[lustre-discuss] [EXTERNAL] [BULK] IOPS monitoring

[Andreas Dilger]

Using "lljobstat" is potentially even better than llstat, if you have JobStats enabled.

As mentioned in the other email, lustrefs-exporter + Grafana is best for large-scale monitoring, vs. node-local debugging.

Cheers, Andreas

On Apr 13, 2026, at 11:31, Vicker, Darby J. (JSC-EG311)[AMENTUM TECHNOLOGY, INC] via lustre-discuss <lustre-discuss at lists.lustre.org> wrote:
> 
> The llstat executable that is part of lustre is my go-to.  Running "llstat -i 1 mdt” on and MDS will give you a pretty good feel for what’s happening on at given MDS.  It’s useful to part these and send to a graphana instance for history and graphing.  
> 
> From: lustre-discuss <lustre-discuss-bounces at lists.lustre.org> on behalf of Robert Pennington via lustre-discuss <lustre-discuss at lists.lustre.org>
> Date: Monday, April 13, 2026 at 4:14 AM
> To: lustre-discuss at lists.lustre.org <lustre-discuss at lists.lustre.org>
> Subject: [EXTERNAL] [BULK] [lustre-discuss] IOPS monitoring
> 
> CAUTION: This email originated from outside of NASA.  Please take care when clicking links or opening attachments.  Use the "Report Message" button to report suspicious messages to the NASA SOC.
> 
> 
> 
>> Hello all,
>> 
>> What’s the best way to monitor Lustre IOPS? I can’t seem to find an up-to-date solution.
>> 
>> Sincerely,
>> Robert Pennington, PhD
>> 
>> HPC/AI Cluster Administrator
>> 
>> Tuebingen AI Center, Universitaet Tuebingen
>> Maria von Linden Str. 1
>> 72076 Tuebingen
>> Germany 
>> 
>> Office number: A-316
> 
> _______________________________________________
> lustre-discuss mailing list
> lustre-discuss at lists.lustre.org
> http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org

---
Andreas Dilger
Principal Lustre Architect
adilger at thelustrecollective.com