Het is tijd om mijn wielerhobby in te zetten voor het goede doel. Met mijn teammaten van het Jeroen Bosch Ziekenhuis neem ik dit jaar deel aan AlpedHuZes om geld in te zamelen voor de strijd tegen kanker. En ik ga niet 6 maar liefst 7 keer omhoog om maar aan te geven dat ik heel erg mijin best wil doen voor het KWF. Steun mij en stort een bedrag. Alle bijdragen helpen!
| Last week I visited VMorld 2011 in Copenhagen. Copenhagen is a great city, I was very pleasently surprised. Clean streets, friendly and very relaxed people, children who seem to look fairly happy, no fat people in hamburger restaurants, a lot of kids wear sporty clothes of their sportclub, public transport is organized well, no fights to get in or out a train or metro, good restaurants and great architecture and design. And at VMworld a lot of the served food was healthy. I like Copenhagen.
This VMworld was my fourth in a row and every year it proves to be an exiting event, this year too. I visited around 14 break-out sessions which were really great. Technical deep dives, Q&A sessions with experts and best practices that are really useable. Also attended spectaculair general sessions with talks about how VMware sees the near future. |
 |
|
What impressed me the most? I don't know were to start. When you look at the future the Datacenters will become automated. vClouds will be used by everyone, big and small businesses. All the manual work done nowadays will be automated. Installing hosts, configuring swithes, deploying VM's, creating secure networks for organizations using the same datacenter, all actions are automized. Repairing is old-school, automatically replacing is faster and therefor more efficient. And the datacenter lights will be turned on just once a week to replace some physical parts. Around 2014 we'll have phsical hosts capable of having 2 physical CPU's with each 16 cores, 300GB of RAM and running around 320 VM's. Company's running 200 virtual servers only need two physical hosts. When you look at new products VMware anounced the intelligent vCenter Operations 5.0 tool for monitoring datacenters, this promises to be a great tool. Operations relies on third party tools to deliver the monitoring results. I had some talks with the vCenter Operations developpers and guess what: 360° Viewpoint is fully compatible! Other newly announced products Applblast, Horizon and Octopus will make a world of change: deploying the same software and sharing files anywhere, anytime, anyplace. And independent of the device and platform you are using. Run Microsoft's Office from an iPad, MAC, Linux and Android device and share your files within seconds with your collegues on the other side of the world. Another great product: VMware Go. It makes an assessment of your physical hosts in your network, determines which host is most appropiate to install the VMware hypervisor on, then it replaces the current OS with the VMware hypervisor and installs back the original Operating System and Apps as a VM on this hypervisor. Incredible. Really cool for small and mid-size business. Even the Borg will be impressed with this assimilation technology. The break-sessions gave me a lot of insight into (Virtual) Memory Management, DRS, HA, vStorage, Storage vMotion, automating ESXi deployment, vSphere best practices, and the list goes on. Also the caveats of stretched clusters and DR and Disaster Prevention were very interesting. And I came in 4th with the NetApp Cycling race, 1/100 of a second behind #3 and 8/100 of #2. The winner got a week of training on Mallorca with the pro's, I missed it by 0.9 seconds. I shifted wrong.. damn. Conclusion: exiting times are waiting us! The cloud is yours. |
Connect a USB pen drive to a Linux OS, in my case a Ubuntu.
Download the vSphere 5 ESXi ISO installation file.
On the Linux machine install syslinux and mboot.
Execute the following commands:
# sudo apt-get install mbr syslinux # sudo /sbin/fdisk /dev/sdb # sudo umount /dev/sdb1 # sudo /sbin/mkfs.vfat -F 32 -n USB /dev/sdb1 # cd /tmp/ # sudo syslinux /dev/sdb1 # sudo install-mbr /dev/sdb # sudo mkdir /mnt/cdrom # sudo mkdir /mnt/usb # sudo mount -o loop /tmp/VMware-VMvisor-Installer-5.0.0-469512.x86_64.iso /mnt/cdrom/ # sudo mount /dev/sdb1 /mnt/usb/ # sudo cp -r /mnt/cdrom/* /mnt/usb/ # sudo mv /mnt/usb/isolinux.cfg /mnt/usb/syslinux.cfg # sudo vi /mnt/usb/syslinux.cfg
In the file /usbdisk/syslinux.cfg, change the line APPEND -c boot.cfg to APPEND -c boot.cfg -p 1.
