc/html/big_sys.shtml
@@ -21,7 +21,7 @@ is best avoided.</p>
<h2>Job Accounting Plugin (JobAcctType)</h2>
-<p>Job accounting relies upon the <i>slurmd</i> daemon on each compute
+<p>Job accounting relies upon the <i>slurmstepd</i> daemon on each compute
node periodically sampling data.
This data collection will take compute cycles away from the application
inducing what is known as <i>system noise</i>.
@@ -32,7 +32,7 @@ is best (<i>jobacct/none</i>).
Consider use of job completion records (<i>JobCompType</i>) for accounting
purposes as this entails far less overhead.
If job accounting is required, configure the sampling interval
-to a relatively large size (e.g. <i>JobAcctParameters="Frequency=300"</i>).
+to a relatively large size (e.g. <i>JobAcctFrequency=300</i>).
Some experimentation may also be required to deal with collisions
on data transmission.</p>
diff --git a/doc/html/configurator.html b/doc/html/configurator.html
index 729bff00e1e..55f8430e395 100644
--- a/doc/html/configurator.html
+++ b/doc/html/configurator.html
@@ -117,7 +117,7 @@ function displayfile()
get_field("JobCompLoc",document.config.job_comp_loc) + "<br>" +
"JobAcctType=jobacct/" + get_radio_value(document.config.job_acct_type) + "<br>" +
get_field("JobAcctLogfile",document.config.job_acct_logfile) + "<br>" +
- "#JobAcctParameters= <br>"+
+ get_field("JobAcctFrequency",document.config.job_acct_frequency) + "<br>" +
"# <br>" +
"# COMPUTE NODES <br>" +
"NodeName=" + document.config.node_name.value +
@@ -222,7 +222,7 @@ to start up (and may eventually time-out) because the NIS server(s)
may not be able to quickly respond to simultaneous requests from
multiple slurmd's. You can instruct slurmd to cache /etc/groups
entries to prevent this from happening by setting
-<B>CacheGroups</B>=1". Reconfiguring ("scontrol reconfig") with
+<B>CacheGroups</B>=1. Reconfiguring ("scontrol reconfig") with
<B>CacheGroups</B>=0 will cause slurmd to purge the cache. Select one
value for <B>CacheGroups</B>:<BR>
<input type="radio" name="cache_groups" value="0" checked>
@@ -392,9 +392,11 @@ SLURM accounts for resource use per job. System specifics can be polled
determined by system type<BR>
Select one value for <B>JobAcctType</B>:<BR>
<input type="radio" name="job_acct_type" value="none" checked> <B>None</B>: No
-extra information gathered<BR>
+job accounting<BR>
<input type="radio" name="job_acct_type" value="linux"> <B>Linux</B>: Specifc
Linux proc table information gathered, use with Linux systems only<BR>
+<input type="text" name="job_acct_frequency" value=""> <B>JobAcctFrequency</B>:
+polling interval in seconds.<BR>
<input type="text" name="job_acct_logfile" value=""> <B>JobAcctLogFile</B>:
Location specification.
This is the location of the text file to be written to (used by Log only).Use a fully
diff --git a/doc/html/jobacctplugins.shtml b/doc/html/jobacctplugins.shtml
index f76ae517cb3..8aa51658cbc 100644
--- a/doc/html/jobacctplugins.shtml
+++ b/doc/html/jobacctplugins.shtml
@@ -5,7 +5,7 @@
<h2> Overview</h2>
<p> This document describes SLURM job accounting plugins and the API that
defines them. It is intended as a resource to programmers wishing to write
-their own SLURM job accounting plugins. This is version 0 of the API.
+their own SLURM job accounting plugins. This is version 1 of the API.
<p>SLURM job accounting plugins must conform to the
@@ -24,62 +24,288 @@ The minor type can be any suitable name
for the type of accounting package. We currently use
<ul>
<li><b>linux</b>—Gathers information from linux proctable and addes this
-information to the standard rusage information already gathered for each
-job.
-<li><b>none</b>—No extra information gathered.
+information to the standard rusage information also gathered for each job.
+<li><b>none</b>—No information gathered.
</ul>
The <b>sacct</b> program can be used to display gathered data from regular
accounting and from these plugins.
<p>The programmer is urged to study
-<span class="commandline">src/plugins/jobacct/linux</span>
+<span class="commandline">src/plugins/jobacct/linux</span> and
+<span class="commandline">src/plugins/jobacct/common</span>
for a sample implementation of a SLURM job accounting plugin.
