How to add a new disk to RAID5

I have a RAID5 consisting of three 10TB HDDs. This RAID5 has a total capacity of 20 TB.

I bought a new 10 TB HDD that I want to use to extend the RAID5: 4 HDDs with a total capacity of 30 TB. The file system on md0 is ext4. Currently, the RAID5 disks are sdc1, sdf1 and sde1. The additional disk is sdd1.

cat /proc/mdstat

The RAID5 is formatted with ext4 and available as md0.

mount

Steps

  1. Prepare new disk
  2. Add disk to RAID
  3. Grow RAID
  4. Extend ext4 files system.

Prepare new disk

First start with the preparation of the new disk. The disk is /dev/sdd and needs to have a partition. I use parted for this. First, create a label of type gpt.

parted -s -a optimal /dev/sdd mklabel gpt

Next is to create the partition using parted. This time, I am using the interface.

parted /dev/sdd

Add disk to RAID

The RAID is a software RAID on Linux, therefore mdadm is used to control the raid. To add a new disk, option –add is used and the raid and new disk are passed as parameters.

mdadm --add /dev/md0 /dev/sdd1

The result of the operation can be seen in mdstat.

cat /poc/mdstat

The new disk is added as a spare device. The (S) behind sdd1 means spare device. In case a device would fail, the spare device will take over automatically and a RAID rebuild will be triggered. This gives me less trouble in case a device fails, as I won’t have to do anything, but it won’t give me more space. The RAID5 is still at 20 TB.

Grow RAID

To make the RAID5 aware of the new disk and that it should be used for data storage, the RAID must be informed to use the new HDD using the grow command.

mdadm --grow --raid-devices=4 /dev/md0

The command informs the RAID that there are now 4 HDDs to be used, instead of 3. This command will trigger a RAID rebuild, as the information must be distributed to the HDDs.

This process will take some time. To learn how to increase the speed the sync, see my other blog about this topic.

The RAID5 consists now of 4 HDD, all working [UUUU]. The size of the RAID is still 20 TB. This is because the md0 has capacity of 30 TB, but the ext4 filesystem is still configured to make use of 20 TB.

Resize ext4 filesystem

To be able to use the 30TB available on the RAID5, you need to resize the file system. First, run an integrity check.

e2fsck -f /dev/md0

After the e2fsck ended without errors, the file system can be extended. This is done by using the tool resize2fs.

resize2fs /dev/md0

After resize2fs completes (can take a while), the size available is now 30TB:

mount /dev/md0 /mnt/md0/

Links

Let the world know

Monitor disk speed in Linux

Running a server allows you to do a lot of stuff from remote. Copying files is one of those tasks you can do from anywhere on the world while being logged on via SSH. For this task it is good to know the speed of read/write to get an idea if it’s working s expected. When sitting in front of your computer, you can see if a HDD is working, in Windows you see a MB/s indication, and in Linux? Not all copy commands show you the transfer rate by standard. Some disk intensive tasks won’t at all (RAID sync).

To monitor disk activities in Linux, several tools are available. One is iostat.

Installation

To install iostat in Debian, you must install the package sysstat

apt-get install sysstat

Execute

To run iostat, just enter iostat in the shell.

iostat

The output will list the captured read / write speed of the available devices. To get a continuous output of the disk activites, run iostat -y 1. This will update the output every second until you end the program.

iostat -y 1

Several options are available to control the output. To get the disk read / write in Mb and not in kB, add the -m flag

iostat -y 1 -m

Using iostat you can see the throughput oft he disks, even when you are running “hidden” tasks like a RAID sync or copy process in another session (screen).

Let the world know

Increase RAID sync rate

Scenario

  • The HDDs are in an external USB case.
  • RAID5 with 3 HDD (10TB)
  • Software RAID5 with mdadm and Debian Linux

Adding a new disk

When you add a new HDD to an existing RAID, a sync is started. In my case I added a 10TB disk to a RAID5. The sync started and as estimated time I got something in the range of days. The estimated time is listed in finish=5384 min.

