Server Fault Stack Exchange
Server Fault Stack Exchange

Questions tagged [high-load]

All aspects of high-loaded systems development and maintenance.

All aspects of high-loaded systems development and maintenance: design, algorithms, architecture, technologies, development, testing, databases, storage, support, hardware, hosting, management.

219 questions
41
votes
2 answers

How to reduce number of sockets in TIME_WAIT?

Ubuntu Server 10.04.1 x86 I've got a machine with a FCGI HTTP service behind nginx, that serves a lot of small HTTP requests to a lot of different clients. (About 230 requests per second in the peak hours, average response size with headers is 650…
Alexander Gladysh
  • 2,343
  • 7
  • 30
  • 47
27
votes
3 answers

How far is "too far off" for ntpd? Can it get there by a sudden jump to heavy load? Can this be overridden?

In many primers on ntpd, like this one, there's always a warning that ntpd will stop resetting the clock "[if] your clock is too far off." How far off is "too far off"? Also, if a server takes a sudden jump to heavy load, for example from total…
user144330
22
votes
6 answers

Why is response time exploding when request frequency drops?

Correction: response time (%D) is μs not ms! 1 This doesn't change anything about the weirdness of this pattern but it means that it is practically way less devastating. Why is response time inversely correlated to request frequency? Shouldn't the…
Raffael
  • 669
  • 7
  • 15
20
votes
6 answers

High Load Average with modest CPU Utilization and almost no IO

The usual explanation for high load average with little cpu use under linux is too much IO (or more properly uninterruptable sleep). I have a service running on a cluster of 2-core VMs that exhibit modest CPU use (~55-70% idle) but higher than 2…
dbenhur
  • 309
  • 1
  • 2
  • 9
18
votes
2 answers

Is it possible to use multiple load balancers to redirect traffic to my application servers?

I'm new to load balancing and I'm wondering if it's possible to use multiple load balancers to redirect traffic to my application servers. I don't really understand how this can be done. Shouldn't a domain name match one to one with a certain…
user3790827
  • 191
  • 1
  • 1
  • 6
18
votes
2 answers

High server load - [jbd2/md1-8] using 99.99% IO

I've been having spike in load over the last week. This usually occurs once or twice a day. I've managed to identify from iotop that [jbd2/md1-8] is using 99.99 % IO. During the high load times there is no high traffic to the server. Server specs…
Alex
  • 181
  • 1
  • 1
  • 3
18
votes
2 answers

HA Proxy - roundrobin vs leastconn

Are there any suggestion on when should I use roundrobin & when should I use leastconn? I am using roundrobin currently and observed the loading of my back-end servesr are not evenly distribution. Of course there might be other issue, but we want to…
Ryan
  • 5,341
  • 21
  • 71
  • 87
15
votes
2 answers

rsync ionice destination

To avoid rsync to starve the system and network we use ionice when starting rsync and set the --bwlimit parameter. For example: ionice -c2 -n7 rsync -aH --bwlimit=30000 /foo root@dest.com:/ This does indeed help to ensure the source server remains…
Jason Smith
  • 251
  • 2
  • 3
13
votes
4 answers

Why is my web server dropping connections with a TCP reset at high load?

I have a small VPS setup with nginx. I want to squeeze as much performance as possible out of it, so I've been experimenting with optimization and load testing. I'm using Blitz.io to do load testing by GETing a small static text file, and running…
Erik Swan
  • 249
  • 1
  • 2
  • 5
13
votes
4 answers

ps aux hanging on high cpu/IO with java processes

I'm having some issues with java process and nrpe checks. We have some processes that sometimes use 1000% cpu on a 32 core system. The system is pretty responsive until you do a ps aux or try to do anything in the /proc/pid#…
Mike
  • 21,910
  • 7
  • 55
  • 79
12
votes
3 answers

CPU0 is swamped with eth1 interrupts

I've got an Ubuntu VM, running inside Ubuntu-based Xen XCP. It hosts a custom FCGI-based HTTP service, behind nginx. Under load from ab the first CPU core is saturated, and the rest is under-loaded. In /proc/interrupts I see that CPU0 serves an…
Alexander Gladysh
  • 2,343
  • 7
  • 30
  • 47
12
votes
5 answers

High Linux loads on low CPU/memory usage

I have quite strange situation, where my CentOS 5.5 box loads are high, but the CPU and memory used are pretty low: top - 20:41:38 up 42 days, 6:14, 2 users, load average: 19.79, 21.25, 18.87 Tasks: 254 total, 1 running, 253 sleeping, 0…
SyRenity
  • 3,159
  • 11
  • 55
  • 79
12
votes
3 answers

The downsides of using nginx as a primary web server?

I've seen millions of websites using nginx as a proxifying webserver working together with Apache. But I've seen very few servers running nginx only as their default webserver. What are the main downsides of such config? I can see some: Inability…
Vladislav Rastrusny
  • 2,581
  • 12
  • 39
  • 56
9
votes
3 answers

Load avg weirdness on Linux Ubuntu

In past few days I've been trying to understand the weirdness which is happening in our infrastructure but I haven't been able to figure it our so I'm turning to you guys to give me some hints. I've been noticing in Graphite, spikes in load_avg…
milosgajdos
  • 1,808
  • 2
  • 21
  • 29
9
votes
4 answers

High load on a nagios server -- How many service checks for a nagios server is too many?

I have a nagios server running Ubuntu with a 2.0 GHz Intel Processor, a RAID10 array, and 400 MB of RAM. It monitors a total of 42 services across 8 hosts, most of which are checked using the check_http plugin even 5 minutes, some every minute.…
Josh
  • 9,001
  • 27
  • 78
  • 124
1
2 3
…
14 15