Skip to main content
Share Your Experience: Take the 2024 Developer Survey
15 votes

At what point is a server considered idle?

A machine is idle when it's not performing the task it's supposed to perform for lack of requests. For example, if you had an email server, you could determine if it was fielding any requests from the ...
Spooler's user avatar
  • 7,136
9 votes
Accepted

Minimal upgrade from one minor CentOS version to the next

No. Not really. If you're running the upgrade, that means new GLIBC and kernel and supporting packages will be downloaded and installed. In addition, any updates to the packages already present on ...
ewwhite's user avatar
  • 198k
8 votes

At what point is a server considered idle?

"Idle" is not a black and white concept: even a busy server can have free CPU cycles, IOPs and memory to run some other application. As a rule of thumb, target CPU load should be at around ~80% and ...
shodanshok's user avatar
  • 50.9k
6 votes
Accepted

LVM cache in writeback mode is equivalent to external ext4 journal?

An external journal isn't the same thing as an LVM cache device at all. An LVM cache on an SSD for writeback wouldn't be volatile, so the concerns with data integrity aren't enormous (except for if ...
Spooler's user avatar
  • 7,136
5 votes
Accepted

Optimizing nginx client_body_buffer_size for web app accepting file uploads

The documentation states what client_body_buffer_size does: Sets buffer size for reading client request body. In case the request body is larger than the buffer, the whole body or only its part is ...
Michael Hampton's user avatar
4 votes

XFS with No Barrier Mount option deprecated

Since kernel 4.13, the nobarrier was removed for XFS. This article suggests two workarounds: switch from XFS to ext4 echo "write through" > /sys/block/$device/queue/write_cache I tried myself the ...
synkro's user avatar
  • 233
3 votes
Accepted

Debugging and fine-tuning the linux process scheduler

Is it correct that the php-fpm process spent as much as 58.130 ms waiting for a CPU to run on? it's time from wake-up to context switch, so I'd say it's time taken to make it finally running (thus — "...
poige's user avatar
  • 9,595
3 votes

Joins performed without indexes: 5568

Joins performed without indexes: 91. This is most likely the biggest culprit for your performance. Your database doesn't have proper indexes for columns that are used in SQL queries that join data ...
Tero Kilkanen's user avatar
3 votes
Accepted

Should a KVM host be doing things other than hosting VMs? (such as ZFS)?

Hypervisors, regardless of the flavor, should be 100% dedicated to serving its VMs. This is for a few reasons. First, one of the key benefits of virtualization is hardware independence - if a host ...
EEAA's user avatar
  • 110k
3 votes
Accepted

winbind authentication through apache slow

I still didn't find how to improve the performance when using mod_authnz_external. However changing to another apache module mod_authnz_ldap got rid of the performance problem. The only downside then ...
Zitrax's user avatar
  • 804
3 votes
Accepted

Speed up backup of just ACLs on large file systems

In my experience, getfacl can be CPU bound by the username-resolving process. Try adding the -n switch, for example issuing getfacl -n -R -p /backup/dir > out_file During benchmarks, pay special ...
shodanshok's user avatar
  • 50.9k
3 votes
Accepted

Very slow write speed (NVME drive) on 10G network

This answer was inspired by comment by @shodanshok, who commented (so I can't upvote his contribution -- posting an answer instead) Edit 2021/06/09 - iperf3 developers identified a possible problem; ...
roman's user avatar
  • 191
3 votes
Accepted

NGINX Redirect to Cached File in Subfolder

You can use a regular expression location to extract the first and second parts of the URI. Use a try_files statement to search the file system for each file in order. For example: location ~ ^(/...
Richard Smith's user avatar
2 votes

Minimal upgrade from one minor CentOS version to the next

Look at the list of packages it wants to update. Remove any that you don't use and aren't dependencies of something you do use. Retry until satisfied with the download size, then deploy the resulting ...
Warren Young's user avatar
2 votes

nginx: configuration file test failed

/proc/cpuinfo shows information about processors on your system. It has nothing to do with nginx worker processes. Your problems are most likely caused by your application server not being able to ...
Tero Kilkanen's user avatar
2 votes
Accepted

How to choose database and backend server locations?

