Bug 825996

Summary: Changing the
Product: [Community] GlusterFS Reporter: Sabyasachi Ruj <ruj.sabya>
Component: io-threadsAssignee: Raghavendra G <rgowdapp>
Status: CLOSED NOTABUG QA Contact:
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 3.2.6CC: amarts, gluster-bugs
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2012-07-11 11:04:51 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:
Attachments:
Description Flags
glusterfsdump created using kill -USR1 none

Description Sabyasachi Ruj 2012-05-29 10:00:45 UTC 
Description of problem:

Here is my volume configuration:

  # gluster volume info testvol
 
  Volume Name: testvol
  Type: Distribute
  Status: Started
  Number of Bricks: 1
  Transport-type: tcp
  Bricks:
  Brick1: server1:/bricks/0001
  Options Reconfigured:
  performance.io-thread-count: 1
  performance.flush-behind: off
  performance.quick-read: off

Now, when I am doing file operations, I can see two glusterfsd threads are writing to the brick. First I found out this using `top -p $(pgrep glusterfsd) -H`, there it was clear that two thread were consuming CPU. Then I performed strace on those threads. I could see both of these threads were doing IO.


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

glusterfs 3.2.6 built on May 16 2012 13:20:35
Repository revision: git://git.gluster.com/glusterfs.git
Copyright (c) 2006-2011 Gluster Inc. <http://www.gluster.com>
GlusterFS comes with ABSOLUTELY NO WARRANTY.
You may redistribute copies of GlusterFS under the terms of the GNU General Public License.



How reproducible:

Steps to Reproduce:
1. Create the above volume configuration.
2. Mount that volume and do IO.
  
Actual results:


Expected results:


Additional info:

I have attached
Comment 1 Sabyasachi Ruj 2012-05-29 10:01:51 UTC 
Created attachment 587374 [details]
glusterfsdump created using kill -USR1
Comment 2 Sabyasachi Ruj 2012-05-29 11:14:23 UTC 
Hey Sorry. I guess I am mixing it up. One thread is doing I guess the network i/o and the other process is actually handling disk/io.

If this is the case, this bug is invalid. Let me confirm once.
Comment 3 Amar Tumballi 2012-07-11 11:04:51 UTC 
when you say 'io-thread-count 1', that means one extra thread will be created other than the main thread which is polling on network.