Search results: Spectrum LSF

Showing 15

Cleanup user details with bqueues -lr output if they don't have running/done jobs in system

Currently, we can find thousands of entries for all users even though they have ZERO STARTED jobs as listed in bqueues -lr output. For some of larger farms, these entries are close to 7M records and bqueues -lr takes long time to return the data. ...

over 2 years ago in Spectrum LSF / Scheduling 3 Delivered

4 VOTE

Prevent LSF from dispatching jobs to problematic hosts consistently

Some jobs are dispatched to a host and failed because the communication between MBATCHD and SBATCHD is failed. The host keeps in OK status and the jobs keep dispatching to the host, which makes the job can not run for a long time. The communicatio...

almost 10 years ago in Spectrum LSF 2 Delivered

2 VOTE

Relax GPU affinity while maintaining Core affinity

Prior to the introduction of the LSB_GPU_NEW_SYNTAX=extend it was possible to have jobs using core affinity and be assigned to cores on a cpu that is not directly connected to the gpu that was assigned to the job. After the introduction it will al...

almost 7 years ago in Spectrum LSF 3 Delivered

2 VOTE

Show specified fields value in bhosts

Here is an example of bhosts -l. bhosts -lHOST pkgpu0049STATUS CPUF JL/U MAX NJOBS RUN SSUSP USUSP RSV DISPATCH_WINDOWclosed_Full 40.00 - 24 24 24 0 0 0 - CURRENT LOAD USED FOR SCHEDULING: r15s r1m r15m ut pg io ls it tmp swp mem slots ngpus Total...

over 7 years ago in Spectrum LSF 1 Delivered

1 VOTE

cores limitations using affinity

In our cluster, we are very often impacted by jobs under reserving cpus (using -n option of bsub). for ex, we have a user group using a GUI for days. When the user close the GUI, it fires thousands of tar/bzip2 -9 to compress log files, making exe...

almost 4 years ago in Spectrum LSF 1 Delivered

2 VOTE

LSF jobs timeout - wall time limit

We are noticing an issue with some LSF jobs timeouts. For example user has submitted a job with 2 hr wall time limit but the compute host where the job is running is flagged as unhealthy by the in-house compute health check script and the host is ...

almost 8 years ago in Spectrum LSF 0 Delivered

2 VOTE

RTM: host group graphs inconsistencies

looking host group graphs (slots utilization for ex), there no way to monitor slots/jobs usable (ie slots reachable by a bsub) in a hostgroup, only "installed" slots are monitored by a max line, not the number minus the closed_Adm, unlicensed, ......

over 9 years ago in Spectrum LSF 1 Delivered

1 VOTE

Increase LSF Explorer auto log out time value

We are using LSF Explorer 10.2.0.9, while running queries for data older than 3-4 weeks which takes more time LSF Explorer auto logs out(300 seconds) while the query is still running in the background. In TS003951590 the concerned IBM support pers...

almost 5 years ago in Spectrum LSF / Reporting/Explorer 0 Delivered

2 VOTE

Need bhosts show extra information about the LSF host status even the host is returned back to EGO

When a host is in closed_EGO or unreach state, how can the admin know this host has ever been closed by LSF(badmin hclose) before EGO takes back the host or host rejoin after the host removal out of cluster. It's true the admin can check the lsb e...

almost 11 years ago in Spectrum LSF 2 Delivered

1 VOTE

If you open/tail one file, you cannot interact with the webpage until you close the file.

With version 10.xx.xx Spectrum LSF Application Center, if you open/tail one file (in the job details, you cannot interact with the webpage until you close the file and therefore cannot open more than 1 file; while with PAC 9.1.5, users were able t...

almost 6 years ago in Spectrum LSF 0 Delivered

Please enter your email address

FILTER BY CATEGORY