This number goes up and down a little bit, but overall result is that the sync will need days. After checking the status again after a while, it still showed days: finish=3437min.

The main problem here Is the rate at which mdadm can sync the data. The value is between 30000K and 43000K. That’s not much given the size of the RAID. There are several tips available on the internet. What help me was to set the stripe_cache_size.

STRIPE_CACHE_SIZE

You set the size of stripe_cache_size for each RAID device (mdX). In case your RAID is md0:

echo 32768 > /sys/block/md0/md/stripe_cache_size

Result

The speed increased to 100000K/sec. That’s close to 3x faster than before. Time went down drastically.

Let the world know

Assign a static IP to DHCP client

After setting up a DHCP server on a Raspberry Pi running Linux I get working leases for my clients. However, these are not static. It can happen that my smartphone gets a new IP address the next it connects: 192.168.0.161 instead of 192.168.0.160. For some clients I want to make sure they always use the same IP. This can achieved with ISC DHCP Server by registering a static lease for a specific MAC.

Example

I’ll use my soundbar for the rest of this blog as an example. The MAC of the network card is bc:30:d9:2a:c9:50. I want to always assign the IP 192.168.0.152 to the soundbar.

Find out client data

To find out the client data like MAC and current lease, check the DHCP server log. Or take a look at the back of the device or its settings to find out the MAC. For the DHCP server log on assigned leases:

sudo systemctl status isc-dhcp-server.service

The last line shows that the DHCP server assigned an IP to a client and shows also the MAC address.

DHCPACK on 192.168.0.152 to bc:30:d9:2a:c9:50

Let’s make sure the MAC bc:30:d9:2a:c9:50 always gets the IP 192.168.0.152.

Configuration

sudo vim /etc/dhcp/dhcpd.conf

This is the DHCP server configuration file. I already configured it for a subnet 192.168.0.x where the server is assigning leases for the IP addresses in the range of 192.168.0.150 to 192.168.0.240.

Inside the subnet configuration, I have to add a configuration for the soundbar for IP 192.168.0.152.

host soundbar {
  hardware ethernet bc:30:d9:2a:c9:50;
  fixed-address 192.168.0.152;
}

The complete dhcpd.conf file will look like this:

subnet 192.168.0.0 netmask 255.255.255.0 {
  range 192.168.0.150 192.168.0.240;
  option routers 192.168.0.1;
  option domain-name "itsfullofstars.de";
  option domain-name-servers 8.8.8.8, 8.8.4.4;
  host soundbar {
    hardware ethernet bc:30:d9:2a:c9:50;
    fixed-address 192.168.0.152;
  }
}

Activate configuration

To activate the new configuration, make either DHCPD load the new configuration from file, or restart the service.

sudo systemctl restart isc-dhcp-server.service

Check the status of the service.

sudo systemctl status isc-dhcp-server.service

Result

The assigned leases can be found in the dhcpd.leases file. All leases assigned are listed here, including the mac address, IP address, start and end time of the lease. If all works out as planned, the soundbar will be in there with the static IP.

sudo more /var/lib/dhcp/dhcpd.leases

 

Let the world know

DHCP Server on Linux with Raspberry Pi

My internet provider is Unitymedia. Their default router comes with a DHCP server. Honestly, it’s one of the worst products I ever had to work with. My private network is 192.168.0.x. The DHCP server of the Unitymedia box is distributing from time to time leases for 192.168.192.x. Changing my private network to 192.168.192.x one is not working, as then the DHCP server picks another address range. Advise from Unitymedia help desk was to reboot the box, which, of course, won’t solve the problem. Because of this error, some of my devices are in a different network: Chromecast won’t work, broken internet connection on smartphones, etc.

I do have a Raspberry Pi (RP) in 24/7 use. My idea is to run my own DHCP server on the RP. This not only solves the DHCP problem, but also gives me more control over the DHCP configuration.