# sudo umount /mnt/usb # sudo umount /mnt/cdrom
Ready, says VMware.
But… When I put the USB drive in the server I get the BOOT prompt. The menu is not loaded correctly.
It has got something to do with that this syslinux kernel version needs another layout of the syslinux.cfg file. Untill I solve this issue I just type on the BOOT command prompt:
mboot.c32 -c boot.cfg -p 1
This works.
Install perl DBI module
# perl -MCPAN -e 'install DBI'
Install the Application Development Client to be able to compile DBB:DB2 perl modules. Download it here:
http://www-01.ibm.com/support/docview.wss?rs=71&uid=swg21256079
To Install the DB2 Application Development Client you need the library libstdc++.so.5. On my Centos 5.5 installation it was missing. I installed it like this:
# rpm -q --whatprovides libstdc++.so.5 # yum -y install compat-libstdc++-33.i386 # find /usr/ -name libstdc++*
Download the correct DB2 "Dataserver Runtime Client" and store it in /tmp, I used v9.7:
http://www-01.ibm.com/support/docview.wss?rs=4020&uid=swg21385217
To Install the DB2 Runtime Client you need the library libaio.so.1. On my Centos 5.5 installation it was missing. I installed it like this:
# rpm -q --whatprovides libaio.so.1 no package provides libaio.so.1 # yum -y install libaio.x86_64 # find /usr/lib64 -name libaio* # /usr/lib64/libaio.so.1
# cd /tmp # tar xvzf ibm_data_server_runtime_client_linuxx64_v97.tar.gz # cd rtcl # ./db2_install
After the installation is finished succesfully you'll see this:
db2_install program completed successfully.
Set environmental variable DB2_HOME:
# export DB2_HOME=/opt/IBM/db2/V9.7
Configure the DB2 Client. This step is needed to prevent error messages like:
SQL10007N Message "-1390" could not be retrieved. Reason code: "3"
when trying to start /opt/ibm/db2/V9.7/bin/db2
# groupadd db2grp1 # groupadd db2fgrp1 # groupadd dasadm1 # useradd -g db2grp1 -m -d /home/db2inst1 db2inst1 # useradd -g db2fgrp1 -m -d /home/db2fenc1 db2fenc1 # useradd -g dasadm1 -m -d /home/dasusr1 dasusr1 # /opt/ibm/db2/V9.7/instance/db2icrt db2inst1
Install perl DDB::DB2 module. Download the modules here: http://httpupdate37.cpanel.net/CPAN/authors/id/I/IB/IBMTORDB2/DBD-DB2-1.81.tar.gz
# cd /tmp # tar zvxf DBD-DB2-1.81.tar.gz # cd DBD-DB2-1.81 # perl Makefile.PL # make # make test # make install
Wow, now we can connect to the database.
