Bug 1610261

Summary: Update the options in ganesha configuration files
Product: [Red Hat Storage] Red Hat Gluster Storage Reporter: Jiffin <jthottan>
Component: nfs-ganeshaAssignee: Jiffin <jthottan>
Status: CLOSED ERRATA QA Contact: Manisha Saini <msaini>
Severity: medium Docs Contact:
Priority: medium    
Version: rhgs-3.4CC: amukherj, dang, ffilz, grajoria, jijoy, jthottan, kkeithle, mbenjamin, pasik, rcyriac, rhinduja, rhs-bugs, sanandpa, sheggodu, skoduri, storage-qa-internal
Target Milestone: ---Keywords: Triaged, ZStream
Target Release: RHGS 3.5.0   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: nfs-ganesha-2.7.3-3 Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2019-10-30 12:15:39 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 1696807    

Description Jiffin 2018-07-31 10:22:33 UTC
Description of problem:
There are three parameters in configuration files which is better to be changed 

change the following

i) up_poll_usec  in FSAL{} in export conf file -- current value default 10us, better to change into 1ms

Reason -- it impact cpu utilization of ganesha process, higher the interval lower cpu utiliization

Patch -- need to be sent in gluster codebase for creating export configuration file script file(Downstream only)

ii.) Entries_HWMark  in CACHEINODE{} in ganesha.conf -- current value 25k, need to change atleast to 100K

Reason -- it impact performance if the workload has more than 25k entries

Patch -- need to modify ganesha.conf next build

iii.) Chunks_HWMark in CACHEINODE{} in ganesha.conf -- current value is default 100K, need to reduced to 1000

Reason -- having more value than Entries_HWMark results more caching at dirent cache and if the entry is missed

in mdcache then it triggers lookup/getattrs for that entry 

Patch change in ganesh.conf next build

Version-Release number of selected component (if applicable):
3.4

Additional info:

Comment 2 Frank Filz 2018-07-31 13:56:00 UTC
I still disagree with setting Chunks_HWMark as low as 1000.

Comment 7 Kaleb KEITHLEY 2018-08-14 12:51:55 UTC
Jiffin, is there a Doc BZ to document the preferred values for these settings in rhgs-3.4?

Comment 21 Manisha Saini 2019-05-29 09:08:47 UTC
Verified this BZ with

# rpm -qa | grep ganesha
nfs-ganesha-2.7.3-3.el7rhgs.x86_64
glusterfs-ganesha-6.0-3.el7rhgs.x86_64
nfs-ganesha-debuginfo-2.7.3-3.el7rhgs.x86_64
nfs-ganesha-gluster-2.7.3-3.el7rhgs.x86_64


As discussed over IRC,this BZ is validated for only two params

 1. Entries_HWMark --> Later was decided to keep this value as 125K instead of 100K
 2. Chunks_HWMark


Changes made in ganesha.conf for the above 2 params are now in place.
========
CACHEINODE {
        Entries_HWMark = 125000;
        Chunk_HWMark = 1000;
}
========

"up_poll_usec" related param is not valid for 3.5.0 release.

Moving this BZ to verified state.

Comment 24 errata-xmlrpc 2019-10-30 12:15:39 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHEA-2019:3252

Comment 25 Red Hat Bugzilla 2023-09-14 04:32:25 UTC
The needinfo request[s] on this closed bug have been removed as they have been unresolved for 1000 days