Preparation

sudo apt-get update
sudo apt-get install isc-dhcp-server

This installs ISC DHCP server. As you can see in the output, starting the DHCP server failed.

sudo systemctl status isc-dhcp-server.service

The error is simply caused because the DHCP server is not configured. Let’s change that.

Configuration

Several parameters must be activated and configured.

sudo vim /etc/dhcp/dhcpd.conf

Lease time

default-lease-time 600;
max-lease-time 7200;

Activate DHCP server

# If this DHCP server is the official DHCP server for the local
# network, the authoritative directive should be uncommented.
authoritative;

Subnet

This configures what IP address are going to be distributed. My private network is 192.168.0.x with the router on 192.168.0.1. As DNS you can use whatever you want, as an example I am using Google DNS servers.

subnet 192.168.0.0 netmask 255.255.255.0 {
  range 192.168.0.150 192.168.0.240;
  option routers 192.168.0.1;
  option domain-name "itsfullofstars.de";
  option domain-name-servers 8.8.8.8, 8.8.4.4;
}

This will give DHCP clients an IP address between .150 and .240, with router .1, Google DNS and sets the domain name to my own.

Deactivate old DHCP server

To not have the DHCP server provided by Unitymedia box still issuing wrong IP address, I am going to deactivate the service via the web interface.

Start DHCP server

After installing and configuring the new DHCP server on RP and deactivating the one from the router box, it’s time to start the new DHCP server.

Result

To see if a IP address is assigned, use this command:

sudo systemctl status isc-dhcp-server.service

Android

Putting my Android device into flight mode and back makes it connect to Wifi again and obtain a new IP address via DHCP. In the DHCP status log, I can see the DHCPDISCOVER from the Android device and that it got the IP address 192.168.0.150 assigned.

Mac

As my Mac always got the wrong IP assigned, I changed it to manual configuration. Change the mode to DHCP, apply and deactivate / activating Wifi.

Soundbar

And my soundbar that got a strange IP address assigned by the Unitymedia router box? Works too!

Chromcast streaming shows the SoundBar is now in the same network.

Let the world know

Apt-get unable to connect to IPv6 address

Recently I had the problem that running apt-get update stalled while trying to connect to an IPv6 address. For instance, on a Raspberry Pi, the update process stalls while trying to connect to archive.raspberrypi.org. All other connections worked fine. Looking at the console output, a difference was that apt was trying to connect to an IPv6 address.

The problem was caused by:

100% [Connecting to archive.raspberrypi.org (2a00:1098:0:80:1000:13:0:8)]

A quick internet search showed that you can force apt to not use IPv6 and only IPv4. As the download worked for IPv4, this seems like a reasonable workaround.

Solution

You can pass a parameter to disable IPv4 to apt-get, or write it to apt config file to make it persistent.

Configuration file

Create a new configuration file. This makes it easy for you to keep the change during updates and to know that you configured this.

sudo vim /etc/apt/apt.conf.d/99disable-ipv6
Insert Acquire::ForceIPv4 "true";
Save
apt-get update

Parameter

To disable IPv6 just once while calling apt, the parameter is Acquire::ForceIPv4=true.

sudo apt-get -o Acquire::ForceIPv4=true update

Result

Loading the package data from archive raspberrypi.org is now ignored and apt-get update works again.

 

Let the world know

Partitioning and formatting a large disk

I got a new 10 TB disk. Before I’ll add this one to a RAID, I want to play around with it, aka: test the drive. Therefore, I’ll need to format the drive to mount it. And before that, I need to create a partition.

FDISK

In the good old days, you used fdisk to partition a HDD. Since a few years, fdisk was replaced by parted as fdisk got some issues with large. Nevertheless, it still works.

Make sure to create a GPT partition table (g), and not the old new partition (n) alternative. Creating a new partition using “n” gives you a 2 TB partition.

Creating a new disklabel of type GPT using “g” gives you 10TB. The create the GPT disklabel and the new partition, use: g & n.