Get the Nagios plugin:
# cd /tmp # wget http://labs.consol.de/wp-content/uploads/2010/12/check_db2_health-1.0.3.tar.gz # tar xvzf check_db2_health-1.0.3.tar.gz # cd check_db2_health-1.0.3 # ./configure # make # make install
Now the script is installed in the default Nagios libexec directory:
# ls /usr/local/nagios/libexec/check_db2_health
Set the environmental setting so the library is found:
# LD_LIBRARY_PATH=/opt/ibm/db2/V9.7/lib64
Check the database:
# check_db2_health --username=root --password=******* --hostname vcenter.r71.nl --mode=connection-time
Make sure the database accepts connections. Login to the vCenter server using ssh and execute the following commands:
# /opt/db2/v9.7.2/adm/db2set DB2COMM=tcpip # /opt/db2/v9.7.2/bin/db2 update dbm cfg using svcename # /opt/db2/v9.7.2/bin/db2 db2 => LIST ACTIVE DATABASES Active Databases Database name = VCDB Applications connected currently = 15 Database path = /storage/db/db2/home/db2inst1/db2inst1/NODE0000/SQL00001/
Beschikbare users:
db2 => GET CONTACTS Name Type Address Max Page Length Description ------------------------------ ---------- ------------------------------ ---------------- ------------------------------ db2inst1 EMAIL db2inst1@localhost Contact added by DB2 Setup Wizard
Login op linux als user db2inst1
~> su - db2inst1 ~> db2set DB2COMM=tcpip ~> db2 update dbm cfg using agentpri system ~> db2 force applications all ~> dbstop ~> dbstart
Problem: when trying to connect the client to the database I get a Segmentation fault error. I'm having trouble solving this issue… Ideas are welcome!
As we all rely more and more on vCenter for managing datacenters with VMware ESX hosts and for delivering our virtual desktops, we have to be sure vCenter services are running smootly. So monitoring is important.
First there was only vCenter on a Windows host. Works fine. Monitoring was possible because you controlled the Windows box and the database. Now there is the new vCenter Virtual Appliance running on SuSE 11.1. Very interesting and probably the way to go in the future. But how to monitor the vCenter services?
For monitoring I mostly use Simple Network Monitoring Protocol (SNMP). Really great, not simple. The vCenter appliance has not got the SNMP daemon installed by default, so some activity is required. Glad we have root access to the machine. Here we go:
Check the Suse version:
# cat /etc/SuSE-release SUSE Linux Enterprise Server 11 (x86_64) VERSION = 11 PATCHLEVEL = 1
Add the repository of this Suse version to YAST, the Suse software manager tool. Use SSH to login to the vCenter server.
# yast
The software is installed. Quit YAST.
 |
 |
 |
 |
Allow the monitoring server to connect to the snmpd daemon:
# vi /etc/hosts.allow sshd: ALL : ALLOW snmpd: ALL : ALLOW ALL: KNOWN
Start the SNMP daemon and make sure it starts after a reboot:
# service snmpd start # chkconfig snmpd on # vi /etc/snmp/snmpd.conf
Change
rwcommunity public 127.0.0.1
to
rwcommunity public
Now try a snmp quey from the monitoring server:
# snmpwalk -v2c -c public vcenter01.r71.nl SNMPv2-MIB::sysDescr.0 = STRING: Linux vcenter01.r71.nl 2.6.32.29-0.3-default SNMPv2-MIB::sysObjectID.0 = OID: NET-SNMP-TC::linux DISMAN-EVENT-MIB::sysUpTimeInstance = Timeticks: (169539) 0:28:15.39 SNMPv2-MIB::sysContact.0 = STRING: Roderick SNMPv2-MIB::sysName.0 = STRING: vcenter01.r71.nl SNMPv2-MIB::sysLocation.0 = STRING: R71 ..
It works!
Now configure your monitoring tool. If you don't have a good monitoring tool (shame on you) then get 360° Viewpoint!
(next I'm trying to figure out how to monitor the DB2 database, will report back later when ready)
 |
 |
Het is 23 september 2012, de nieuwe Specialized Tarmac SL3 frame is uitgebracht. De snelheid van het frame in combinatie met de juiste wielrenner komt overeen en overtreft zelfs die van een Neutrino deeltje waardoor tijdreizen op deze fiets naar het verleden mogelijk is.