Physical location does not necessarily correlate with "logical" network location, so you'll need to do more testing and research. You want your application to be as "close" (in terms of network ...
EEAA's user avatar
  • 110k
2 votes
Accepted

Tune when sssd goes into offline mode

reconnection_retries (integer) Number of times services should attempt to reconnect in the event of a Data Provider crash or restart before they give up Default: 3 Reduce to 1 and look at the ...
Sergey Ovsienko's user avatar
2 votes

mysqld taking 800% CPU since traffic increased on my site, how to fix

Suggestions to consider for your my.cnf-ini [mysqld] section with current information analyzed. key_cache_age_threshold=64800 # from 300 second discard, only to RD again innodb_lru_scan_depth=100 # ...
Wilson Hauck's user avatar
2 votes

mysqld taking 800% CPU since traffic increased on my site, how to fix

Suggestions to consider for your my.cfg-ini [mysqld] section max_connections=400 # from 2000 - max_used_connections was 109 in 2 hours thread_cache_size=100 # from autocalc CAP at 100 to avoid OOM &...
Wilson Hauck's user avatar
2 votes

Joins performed without indexes: 5568

There are at least two issues with your configuration: Overallocated max_connections. Your highest connection usage is 4 and you allocated 440 connections which caused memory overconsumption (each ...
AlexD's user avatar
  • 10.4k
2 votes
Accepted

Nginx auto minify html output

Google Pagespeed for Nginx does minification, and a bunch of other things. However when I benchmarked it didn't make enough of a difference to bother with, given my site was already well optimized. ...
Tim's user avatar
  • 33k
2 votes

Fastest possible configuration for a local MySQL database, no security needed?

It depends on how much RAM you have. query_cache_size is too big. Don't allocate more than 50M. In most servers it is better to turn it off. key_buffer_size = 1332M innodb_buffer_pool_size = 27M ...
Rick James's user avatar
  • 2,618
2 votes

Very slow write speed (NVME drive) on 10G network

Using my 2 servers connected over 10Gbit ethernet and your way of testing shows that the NVMe drives I use are slower than I expected when using iperf3. I think you might see the disks are actually ...
Marco's user avatar
  • 307
2 votes
Accepted

Where is my bottleneck: RAM, CPU or disk?

RAM usage is at ~73% with almost no swap space being used, and all your cores appear to be maxed out. So RAM size and speed are fine, and the bottleneck is certainly the CPU. Based off your ...
slightly_toasted's user avatar
1 vote

Fastest possible configuration for a local MySQL database, no security needed?

Most of the settings you have in there are either pointless or downright counter-productive. Remove them all. Make sure you are only using InnoDB. Adjust innodb_buffer_pool size to 50-80% of your RAM ...
Gordan Bobić's user avatar
1 vote
Accepted

Fastest possible configuration for a local MySQL database, no security needed?

There are a bunch of scripts that connects to your MySQL server, analyzes some variables and performance statistics of your running system and proposes configurations changes. The most famous are ...
JucaPirama's user avatar
1 vote
Accepted

How to convert swap partition from RAID1 to RAID0 using mdadm

Please find below a simple way to achieve the requested.I did not really need help. Mostly sharing to whoever it might help. To start you may want to check the details of /dev/md/s sudo mdadm --...
dankilev's user avatar
  • 166
1 vote

Joins performed without indexes: 5568

Sorry about the unfitting first answer - I obviously was too tired yesterday and misunderstood your question. While using connection pools would help with the overall performance of the application, ...
Phillip -Zyan K Lee- Stockmann's user avatar
1 vote

At what point is a server considered idle?

What operating system is it and what is/was the server's purpose? If I had no other information and I needed to determine is a server was idle I would check to see which services were installed/...
apocalysque's user avatar
1 vote

How to speed up connection with a server in China

The "Great Firewall" is filtering all connections into and out of China. Generally, all connections suffer, there's a lot of packet loss as well. As long as the Chinese don't change this (highly ...
Zac67's user avatar
  • 11.9k

Only top scored, non community-wiki answers of a minimum length are eligible