PARTED

An alternative to fdisk is parted. Parted is newer than fdisk and there are GUIs available to make it easier to end users to use it. Parted allows to pass parameters to it and do all the partition and sizing stuff in one command.

parted -s -a optimal /dev/sde mkpart primary ext4 0% 100%

Create Filesystem

Finally, after the HDD is partitioned, it’s time to format the partition with EXT4. Of course you can use a different filesystem.

mkfs.ext4 /dev/sde1

mount /dev/sde1 /mnt/sde/

Let the world know

Ideen für 2019

2019 ist schon ein paar Tage alt, Zeit die Ideen zu sammeln und einen Plan für 2019 und 2020 zu Erstellen. Seit meiner Rückkehr nach Deutschland habe ich die Idee den SAP Community Erfolg aus Brasilien in Karlsruhe und Umgebung zu wiederholen.

Rückblick

In Brasilien habe ich maßgeblich dazu beigetragen das die SAP Inside Tracks von einem Event von wenigen Nerds zu einem Erfolg wurde. Der Prozess lief leider nicht ganz reibungslos, da man hierbei festgezurrte Denkmuster durchbrechen muss, und nicht jeder mag Veränderungen. Davon profitieren ja, aber aktiv mitgestalten? Was zählt ist aber auch das Resultat. Von einem SIT in Sao Paulo mit etwas 20 Leuten hin zu einer ganzen Menge an SITs in verschiedenen Städten mit um die 100 Teilnehmern pro Event. Ganz neue Themen wurden erschlossen, neue Leute kamen hinzu, neue Freundschaften geschlossen.

In Rio habe ich mich gegen das SIT Konzept entschieden und Meetups organisiert. Der Vorteil ist das diese keine ganztägige Veranstaltung sind wo man möglichst viel Themenkomplexe abdecken will. Während SIT schon das “Schimpfwort” SAP enthält, ist ein Meetup ein etablierter Begriff, in und außerhalb der IT. Das hilft wirklich wenn man erklären muss was man vorhat und was einen erwartet.

Was war der Unterschied? Mehrmals im Jahr eine Veranstaltung mit etwa 3 Stunden Vorträge auf 2 bis 4 Stunden verteilt und ein klarer Fokus auf ein Thema. Vor allem die Möglichkeit sich auf ein Thema zu fokussieren kam sehr gut an. Es gibt keine Unterbrechung in der Agenda, kein Vortrag morgens und erst wieder abends zu ABAP. Die Gruppe ist dadurch auch homogener, was gut und schlecht ist, aber: unter Gleichen findet man einfacher Anschluß und kann die Probleme besser besprechen. Die Fortführung der Meetups ist auch einfacher. Auch heute noch finden meine Meetups SAP Rio de Janeiro erfolgreich statt, ohne das ich noch komplett involviert bin. Es gab Veranstaltungen zu ABAP, Fiori, Cloud, S/4, etc. 2015 fand der erste Meetup statt, bis heute haben weit über 600 Leute teilgenommen. Weit über 1.000 Leute haben sich registriert. Für SAP Themen ist es die größte Veranstaltung im Bundestaat Rio, wenn nicht sogar für alle Bundesländer mit Ausnahme Sao Paulo (hier findet das SAP Forum von der SAP statt)., oder sogar für ganz Südamerika.

Die Idee

Und für 2019 und die Region Nord-Baden / Karlsruhe? Es gibt auf jeden Fall keinen Bedarf für einen SAP Inside Track. Dazu finde ich das Konzept einer großen Veranstaltung mit Vorträgen aus dem ganzen Land + Nachbarländer nicht mehr zeitgemäß. Dafür sind die Themen aus der SAP Welt zu komplex und zu schnelllebig. SAP stellt neue Funktionen alle 2 Wochen in der SAP Cloud bereit, darüber mit 1 – 2 Jahren Verspätung zu berichten ist einfach zu spät.

Der Plan

Etablieren einer Vortragsreihe im Meetup Format. Motto: think global, act local.

