CONDOR_USERLOG(1)

📦
version 25.14.1 · 12 versions
, in fedora 43

👑 CONDOR_USERLOG 👑

okay bestie, welcome to the HTCondor Manual — this is the page for condor_userlog, and it is genuinely iconic.

✨ NAME ✨

condor_userlog — the way it displays and summarizes job statistics pulled straight out of your job log files. that's the whole vibe. it reads the logs, it gives you the receipts. 🧾✨

🌸 SYNOPSIS 🌸

condor_userlog [-help ] [-total | -raw ] [-debug ] [-evict ] [-j cluster | cluster.proc] [-all ] [-hostname ] logfile ...
heads up, bestie: this page is long and it comes in parts, so don't panic when the synopsis keeps unfolding. we're just getting started 💖

✨ DESCRIPTION ✨

okay so here's the tea, bestie 🫖 — condor_userlog parses the info sitting inside job log files and shows you summaries, like, per workstation allocation and per job. If you're wondering how to even get HTCondor writing a log file for your jobs in the first place, that's a whole other manual page — check condor_submit. 💅

🌸 the per-allocation record (when -total isn't in the chat)

If -total is not specified, condor_userlog will first display a record for each workstation allocation, and that record is giving:

  • Job — the cluster/process id of the HTCondor job. 👑
  • Host — the host where the job ran. By default, it's the host's IP address that gets displayed, but if you pass -hostname, then the host name shows up instead. ✨
  • Start Time — the time (month/day hour:minute) when the job began running on the host. ⏰
  • Evict Time — the time (month/day hour:minute) when the job was evicted from the host. 😭
  • Wall Time — the time (days+hours:minutes) for which this workstation was allocated to the job. 🧱
  • Good Time — the allocated time (days+hours:min) which contributed to the completion of this job. And listen: if the job exited during the allocation, then this value will equal Wall Time. Otherwise, it will be 0+00:00; self-checkpoints are presently ignored. 💀
  • CPU Usage — the CPU time (days+hours:min) which contributed to the completion of this job. 🔥

🦋 then the summary per host

After that, condor_userlog displays summary statistics per host, and those go a little something like:

  • Host/Job — the IP address or host name for the host. 🎀
  • Wall Time — the workstation time (days+hours:minutes) allocated by this host to the jobs specified in the query. By default, all jobs in the log are included in the query. 💖
  • Good Time — the time (days+hours:minutes) allocated on this host which contributed to the completion of the jobs specified in the query. ✨
  • CPU Usage — the CPU time (days+hours:minutes) obtained from this host which contributed to the completion of the jobs specified in the query. 🔥
  • Avg Alloc — the average length of an allocation on this host (days+hours:minutes). 📏
  • Avg Lost — the average amount of work lost (days+hours:minutes) when a job was evicted from this host. 😭
  • Goodput — this percentage is computed as Good Time divided by Wall Time. 🧮
  • Util. — this percentage is computed as CPU Usage divided by Good Time. 🧮

🌈 then the summary per job

Next up, condor_userlog displays summary statistics per job, and she's serving:

  • Host/Job — the cluster/process id of the HTCondor job. 👑
  • Wall Time — the total workstation time (days+hours:minutes) allocated to this job. 🧱
  • Good Time — the total time (days+hours:minutes) allocated to this job which contributed to the job's completion. ✨
  • CPU Usage — the total CPU time (days+hours:minutes) which contributed to this job's completion. 🔥
  • Avg Alloc — the average length of a workstation allocation obtained by this job (days+hours:minutes). 📏
  • Avg Lost — the average amount of work lost (days+hours:minutes) when this job was evicted from a host; self-checkpoints are presently ignored. 💀
  • Goodput — this percentage is computed as Good Time divided by Wall Time. 🧮
  • Util. — this percentage is computed as CPU Usage divided by Good Time. 🧮

And finally? condor_userlog displays a summary for all hosts and jobs. the end, iconic behavior, no notes 💅🫶

✨ OPTIONS ✨

okay bestie here's the full menu of flags, no skips, we're eating all of them 💅

  • -help — throws you a lil quick rundown of all the supported options. like a vibe check for your brain 🧠
  • -total — shows you only the job totals. that's it. just the receipts, nothing else 🧾
  • -raw — displays raw data only. no garnish, no glow-up, straight off the grill 🔥
  • -debug — debug mode. for when things are giving mystery and you need the tea ☕👀
  • -j — selects a specific cluster or a cluster.proc. pick your fave, we're being specific 🎯
  • -evict — selects only the allocations that ended because of eviction. yeah. the sad ones. rest in peace bestie 😭👻
  • -all — selects all the clusters and all the allocations. no filter, no skips, maximum range 🌈✨
  • -hostname — displays the host name instead of the IP address. way prettier, way more readable, we love to see it 🏠💖

🌸 GENERAL REMARKS 🌸

lowkey important info, don't scroll past this one:

  • the HTCondor job log file format does not have a year field in the timestamp, so every single entry is assumed to happen in the current year. yeah. that's just how it is 🫠
  • so if an allocation begins in one year and ends in the next one? it gets silently ignored. no warning, no heads up, just gone. ghosted fr fr 👻 the silence is the message 😭

💀 EXIT STATUS 💀

  • 0 (zero) — slay! everything worked, we're thriving 👑✨
  • 1 (one) — nope, it failed. it's giving tragedy 🫶

🎀 AUTHOR 🎀

the HTCondor Team. legends only, no cap 💜

1990-2026, Center for High Throughput Computing, Computer Sciences Department, University of Wisconsin-Madison, Madison, WI, US. Licensed under the Apache License, Version 2.0.

and that's the whole page bestie, we ate and left no crumbs 🍓🧚 this was CONDOR_USERLOG(1) from Oct 16, 2026 💖 go forth and read your logs like the main character you are ✨💅🦋

📦 where this page comes from