|
Problem: vCenter crashes also after a restart of the service. SQL Express is used (has a 4GB limit). Event viewer error log: Solution: Download a script from the VMware Website KB article for vCenter2 or vCenter4. Tested by me on vCenter5. Install Microsoft Visual Studio on your vCenter server. Do this by right clicking a CMD shortcut, run as administrator, start the MSI file. You can also run this script from the commandline (no visual studio needed). In the script change the flag @DELETE_DATA from 0 to 1 so that it actually deletes the rows. cd "C:\Program Files (x86)\Microsoft SQL Server\90\Tools\Binn" OSQL.EXE -E -S SERVERNAME\SQLEXP_VIM -i C:\file_containing_the_sql_script.sql This was the result output: VirtualCenter database user: dbo Logged in user: dbo Cutoff date: '2011/03/25' Batch size: 10000 2011-09-21 12:18:08 starting... 2011-09-21 12:18:08 VPX_TASK: no matching rows to delete. 2011-09-21 12:18:10 VPX_EVENT_ARG: no matching rows to delete. 2011-09-21 12:18:10 VPX_EVENT: no matching rows to delete. 2011-09-21 12:18:10 VPX_HIST_STAT1: no matching rows to delete. 2011-09-21 12:18:10 VPX_SAMPLE_TIME1: no matching rows to delete. 2011-09-21 12:18:10 VPX_HIST_STAT2: no matching rows to delete. 2011-09-21 12:18:10 VPX_SAMPLE_TIME2: no matching rows to delete. 2011-09-21 12:18:10 VPX_HIST_STAT3: no matching rows to delete. 2011-09-21 12:18:10 VPX_SAMPLE_TIME3: no matching rows to delete. 2011-09-21 12:18:11 VPX_HIST_STAT4: will attempt to delete 62633 rows. 2011-09-21 12:18:11 completed 10000 rows... 2011-09-21 12:18:12 completed 20000 rows... 2011-09-21 12:18:13 completed 30000 rows... 2011-09-21 12:18:13 completed 40000 rows... 2011-09-21 12:18:14 completed 50000 rows... 2011-09-21 12:18:15 completed 60000 rows... 2011-09-21 12:18:16 VPX_HIST_STAT4: deleted 62633 total rows. 2011-09-21 12:18:16 VPX_SAMPLE_TIME4: will attempt to delete 298 rows. 2011-09-21 12:18:16 VPX_SAMPLE_TIME4: deleted 298 total rows. ****************** SUMMARY ******************* Deleted 0 rows from VPX_TASK table. Deleted 0 rows from VPX_EVENT_ARG table. Deleted 0 rows from VPX_EVENT table. Deleted 0 rows from VPX_HIST_STAT1 table. Deleted 0 rows from VPX_SAMPLE_TIME1 table. Deleted 0 rows from VPX_HIST_STAT2 table. Deleted 0 rows from VPX_SAMPLE_TIME2 table. Deleted 0 rows from VPX_HIST_STAT3 table. Deleted 0 rows from VPX_SAMPLE_TIME3 table. Deleted 62633 rows from VPX_HIST_STAT4 table. Deleted 298 rows from VPX_SAMPLE_TIME4 table.
After this shrink the database file by removing the newly created empty space. Go Visual Studio > Select Database VIM_VCDB > Tasts > Shrink > Database I'm not a DBA guy but I think you can do the same with a SQL statement. Add this to a file and run it like above with OSQL.EXE. DBCC SHRINKDATABASE SQLEXP_VIM After this start the VI client, press CTRL+SHIFT+I and adjust the amount of logged data. After 2 days the databse reached 4GB again! I tried the script I found in this article. The script seems to run forever, so I checked it the next day. The database size shrunk from 4GB to 400MB. I don't know if this was a permanent solution cause the next day in my home lab I switched to the vCenter Virtual Appliance. |
Howto perform a scripted ESXi 5.0 installation via HTTP and not TFTP? You can find the answer here. And the WHY question? Because HTTP is faster, can handle larger files, does not cost performance of your PXE boot server, and you not limited to your network (via internet is possible too).
For ESXi4 it was fairly easy to install via HTTP. VMware changed this into something more complicated, maybe to promote their integrated installation feature. The VMware vSphere documentation also lacks some information. This is the complete setup:
Install a PXE boot server.
