achille

/* 
 * Auto-tuning delete that allows for removal of large amounts of data 
 * without impacting performance. Configurable to a target load amount.
 *
 * How it works: 
 * TL;DR: Delete a small slice every second; Vary the size of each slice 
 *        based on how long the previous delete took; sleep; repeat. 
 *
 * TODO: Modify this to allow for deletion based on objectid's date 
 *       which is embedded in the first four bytes.

mlafeldt

Shakespeare Sonnet++ Postmortem (incident #465)

Date

2015-10-21

Authors

• jennifer
• martym

ogrisel

model name      : Intel(R) Xeon(R) CPU @ 2.30GHz                                                                                   
Loading arrays to memory                                                                                                           
Loading speed: 0.670GB/s                                                                                                           
                                                                                                                                   
timing bandwidth for sequential memory access                                                                                      
bandwidth: 9.0 GB/s                                                                                                                
                                                                                                                                   
n_workers=2 (threads)                                                                               

brendangregg

There are some quick benchmarks for the "Decade of Wasted Cores" patches on Linux 4.1. I had to add "extern int sched_max_numa_distance;" to arch/x86/kernel/smpboot.c for Linux 4.1 to compile. Brief analysis during the benchmarks using time(1) and mpstat(1) to check runtimes, usr/sys time, and per-CPU balance; iostat(1) to check for disk bottlenecks.

Summary: no significant difference seen in these tests.

c3.8xlarge (32 CPU) PV 1-node NUMA

The patch shouldn't make a difference to this 1-node system, but I felt it worth checking, especially since most of our systems are 1-node.

atcuno

The purpose of this document is to make recommendations on how to browse in a privacy and security conscious manner. This information is compiled from a number of sources, which are referenced throughout the document, as well as my own experiences with the described technologies.

I welcome contributions and comments on the information contained. Please see the How to Contribute section for information on contributing your own knowledge.

Table of Contents

• Motivation
• Audience
• TL;DR?
• Updates
• Technologies that Affect Security and Privacy

kiantabrizi

var dbNames = [];
var sum = 0;
var dblist = db.getMongo().getDBs();
for (var key in dblist) {
    if(key === "databases") {
        for (var i = 0; i < dblist.databases.length; i++){
            if (dblist.databases[i].name !== "local") {
                dbNames.push(dblist.databases[i].name);
            }
        }

lewisd32

echo "Flipping tables! (╯°□°）╯︵ ┻━┻"

num_rules=3

real=3    # exposed to the ELB as port 443
test=4    # used to install test certs for domain verification
health=5  # used by the ELB healthcheck

blue_prefix=855
green_prefix=866

vdm

ssh ubuntu@n.n.n.n "bash -s -x" -- <ixgbevf-upgrade.sh

bmaurer

diff --git a/include/net/tcp.h b/include/net/tcp.h
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -285,6 +285,7 @@
 extern int sysctl_tcp_adv_win_scale;
 extern int sysctl_tcp_tw_reuse;
 extern int sysctl_tcp_frto;
+extern int sysctl_tcp_syn_acceptq_pct;
 extern int sysctl_tcp_low_latency;
 extern int sysctl_tcp_dma_copybreak;

comerford

// kills long running ops in MongoDB (taking seconds as an arg to define "long")
// attempts to be a bit safer than killing all by excluding replication related operations
// and only targeting queries as opposed to commands etc.
 
killLongRunningOps = function(maxSecsRunning) {
    currOp = db.currentOp();
    for (oper in currOp.inprog) {
        op = currOp.inprog[oper-0];
        if (op.secs_running > maxSecsRunning && op.op == "query" && !op.ns.startsWith("local")) {
            print("Killing opId: " + op.opid