Description of problem: $ condor_q -h Usage: condor_q [options] where [options] are -global Get global queue -submitter <submitter> Get queue of specific submitter -run Get information about running jobs -hold Get information about jobs on hold -dag Sort DAG jobs under their DAGMan -expert Display shorter error messages -avgqueuetime Average queue time for uncompleted jobs $ condor_q -global .... 20 jobs; 20 idle, 0 running, 0 held $ condor_q -submitter condor .... 20 jobs; 20 idle, 0 running, 0 held $ condor_q -run ---->....(without information: 20 jobs; 20 idle, 0 running, 0 held) <----- $ condor_q -hold .... ----> 0 jobs; 0 idle, 0 running, 0 held <---- wrong numbers $ condor_q -dag .... 20 jobs; 20 idle, 0 running, 0 held $ condor_q -expert .... 20 jobs; 20 idle, 0 running, 0 held $ condor_q -avgqueuetime ----> Error: unrecognized argument -avgqueuetime <---- parameter from help, which doesn't work ... help message Version-Release number of selected component (if applicable): How reproducible: 100% Actual results: There are some dissimilar outputs from condor_q command with parameters. Expected results: There will not be -avgqueuetime in parameters or the functionality of this parameter will be in condor_q. All other outputs from condor_q with above parameters will output correct information about jobs on last line. Additional info:
Three issues here - Re -avgqueuetime: https://condor-wiki.cs.wisc.edu/index.cgi/tktview?tn=1629 Re -run: definitely unfortunate Re -hold: probably a bug
Re -avgqueuetime, commit 914fe998bba1d9dc1d9e6813b9b98e8c6bea294c Author: Matthew Farrellee <matt@redhat> Date: Wed Jan 12 09:04:55 2011 -0500 Removed -avgqueuetime from condor_q -h when built without postgresql, #1629 diff --git a/src/condor_q.V6/queue.cpp b/src/condor_q.V6/queue.cpp index 91054a0..85d9342 100644 --- a/src/condor_q.V6/queue.cpp +++ b/src/condor_q.V6/queue.cpp @@ -1854,10 +1854,10 @@ usage (char *myName) "\t\t\tPerform a direct query to the rdbms\n" "\t\t\tor to the schedd without falling back to the queue\n" "\t\t\tlocation discovery algortihm, even in case of error\n" + "\t\t-avgqueuetime\t\tAverage queue time for uncompleted jobs\n" #else "\t\t-direct <schedd>\tPerform a direct query to the schedd\n" #endif - "\t\t-avgqueuetime\t\tAverage queue time for uncompleted jobs\n" "\t\t-version\t\tPrint the Condor Version and exit\n" "\t\trestriction list\n" "\twhere each restriction may be one of\n" -- Available for 7.5.6.
Re -run, commit 9cc888274722aeda9935615dcd93107b84be4f2e Author: jbasney <jbasney> Date: Thu Jan 21 18:49:56 1999 +0000 add -run option, which only displays running jobs (unless another constraint is given) and shows the remote host on which each job is running ... @@ -527,6 +577,27 @@ show_queue( char* scheddAddr, char* scheddName, char* scheddMachine ) if( verbose ) { jobs.fPrintAttrListList( stdout ); + } else if ( run ) { + summarize = false; + AttrListPrintMask mask; + printf( " %-7s %-14s %11s %12s %-16s\n", "ID", "OWNER", + "SUBMITTED", "CPU_USAGE", ... The lack of summary is intentional.
Re -hold, ommit 2128193f894e47e903b5b53ac6e497e0ccca315c Author: tannenba <tannenba> Date: Tue Sep 24 22:00:40 2002 +0000 added "-hold" (or -held) option to condor_q. derek: this was a bug. really. honest. ;^) p.s. derek: don't hurt me. i am weak. i need am sick. ... + } else if ( show_held ) { + if (!setup_mask) { + mask.registerFormat ("%4d.", ATTR_CLUSTER_ID); + mask.registerFormat ("%-3d ", ATTR_PROC_ID); + mask.registerFormat ( (StringCustomFmt) format_owner, + ATTR_OWNER, "[????????????] " ); + mask.registerFormat(" ", "*bogus*", " "); // force space + mask.registerFormat ( (IntCustomFmt) format_q_date, + ATTR_ENTERED_CURRENT_STATUS, "[?????????? + mask.registerFormat(" ", "*bogus*", " "); // force space + mask.registerFormat( "%-43.43s\n", ATTR_HOLD_REASON ); + setup_mask = true; + usingPrintMask = true; + } ... Unlike other options, this one relies on the default setting for "summarize".
Re -hold, I would expect that the summary line reports a summary of what was displayed. Since there are no held jobs, having "0 jobs; 0 idle, 0 running, 0 held" is reasonable output.
Summary - -avgqueuetime: fixed upstream for 7.5.6 -run: intentional, NOTABUG, reasonable RFE -hold: NOTABUG
Technical note added. If any revisions are required, please edit the "Technical Notes" field accordingly. All revisions will be proofread by the Engineering Content Services team. New Contents: C: condor_q compiled without postgresql suggests -avgqueuetime as an argument (from -help) C: condor_q ignores the argument, with an error F: condor_q -h updated to not report -avgqueuetime when compiled without postgresql R: No more confusion
Tested on RHEL6.1/5.6 x x86_64/i386 with condor-7.6.1-0.1 and it works. -->VERIFIED There is correct output of condor_q -avgqueuetime: Error: unrecognized argument -avgqueuetime
An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on therefore solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHEA-2011-0889.html