Install the gPXELINUX.0 image:
yum install gpxe-bootimgs find /usr/ -name *gpxe* cp /usr/share/syslinux/gpxelinux.0 /tftpboot/
The DHCP config for Apache:
# pxe boot stuff allow booting; allow bootp; # gPXE options option space gpxe; option gpxe-encap-opts code 175 = encapsulate gpxe; option gpxe.bus-id code 177 = string; class "pxeclients" { match if substring(option vendor-class-identifier, 0, 9) = "PXEClient"; next-server 10.0.2.11; if not exists gpxe.bus-id { filename "/gpxelinux.0"; } }
Now edit the /tftproot/pxelinux.conf/default file which will be loaded by the gPXE bootloader.
LABEL ESXi 5.0 KickStart and HTTP KERNEL http://10.0.2.14:8080/vSphere/ESXi_5.0/MBOOT.C32 APPEND -c http://10.0.2.14:8080/vSphere/ESXi_5.0/BOOT.CFG ks=http://10.0.2.14:8080/vSphere/ESXi_5.0/ks.cfg +++ IPAPPEND 1
Then the BOOT.CFG file from the extracted VMware's ISO was edited:
bootstate=0 title=Loading ESXi installer kernel=http://10.0.2.14:8080/vSphere/ESXi_5.0/TBOOT.B00 kernelopt=runweasel #kernelopt=ks=http://10.0.2.14:8080/vSphere/ESXi_5.0/esxi_ksFiles/ks.cfg modules=http://10.0.2.14:8080/vSphere/ESXi_5.0/B.B00 --- http://10.0.2.14:8080/vSphere/ESXi_5.0/USEROPTS.GZ --- http://10.0.2.14:8080/vSphere/ESXi_5.0/K.B00 --- http://10.0.2.14:8080/vSphere/ESXi_5.0/A.B00 --- http://10.0.2.14:8080/vSphere/ESXi_5.0/S.V00 --- http://10.0.2.14:8080/vSphere/ESXi_5.0/WEASELIN.I00 --- http://10.0.2.14:8080/vSphere/ESXi_5.0/TOOLS.T00 --- http://10.0.2.14:8080/vSphere/ESXi_5.0/IMGDB.TGZ --- http://10.0.2.14:8080/vSphere/ESXi_5.0/IMGPAYLD.TGZ
I use a Linux webserver, capital sensitive. I had to change all the lowercase letters in the filenames into captials. And mind this: the real filenames are different then the ones named in the vSphere install and configure guide.
Create a ks.cfg file and make it available on your webserver:
accepteula #dryrun install --firstdisk --overwritevmfs #url --url http://10.0.2.11/vmware/esxi/5.0/ks.cfg rootpw password reboot network --bootproto=static --ip=10.0.2.100 --gateway=10.0.1.254 --nameserver=10.0.2.11 --netmask=255.0.0.0 --hostname=esxi5.r71.nl --addvmportgroup=1 %firstboot --interpreter=busybox # enable HV (Hardware Virtualization to run nested 64bit Guests + Hyper-V VM) #grep -i "vhv.allow" /etc/vmware/config || echo "vhv.allow = \"TRUE\"" >> /etc/vmware/config # enable & start remote ESXi Shell (SSH) vim-cmd hostsvc/enable_ssh vim-cmd hostsvc/start_ssh # enable & start ESXi Shell (TSM) vim-cmd hostsvc/enable_esx_shell vim-cmd hostsvc/start_esx_shell
Good luck.
Alternative approach, same result:
Edit the dhcpd.conf as follows:
allow booting; allow bootp; option option-128 code 128 = string; option option-129 code 129 = text; next-server 10.0.2.11; #filename "/pxelinux.0"; if ((exists user-class) and (option user-class = "gPXE")) { filename "http://10.0.2.14:8080/gpxe/boot.php"; } else { filename "undionly.kpxe"; }
Add the php file to the webserver with the following content:
<?php echo "#!gpxe\n"; echo 'chain pxelinux.0' ."\n"; ?>
Add the following files to the webserver:
Add the following file to the tftpserver:
That should work too.