<p class="footer"><a href="#top">top</a>
<h2>API Functions</h2>
-The job accounting API uses hooks in the slurmstepd.
+The job accounting API uses hooks in the slurmctld, slurmd, and slurmstepd.
<p>All of the following functions are required. Functions which are not
implemented must be stubbed.
-
-
<h4>Functions called by all slurmstepd processes</h4>
-<p class="commandline">int jobacct_p_init(int frequency)
+<p class="commandline">int jobacct_p_startpoll(int frequency)
<p style="margin-left:.2in"><b>Description</b>:
-jobacct_p_init() is called at the beginning of each process at the start of the
-slurmstepd, this starts a thread that should poll information to be gathered
-at the end of the process. Put global initialization here.
+jobacct_p_startpoll() is called at the start of the slurmstepd,
+this starts a thread that should poll information to be queried at any time
+during throughout the end of the process.
+Put global initialization here.
<p style="margin-left:.2in"><b>Arguments</b>:
-<span class="commandline">frequency</span> (input) poll frequency for gathering
+<span class="commandline">frequency</span> (input) poll frequency for polling
thread.
<p style="margin-left:.2in"><b>Returns</b>:
<span class="commandline">SLURM_SUCCESS</span> on success, or
<span class="commandline">SLURM_FAILURE</span> on failure.
-<p class="commandline">int jobacct_p_fini(slurmd_job_t *job)
+<p class="commandline">int jobacct_p_endpoll()
<p style="margin-left:.2in"><b>Description</b>:
-jobacct_p_fini() is called when the process is finished gathered information is
-placed in the slurmd_job_t structure. This function will also stop the
-gathering thread.
+jobacct_p_endpoll() is called when the process is finished to stop the
+polling thread.
<p style="margin-left:.2in"><b>Arguments</b>:
-<span class="commandline">job</span> (input) structure to hold information
-gathered from gathering thread.
+<span class="commandline">none</span>
<p style="margin-left:.2in"><b>Returns</b>:
<span class="commandline">SLURM_SUCCESS</span> on success, or
<span class="commandline">SLURM_FAILURE</span> on failure.
-<p class="commandline">int jobacct_p_suspend()
+<p class="commandline">void jobacct_p_suspendpoll()
<p style="margin-left:.2in"><b>Description</b>:
-jobacct_p_suspend() is called when the process is suspended by the slurmctld.
+jobacct_p_suspendpoll() is called when the process is suspended or resumed.
This causes the polling thread to halt until the process is resumed.
<p style="margin-left:.2in"><b>Arguments</b>:
<span class="commandline">none</span>
<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">none</span>
+
+<p class="commandline">int jobacct_p_add_task(pid_t pid, uint16_t tid)
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_add_task() used to add a task to the poller.
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline"> pid</span> (input) Process id
+<span class="commandline"> tid</span> (input) slurm global task id
+<p style="margin-left:.2in"><b>Returns</b>:
<span class="commandline">SLURM_SUCCESS</span> on success, or
<span class="commandline">SLURM_FAILURE</span> on failure.
+
+<p class="commandline">jobacctinfo_t *jobacct_p_stat_task(pid_t pid)
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_stat_task() used to get most recent information about task.
+DO NOT FREE the information returned by this function!
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline"> pid</span> (input) Process id
+<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">jobacctinfo structure pointer</span> on success, or
+<span class="commandline">NULL</span> on failure.
+
+<p class="commandline">int jobacct_p_remove_task(pid_t pid)
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_remove_task() used to remove a task from the poller.
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline"> pid</span> (input) Process id
+<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">SLURM_SUCCESS</span> on success, or
+<span class="commandline">SLURM_FAILURE</span> on failure.
+
+<p class="footer"><a href="#top">top</a>
+
+<h4>Functions called by the slurmctld process</h4>
+
+<p class="commandline">int jobacct_p_init_slurmctld(char *job_acct_log)
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_init_slurmctld() is called at the start of the slurmctld,
+this opens the logfile to be written to.
+Put global initialization here.
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline">job_acct_log</span> (input) logfile name.
+<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">SLURM_SUCCESS</span> on success, or
+<span class="commandline">SLURM_FAILURE</span> on failure.
+
+<p class="commandline">int jobacct_p_fini_slurmctld()
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_fini_slurmctld() is called at the end of the slurmctld,
+this closes the logfile.
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline">none</span>
+<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">SLURM_SUCCESS</span> on success, or
+<span class="commandline">SLURM_FAILURE</span> on failure.
