[gpfsug-discuss] Checking for Stale File Handles

[Alexander John Mamach]

Hi Fred,

We sometimes find a node will show that GPFS is active when running mmgetstate, but one of our GPFS filesystems, (such as our home or projects filesystems) are inaccessible to users, while the other GPFS-mounted filesystems behave as expected. Our current node health checks don’t always detect this, especially when it’s for a resource-based mount that doesn’t impact the node but would impact jobs trying to run on the node.

If there is something native to GPFS that can detect this, all the better, but I’m simply unaware of how to do so.

Thanks,

Alex

Senior Systems Administrator

Research Computing Infrastructure
Northwestern University Information Technology (NUIT)

2020 Ridge Ave
Evanston, IL 60208-4311

O: (847) 491-2219
M: (312) 887-1881
www.it.northwestern.edu

________________________________
From: gpfsug-discuss-bounces at spectrumscale.org <gpfsug-discuss-bounces at spectrumscale.org> on behalf of Frederick Stock <stockf at us.ibm.com>
Sent: Friday, August 9, 2019 1:03:09 PM
To: gpfsug-discuss at spectrumscale.org <gpfsug-discuss at spectrumscale.org>
Cc: gpfsug-discuss at spectrumscale.org <gpfsug-discuss at spectrumscale.org>
Subject: Re: [gpfsug-discuss] Checking for Stale File Handles

Are you able to explain why you want to check for stale file handles?  Are you attempting to detect failures of some sort, and why do the existing mechanisms in GPFS not provide the functionality you require?

Fred
__________________________________________________
Fred Stock | IBM Pittsburgh Lab | 720-430-8821
stockf at us.ibm.com


----- Original message -----
From: Alexander John Mamach <alex.mamach at northwestern.edu>
Sent by: gpfsug-discuss-bounces at spectrumscale.org
To: "gpfsug-discuss at spectrumscale.org" <gpfsug-discuss at spectrumscale.org>
Cc:
Subject: [EXTERNAL] [gpfsug-discuss] Checking for Stale File Handles
Date: Fri, Aug 9, 2019 1:46 PM


Hi folks,



We’re currently investigating a way to check for stale file handles on the nodes across our cluster in a way that minimizes impact to the filesystem and performance.



Has anyone found a direct way of doing so? We considered a few methods, including simply attempting to ls a GPFS filesystem from each node, but that might have false positives, (detecting slowdowns as stale file handles), and could negatively impact performance with hundreds of nodes doing this simultaneously.



Thanks,



Alex



Senior Systems Administrator

Research Computing Infrastructure
Northwestern University Information Technology (NUIT)

2020 Ridge Ave
Evanston, IL 60208-4311

O: (847) 491-2219
M: (312) 887-1881
www.it.northwestern.edu



_______________________________________________
gpfsug-discuss mailing list
gpfsug-discuss at spectrumscale.org
http://gpfsug.org/mailman/listinfo/gpfsug-discuss<https://urldefense.proofpoint.com/v2/url?u=http-3A__gpfsug.org_mailman_listinfo_gpfsug-2Ddiscuss&d=DwMFaQ&c=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws&r=DWNlS6sDn1OrnGOrEhXtg2NFm8p2BHsvrT1P2_F36Mg&m=OsNhvPWE6Qrbzn5DFPdDOj4Rx4ujTI8EnYl4t9v-KXQ&s=fMvliztGa5f_ZKOREibEaRgtHOw_jb3qvBlVVfQj0SY&e=>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://gpfsug.org/pipermail/gpfsug-discuss_gpfsug.org/attachments/20190809/570d19c3/attachment-0002.htm>