GlusterFs - high load 90-107% CPU

Posted by Sara on Server Fault See other posts from Server Fault or by Sara
Published on 2013-02-19T05:17:34Z Indexed on 2013/11/13 16:00 UTC
Read the original article Hit count: 435

Filed under:

Performance

|

cpu

|

load

|

high-load

|

glusterfs

I try and try and try to performance and fix problem with gluster, i try all. I served on gluster webpages, php files, images etc. I have problem after update from 3.3.0 to 3.3.1. I try 3.4 when i think maybe fix it but still the same problem. I temporarily have 1 brick, but before upgrade will be fine.

Config:
Volume Name: ...
Type: Replicate
Volume ID: ...
Status: Started
Number of Bricks: 0 x 2 = 1
Transport-type: tcp
Bricks:
Brick1: ...:/...
Options Reconfigured:
cluster.stripe-block-size: 128KB
performance.cache-max-file-size: 100MB
performance.flush-behind: on
performance.io-thread-count: 16
performance.cache-size: 256MB
auth.allow: ...
performance.cache-refresh-timeout: 5
performance.write-behind-window-size: 1024MB

I use fuse, hmm "Maybe the high load is due to the unavailable brick" i think about it, but i cant find information on how to safely change type of volume. Maybe u know how?

© Server Fault or respective owner

Related posts about Performance

Improving VPN performance - stronger encryption = more performance?

as seen on Server Fault - Search for 'Server Fault'
I have a site-to-site VPN set up with two SonicWall's (a TZ170 and a Pro1260). It was suggested to me that turning off encryption (so the VPN is tunneling only) would improve performance. (I'm not concerned with security, because the VPN is running over a trusted line.) Using FTP and HTTP transfers… >>> More
Inaccurate performance counter timer values in Windows Performance Monitor

as seen on Stack Overflow - Search for 'Stack Overflow'
I am implementing instrumentation within an application and have encountered an issue where the value that is displayed in Windows Performance Monitor from a PerformanceCounter is incongruent with the value that is recorded. I am using a Stopwatch to record the duration of a method execution, then… >>> More
Excel-based Performance Reviews transformed into Web Application for Performance Management

as seen on Geeks with Blogs - Search for 'Geeks with Blogs'
HR TMS provides enterprise talent management solutions for healthcare, retail and corporate customers, focusing on performance management, compensation management and succession planning. As the competency of nurses and other healthcare workers is critical, the government, via the Joint Commission… >>> More
How to save a perfmon Performance Counter as a textfile (Reliability and Performance Monitor Version

as seen on Server Fault - Search for 'Server Fault'
Now the file gets saved as blg, but I would like a txt versin to import in Excel. >>> More
SQLAuthority News – A Successful Performance Tuning Seminar at Pune – Dec 4-5, 2010

as seen on SQL Authority - Search for 'SQL Authority'
This is report to my third of very successful seminar event on SQL Server Performance Tuning. SQL Server Performance Tuning Seminar in Colombo was oversubscribed with total of 35 attendees. You can read the details over here SQLAuthority News – SQL Server Performance Optimizations Seminar – Grand… >>> More

Related posts about cpu

Speedstep and Intel x5570?

as seen on Server Fault - Search for 'Server Fault'
Hi, My new server has 2 x X5570 CPUs. Now here is the output of grep -i hz /proc/cpuinfo model name : Intel(R) Xeon(R) CPU X5570 @ 2.93GHz cpu MHz : 1600.231 model name : Intel(R) Xeon(R) CPU X5570 @ 2.93GHz cpu MHz : 1600.231 model name : Intel(R) Xeon(R) CPU … >>> More
VMware ESXi - varying CPU time (CPU reservation)

as seen on Server Fault - Search for 'Server Fault'
Hello! I'm running FreeBSD 7.2 under VMware ESXi 3.5. Host has 2 physical CPUs and the BSD box is currently the only running VM. Only one virtual CPU is assigned to the VM. When measuring CPU time of a specific program, I get very different results from time to time. Processor usage is reported… >>> More
How to calculate CPU % based on raw CPU ticks in SNMP

as seen on Server Fault - Search for 'Server Fault'
According to http://net-snmp.sourceforge.net/docs/mibs/ucdavis.html#scalar_notcurrent ssCpuUser, ssCpuSystem, ssCpuIdle, etc are deprecated in favor of the raw variants (ssCpuRawUser, etc). The former values (which don't cover things like nice, wait, kernel, interrupt, etc) returned a percentage… >>> More
disparity between `top`'s given CPU % and process CPU usage total

as seen on Super User - Search for 'Super User'
I've noticed that there are sometimes (large) differences between the reported total CPU usage and a summation of the per-process CPU utilization given by apps like top and wmtop. As an example: I recently ran a git filter-branch --index-filter on a fairly large repo, with the index-filter command… >>> More
How to read oom-killer syslog messages?

as seen on Server Fault - Search for 'Server Fault'
I have a Ubuntu 12.04 server which sometimes dies completely - no SSH, no ping, nothing until it is physically rebooted. After the reboot, I see in syslog that the oom-killer killed, well, pretty much everything. There's a lot of detailed memory usage information in them. How do I read these logs… >>> More