+
+<p class="commandline">
+int jobacct_p_job_start_slurmctld(struct job_record *job_ptr)
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_job_start_slurmctld() is called at the allocation of a new job in
+the slurmctld, this prints out beginning information about a job.
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline">job_ptr</span> (input) information about the job in
+slurmctld.
+<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">SLURM_SUCCESS</span> on success, or
+<span class="commandline">SLURM_FAILURE</span> on failure.
+
+<p class="commandline">
+int jobacct_p_job_complete_slurmctld(struct job_record *job_ptr)
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_job_complete_slurmctld() is called at the end of a job in
+the slurmctld, this prints out ending information about a job.
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline">job_ptr</span> (input) information about the job in
+slurmctld.
+<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">SLURM_SUCCESS</span> on success, or
+<span class="commandline">SLURM_FAILURE</span> on failure.
+
+<p class="commandline">
+int jobacct_p_step_start_slurmctld(struct step_record *step_ptr)
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_step_start_slurmctld() is called at the allocation of a new step in
+the slurmctld, this prints out beginning information about a step.
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline">step_ptr</span> (input) information about the step in
+slurmctld.
+<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">SLURM_SUCCESS</span> on success, or
+<span class="commandline">SLURM_FAILURE</span> on failure.
+
+<p class="commandline">
+int jobacct_p_step_complete_slurmctld(struct step_record *step_ptr)
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_step_complete_slurmctld() is called at the end of a step in
+the slurmctld, this prints out ending information about a step.
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline">step_ptr</span> (input) information about the step in
+slurmctld.
+<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">SLURM_SUCCESS</span> on success, or
+<span class="commandline">SLURM_FAILURE</span> on failure.
+
+<p class="commandline">
+int jobacct_p_suspend_slurmctld(struct job_record *job_ptr)
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_suspend_slurmctld() is called when a job is suspended or resumed in
+the slurmctld, this prints out information about the suspension of the job
+to the logfile.
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline">job_ptr</span> (input) information about the job in
+slurmctld.
+<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">SLURM_SUCCESS</span> on success, or
+<span class="commandline">SLURM_FAILURE</span> on failure.
+
+<p class="footer"><a href="#top">top</a>
+
+<h4>Functions common to all processes</h4>
+
+<p class="commandline">
+int jobacct_p_init_struct(jobacctinfo_t *jobacct, uint16_t tid)
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_init_struct() is called to set the values of a jobacctinfo_t to
+initial values.
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline">jobacct</span>
+(input/output) structure to be altered.
+<span class="commandline">tid</span>
+(input) id of the task send in (uint16_t)NO_VAL if no specfic task.
+<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">SLURM_SUCCESS</span> on success, or
+<span class="commandline">SLURM_FAILURE</span> on failure.
+
+<p class="commandline">jobacctinfo_t *jobacct_p_alloc(uint16_t tid)
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_alloc() used to alloc a pointer to and initialize a
+new jobacctinfo structure.<br>
+You will need to free the information returned by this function!
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline">tid</span>
+(input) id of the task send in (uint16_t)NO_VAL if no specfic task.
+<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">jobacctinfo structure pointer</span> on success, or
+<span class="commandline">NULL</span> on failure.
+
+<p class="commandline">void jobacct_p_free(jobacctinfo_t *jobacct)
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_free() used to free the allocation made by jobacct_p_alloc().
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline">jobacct</span>
+(input) structure to be freed.
+<span class="commandline">none</span>
+<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">none</span>
+
+<p class="commandline">
+int jobacct_p_setinfo(jobacctinfo_t *jobacct,
+ enum jobacct_data_type type, void *data)
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_setinfo() is called to set the values of a jobacctinfo_t to
+specific values based on inputs.
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline">jobacct</span>
+(input/output) structure to be altered.
+<span class="commandline">type</span>
+(input) enum of specific part of jobacct to alter.
+<span class="commandline">data</span>
+(input) corresponding data to set jobacct part to.
+<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">SLURM_SUCCESS</span> on success, or
+<span class="commandline">SLURM_FAILURE</span> on failure.
+
+<p class="commandline">
+int jobacct_p_getinfo(jobacctinfo_t *jobacct,
+ enum jobacct_data_type type, void *data)
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_getinfo() is called to get the values of a jobacctinfo_t
+specific values based on inputs.
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline">jobacct</span>
+(input) structure to be queried.
+<span class="commandline">type</span>
+(input) enum of specific part of jobacct to get.
+<span class="commandline">data</span>
+(output) corresponding data to from jobacct part.