  • Anzahl: 2 bis 4 mal im Jahr.
  • Dauer: 2 bis 4 Stunden
  • Tag: unter der Woche, zwischen Dienstag, Mittwoch oder Donnerstag
  • 3 bis 4 Vortragende
  • etwa 30 Minuten pro Vortrag
  • Rahmen: Keynote, Vorträge, Wrap-up
  • Fokus auf 1 Thema (Cloud, C4C, Analytics, Mobile, Fiori, Personas, Design Thinking, etc.)
  • Offen für alle: SAP oder nicht SAP, Student, Anfänger, Berufserfahren, CTO
  • Kostenlos
  • So viel hands-on wie möglich, so viel Theorie wie nötig

Was wird benötigt

  • Motivierte Leute. Ob einfach hingehen um zu lernen oder einen Vortrag zu halten. Vor 20 Leuten einen Vortrag zu halten motiviert mehr als vor 20 leeren Stühlen.
  • Räume. Irgendwo muss die Veranstaltung ja stattfinden.
  • Mehr Leute. So etwas steht und fällt mit den Menschen.

Den Anfang habe ich mal gemacht und versuche interessierte Leute im Raum KA zu finden. Der Stammtisch SAP Karlsruhe existiert, einfach vorbeikommen. Auch auf Twitter (@tobiashofmann) kann man mit mir Kontakt aufnehmen. Oder über LinkedIn, Xing, E-Mail, etc.

Der nächste Stammtisch SAP Karlsruhe findet am 30.1. statt. Mehr infos hier.

Let the world know

State of the art documentation from SAP

SAP is investing heavily in marketing the Fiori for iOS and the SDK. In case you are slightly interested in Fiori and UX in general in SAP, for sure you heard a lot about the SDK. 2 ½ years after the announcement the Fiori Design guidelines include an iOS section, there are SAP Developers tutorials, a special iPad app for learning its usage is available, even Apple has set up a Fiori page. Current version of the SDK is 3.0, and now there is even an Android version available (with much less marketing activities).

If you want to write an app with the SDK, make sure you have an iPad. The online SDK documentation is available too, but offers less benefit than the Fiori Mentor app. In case you are wondering why the SDK documentation is not good enough: I suggest you take a look at it. For instance, the documentation for the map component.

As you can see, you see … not much.

No images, therefore: good luck in finding out what the UI control should look like. A look at the page source code reveals that the images are only visible to SAP employees with access to SAP’s intranet.

Server github.wdf.sap.corp is not accessible from the internet. In case you are wondering how to find the SDK documentation: SAP Cloud Platform SDK for iOS Assistant contains a link in its help to the API. And of course: Google. So yes, made available to the interested developer. No S-, I- or D-User required.

In case you are one of the few that develop apps using Fiori for iOS SDK, ask your manager to get an iPad. The public available SDK documentation is already not easily consumable (use the Fiori Mentor app), does not include complete sample code and comes with missing images.

SAP is now pushing the intelligent enterprise. Let’s hope that it will be intelligent enough to test if the public available documentation is complete.

 

Let the world know

Remove last n characters of file in MacOs

With MacOS and finder you can easily substitute characters of files using the rename functionality. Just select 2 or more files, right click, and inform the character you want to substitute, like _ with space.

To remove the last N characters from a file that looks like Text-2018221112.mp4 to Text is more complicated. The rename dialog does not understand regex. What you can use is the shell and rename

Install rename

brew install rename

Go to the directory with the files and run

rename -n 's/.{11}.mp4/\.mp4/' *

Rename uses the well known sed syntax s/char/replace/.

  • -n runs the replace in simulation mode. It will print the result, without renaming the files yet. Perfect for testing.
  • {11} number of characters to replace
  • \.mp4 is to insert mp4 again, as .{}11.mp4 will replace also the file suffix

In case the output matches your goal, run rename without -n and the files will be renamed.

rename 's/.{11}.mp4/\.mp4/' *

Let the world know