• Home
  • Work
  • Blog
  • Security
  • Links
  • Archive for the ‘Hardware’ Category

    RIPE-Atlas Anchor installation

    2016 - 02.28

    Ripe-Atlas project Anchor installation.

    Anchor is high capacity RIPE-Atlas probe for internet measurements.

    Anchor mainly consist of  Soekris Net6501-70 board (can find on kd85.com) and running Centos 6.x.



    Samsung SSD 840 PRO – performance degradation

    2014 - 12.03

    About year ago I wrote blogpost about endurance and performance test of Samsung SSD 840 PRO. Some things has changed, especially firmware of disks.

    I performed test on disk with firmware DXM04B0Q. When you want to buy this disk now you probably have firmware DXM05B0Q or DXM06B0Q.

    Problem is, both new firmwares has the same performance degradation issue.

    Since my first blogpost we bought about 100 disks and after few months we’re started observing some problems on database servers. Servers and disks was very slow, we are talking about 10 MB/s continuous write speed.

    First of all, we repeat our direct disk test, then we started to test disk with different filesystems.



    Samsung SSD 840: Endurance Destruct Test

    2013 - 10.10

    When you operate a Datacenter with many servers you also probably have big number of installed disks. In most cases even you have cluster, especialy in small companies, there is still SPOF (Single point of failure) somewhere.

    In this SPOF it’s critical to know condition of the system. When system using HDD it’s important to know HDD condition to prevent failures, plan maintenance etc.

    We will talk about Samsung SSD 840 PRO series. We realize this SSD has very good performance and lifetime. Before we use it in production we must know how to monitor condition. There is many articles and technical specification but we had a lot of questions without answer.


    (Update: 10.10.2013)

    (Samsung performance issues, read new post, 03.12.2014)


    Forgotten Skills: Build your own linux kernel

    2013 - 09.23

    Firt we will talk about WHY. Many people have many opinions on that and I’d like to present you mine.

    I work as System Specialist. During my work I often came to point that normal kernel from distribution doesn’t work. Whole or part of server was unusable because one of these common reasons:

    • Virtualization support with combination of new HW
    • Missing driver for some Raid Controllers
    • Missing driver for network cards
    • Security hardening patches for specific systems

    I don’t remember which of the reason was the first when I must build my own kernel to be able use server or computer with all it’s hardware. In few next years another big reason appear.

    • New kernel have fixed some performance issues

    Now we can talk about how to do it.


    IBM server x336 modification SCSI to SATA – 2 (LED Indicators)

    2012 - 05.17

    This is small update of article about modification server IBM x336 from SCSI to SATA.

    Today I’d like to talk about front LED indicators.

    Our cheap sata backplane has two outputs for LED indicators. One is for power and one for activity.


    I’ve bought two colors led for indication power and activity of each HDD. As I dicovered later activity output not work properly.


    IBM server x336 modification SCSI to SATA

    2012 - 05.09

    Few months ago I bought some older IBM servers. Two cheap iBM x336 with 4 GB RAM and two SCSI disks.

    After first look I decided to use ona server as my new server with two 146 GB SCSI disks and FreeBSD installed and second server modify for two SATA disks. SCSI disks are great but more expensive and lower capacity then SATA disks.

    I’d like to explain how to modify IBM X336 SCSI version and using it with SATA disks. During modification there was many obstacles that I need to solve.