+<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">SLURM_SUCCESS</span> on success, or
+<span class="commandline">SLURM_FAILURE</span> on failure.
+
+<p class="commandline">
+void jobacct_p_aggregate(jobacctinfo_t *dest, jobacctinfo_t *from)
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_aggregate() is called to aggregate and get max values from two
+different jobacctinfo structures.
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline">dest</span>
+(input/output) initial structure to be applied to.
+<span class="commandline">from</span>
+(input) new info to apply to dest.
+<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">none</span>
+
+<p class="commandline">
+void jobacct_p_pack(jobacctinfo_t *jobacct, Buf buffer)
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_pack() pack jobacctinfo_t in a buffer to send across the network.
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline">jobacct</span>
+(input) structure to pack.
+<span class="commandline">buffer</span>
+(input/output) buffer to pack structure into.
+<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">none</span>
+
+<p class="commandline">
+void jobacct_p_unpack(jobacctinfo_t *jobacct, Buf buffer)
+<p style="margin-left:.2in"><b>Description</b>:
+jobacct_p_unpack() unpack jobacctinfo_t from a buffer received from
+the network.
+You will need to free the jobacctinfo_t returned by this function!
+<p style="margin-left:.2in"><b>Arguments</b>:
+<span class="commandline">jobacct</span>
+(input/output) structure to fill.
+<span class="commandline">buffer</span>
+(input) buffer to unpack structure from.
+<p style="margin-left:.2in"><b>Returns</b>:
+<span class="commandline">SLURM_SUCCESS</span> on success, or
+<span class="commandline">SLURM_FAILURE</span> on failure.
+
<p class="footer"><a href="#top">top</a>
<h2>Parameters</h2>
@@ -90,11 +316,10 @@ plugins, the job accounting API counts on three parameters:
<dd>Specifies which plugin should be used.
<dt><span class="commandline">JobAcctFrequency</span>
<dd>Let the plugin know how long between pollings.
+<dt><span class="commandline">JobAcctLogFile</span>
+<dd>Let the plugin the name of the logfile to use.
</dl>
-<p class="footer"><a href="#top">top</a>
-
-
<h2>Versioning</h2>
<p> This document describes version 1 of the SLURM Job Accounting API. Future
releases of SLURM may revise this API. A job accounting plugin conveys its
diff --git a/doc/html/quickstart_admin.shtml b/doc/html/quickstart_admin.shtml
index 570f3b75d09..bee2b436f69 100644
--- a/doc/html/quickstart_admin.shtml
+++ b/doc/html/quickstart_admin.shtml
@@ -533,6 +533,8 @@ SlurmdSpoolDir = /tmp/slurmd
SlurmdTimeout = 300
TreeWidth = 50
JobAcctLogFile = /tmp/jobacct.log
+JobAcctFrequncy = 5
+JobAcctType = jobacct/linux
SLURM_CONFIG_FILE = /etc/slurm/slurm.conf
StateSaveLocation = /usr/local/tmp/slurm/adev
SwitchType = switch/elan
diff --git a/doc/man/man5/slurm.conf.5 b/doc/man/man5/slurm.conf.5
index 0b1430c8404..cc03977227a 100644
--- a/doc/man/man5/slurm.conf.5
+++ b/doc/man/man5/slurm.conf.5
@@ -141,38 +141,17 @@ May not exceed 65533.
Define the job accounting mechanism type.
Acceptable values at present include "jobacct/linux" and "jobacct/none".
The default value is "jobacct/none".
-In order to use the \fBsacct\fR tool, "jobacct/linux" must be configured.a
-\fBJobAcctLoc\fR
+In order to use the \fBsacct\fR tool, "jobacct/linux" must be configured.
+.TP
+\fBJobAcctLogFile\fR
Define the location where job accounting logs are to be written.
For jobacct/none this parameter is ignored.
For jobacct/linux this is the fully-qualified file name for the data file.
.TP
-\fBJobAcctParameters\fR
-Define any parameters to pass to the job accounting plugin, in a
-quoted and comma-separated list.
+\fBJobAcctFrequency\fR
+Define the polling frequencys to pass to the job accounting plugin.
For jobacct/none this parameter is ignored.
-For jobacct/linux the parameters available are
-.RS
-.TP
-\fBFrequency=N\fR, where N is the number of seconds between
-sampling the memory usage stats (psize, vsize). If N=0, no sampling
-is done and psize and vsize are returned as 0.
-.TP
-\fBMaxSendRetries=N\fR, where N is the number of times to try to send
-an accounting message before giving up.
-.TP
-\fBMaxSendRetryDelay=N\fR, pause for 1 to MaxSendRetryDelay
-seconds between attempts to deliver an
-accounting message.
-.TP
-\fBStaggerSlotSize=N\fR - For a process that might be sending
-a message at the same time as N other processes in the job, where N is
-10 or greater, each process will pause a bit before trying to send its
-message. For N tasks, N "staggered timeslots" are defined, in increments
-of (StaggerSlotSize*.001) seconds. The first process sends its message
-immediately, the second process pauses one increment before sending,
-the third process pauses two increments before sending, and so on.
-.RE
+For jobacct/linux the parameter is a number is seconds between polls.
.TP
\fBJobCompLoc\fR
The interpretation of this value depends upon the logging mechanism
diff --git a/etc/slurm.conf.example b/etc/slurm.conf.example
index 5643a02dd3c..0488cfa5d43 100644
--- a/etc/slurm.conf.example
+++ b/etc/slurm.conf.example
@@ -270,9 +270,9 @@
#
# o Define the job accounting mechanism to use.
#
-# "jobacct/linux" : Instigates a polling thread to gather information
+# "jobacct/linux" : Job accouting information
# from the linux proc table
-# "jobacct/none" : Do not have polling thread.
+# "jobacct/none" : No job accouting information.
#
JobAcctType=jobacct/none
@@ -285,9 +285,8 @@ JobAcctType=jobacct/none
#
# o Define the log file for job accounting this will be written on the
# same node the slurmctld is being ran on.
-# This log is kept even if JobAcctType=jobacct/none
#
-JobAcctLogFile=/var/log/slurm_accounting.log
+#JobAcctLogFile=/var/log/slurm_jobacct.log
#
# o Define the places to look for SLURM plugins. This is a
diff --git a/src/common/slurm_jobacct.h b/src/common/slurm_jobacct.h
index f9fb83f7183..a04c6986c3e 100644
--- a/src/common/slurm_jobacct.h
+++ b/src/common/slurm_jobacct.h
@@ -55,7 +55,7 @@
typedef struct slurm_jobacct_context * slurm_jobacct_context_t;
-/* common for both slurmctld and slurmstepd */
+/* common */
extern int jobacct_g_init_struct(jobacctinfo_t *jobacct, uint16_t tid);
extern jobacctinfo_t *jobacct_g_alloc(uint16_t tid);
extern void jobacct_g_free(jobacctinfo_t *jobacct);
diff --git a/src/plugins/jobacct/common/common_slurmstepd.c b/src/plugins/jobacct/common/common_slurmstepd.c
index 9d4655dc72b..62184d75392 100644
--- a/src/plugins/jobacct/common/common_slurmstepd.c
+++ b/src/plugins/jobacct/common/common_slurmstepd.c
@@ -37,7 +37,7 @@ pthread_mutex_t jobacct_lock = PTHREAD_MUTEX_INITIALIZER;
extern int common_endpoll()
{
fini = true;
-
+
return SLURM_SUCCESS;
}
extern int common_add_task(pid_t pid, uint16_t tid)
diff --git a/src/plugins/jobacct/common/jobacct_common.c b/src/plugins/jobacct/common/jobacct_common.c
index 29d79a37786..af562e86f52 100644
--- a/src/plugins/jobacct/common/jobacct_common.c
+++ b/src/plugins/jobacct/common/jobacct_common.c
@@ -239,7 +239,15 @@ extern void common_aggregate(struct jobacctinfo *dest,
}
dest->tot_pages += from->tot_pages;
-
+ if((dest->min_cpu > from->min_cpu)
+ || (dest->min_cpu == (uint32_t)NO_VAL)) {
+ if(from->min_cpu == (uint32_t)NO_VAL)
+ from->min_cpu = 0;
+ dest->min_cpu = from->min_cpu;
+ dest->min_cpu_task = from->min_cpu_task;
+ }
+ dest->tot_cpu += from->tot_cpu;
+
if(dest->max_vsize_task == (uint16_t)NO_VAL)
dest->max_vsize_task = from->max_vsize_task;
@@ -266,15 +274,6 @@ extern void common_aggregate(struct jobacctinfo *dest,
dest->rusage.ru_stime.tv_usec -= 1E6;
}
- if((dest->min_cpu > from->min_cpu)
- || (dest->min_cpu == (uint32_t)NO_VAL)) {
- if(from->min_cpu == (uint32_t)NO_VAL)
- from->min_cpu = 0;
- dest->min_cpu = from->min_cpu;
- dest->min_cpu_task = from->min_cpu_task;
- }
- dest->tot_cpu += from->tot_cpu;
-
dest->rusage.ru_maxrss += from->rusage.ru_maxrss;
dest->rusage.ru_ixrss += from->rusage.ru_ixrss;
dest->rusage.ru_idrss += from->rusage.ru_idrss;
diff --git a/src/sacct/print.c b/src/sacct/print.c
index bda72562396..be806f36ffa 100644
--- a/src/sacct/print.c
+++ b/src/sacct/print.c
@@ -27,49 +27,40 @@
\*****************************************************************************/
#include "sacct.h"
+#define FORMAT_STRING_SIZE 32
char *_decode_status(int status);
char *_elapsed_time(long secs, long usecs);
char *_elapsed_time(long secs, long usecs)
{
- int days, hours, minutes, seconds;
- char daybuf[10],
- hourbuf[4],
- minbuf[4];
- static char outbuf[20]; /* this holds LOTS of time! */
- div_t res;
-
- daybuf[0] = 0;
- hourbuf[0] = 0;
- minbuf[0] = 0;
-
- res = div(usecs+5000, 1e6); /* round up the usecs, then */
- usecs /= 1e4; /* truncate to .00's */
-
- res = div(secs+res.quot, 60*60*24); /* 1 day is 24 hours of 60
- minutes of 60 seconds */
- days = res.quot;
- res = div(res.rem, 60*60);
- hours = res.quot;
- res = div(res.rem, 60);
- minutes = res.quot;
- seconds = res.rem;
- if (days) {
- snprintf(daybuf, sizeof(daybuf), "%d-", days);
- snprintf(hourbuf, sizeof(hourbuf), "%02d:", hours);
- } else if (hours)
- snprintf(hourbuf, sizeof(hourbuf), "%2d:", hours);
- if (days || hours)
- snprintf(minbuf, sizeof(minbuf), "%02d:", minutes);
- else if (minutes)
- snprintf(minbuf, sizeof(minbuf), "%2d:", minutes);
- if (days || hours || minutes)
- snprintf(outbuf, sizeof(outbuf), "%s%s%s%02d.%02ld",
- daybuf, hourbuf, minbuf, seconds, usecs);
+ static char str[FORMAT_STRING_SIZE];
+ long days, hours, minutes, seconds;
+
+ while (usecs >= 1E6) {
+ secs++;
+ usecs -= 1E6;
+ }
+
+ seconds = secs % 60;
+ minutes = (secs / 60) % 60;
+ hours = (secs / 3600) % 24;
+ days = secs / 86400;
+
+ if (days)
+ snprintf(str, FORMAT_STRING_SIZE,
+ "%ld-%2.2ld:%2.2ld:%2.2ld",
+ days, hours, minutes, seconds);
+ else if (hours)
+ snprintf(str, FORMAT_STRING_SIZE,
+ "%ld:%2.2ld:%2.2ld",
+ hours, minutes, seconds);
else
- snprintf(outbuf, sizeof(outbuf), "%2d.%02ld", seconds, usecs);
- return(outbuf);
+ snprintf(str, FORMAT_STRING_SIZE,
+ "%ld:%2.2ld",
+ minutes, seconds);
+
+ return str;
}
void print_fields(type_t type, void *object)
@@ -701,10 +692,10 @@ void print_rss(type_t type, void *object)
switch(type) {
case HEADLINE:
- printf("%8s", "Rss");
+ printf("%22s", "MAX_RSS/Task - AVE");
break;
case UNDERSCORE:
- printf("%8s", "------");
+ printf("%22s", "----------------------");
break;
case JOB:
printf("%8ld", job->rusage.ru_maxrss);
@@ -887,16 +878,18 @@ void print_cputime(type_t type, void *object)
switch(type) {
case HEADLINE:
- printf("%10s", "SystemTime");
+ printf("%15s", "SystemTime");
break;
case UNDERSCORE:
- printf("%10s", "------");
+ printf("%15s", "----------");
break;
case JOB:
- printf("%10.2f", job->sacct.min_cpu);
+ printf("%15s",
+ _elapsed_time((int)job->sacct.min_cpu, 0));
break;
case JOBSTEP:
- printf("%10.2f", step->sacct.min_cpu);
+ printf("%15s",
+ _elapsed_time((int)step->sacct.min_cpu, 0));
break;
}
}
--
GitLab