A short overview about how to install an ESXi5 server via PXE boot using the following steps.
mkdir /mnt/cdrom mkdir /tftpboot/images/vsphere/esxi5.0 mount -o loop VMware-VMvisor-Installer-5.0.0-469512.x86_64.iso /media/tmp or mount /dev/cdrom/ /mnt/cdrom cp -rav /mnt/cdrom/* /tftpboot/images/vsphere/esxi5.0 chmod -R +w /tftpboot/images/vsphere/esxi5.0
vi /tftpboot/images/vsphere/esxi5.0/boot.cfg bootstate=0 title=Loading ESXi installer prefix=/images/vmware/esxi/5.0/ kernel=tboot.b00 #kernelopt=runweasel modules=b.b00 --- useropts.gz --- k.b00 --- a.b00 --- ata-pata.v00 --- ata-pata.v01 --- ata-pata.v02 --- ata-pata.v03 --- ata-pata.v04 --- ata-pata.v05 --- ata-pata.v06 --- ata-pata.v07 --- block-cc.v00 --- ehci-ehc.v00 --- s.v00 --- weaselin.i00 --- ima-qla4.v00 --- ipmi-ipm.v00 --- ipmi-ipm.v01 --- ipmi-ipm.v02 --- misc-cni.v00 --- misc-dri.v00 --- net-be2n.v00 --- net-bnx2.v00 --- net-bnx2.v01 --- net-cnic.v00 --- net-e100.v00 --- net-e100.v01 --- net-enic.v00 --- net-forc.v00 --- net-igb.v00 --- net-ixgb.v00 --- net-nx-n.v00 --- net-r816.v00 --- net-r816.v01 --- net-s2io.v00 --- net-sky2.v00 --- net-tg3.v00 --- ohci-usb.v00 --- sata-ahc.v00 --- sata-ata.v00 --- sata-sat.v00 --- sata-sat.v01 --- sata-sat.v02 --- sata-sat.v03 --- scsi-aac.v00 --- scsi-adp.v00 --- scsi-aic.v00 --- scsi-bnx.v00 --- scsi-fni.v00 --- scsi-hps.v00 --- scsi-ips.v00 --- scsi-lpf.v00 --- scsi-meg.v00 --- scsi-meg.v01 --- scsi-meg.v02 --- scsi-mpt.v00 --- scsi-mpt.v01 --- scsi-mpt.v02 --- scsi-qla.v00 --- scsi-qla.v01 --- uhci-usb.v00 --- tools.t00 --- imgdb.tgz --- imgpayld.tgz build= updated=0
bootstate=0 title=Loading ESXi installer kernel=http://10.0.2.14:8080/vSphere/ESXi_5.0/tboot.b00 kernelopt=runweasel #kernelopt=ks=http://10.0.2.14:8080/esxi_ksFiles/ks.cfg modules=http://10.0.2.14:8080/vSphere/ESXi_5.0/b.b00 --- http://10.0.2.14:8080/vSphere/ESXi_5.0/useropts.gz --- http://10.0.2.14:8080/vSphere/ESXi_5.0/k.b00 --- http://10.0.2.14:8080/vSphere/ESXi_5.0/a.b00 --- http://10.0.2.14:8080/vSphere/ESXi_5.0/s.v00 --- http://10.0.2.14:8080/vSphere/ESXi_5.0/weaselin.v00 --- http://10.0.2.14:8080/vSphere/ESXi_5.0/tools.t00 --- http://10.0.2.14:8080/vSphere/ESXi_5.0/imgdb.tgz --- http://10.0.2.14:8080/vSphere/ESXi_5.0/imgpayld.tgz
accepteula #dryrun install --firstdisk --overwritevmfs #url --url http://10.0.2.11/vmware/esxi/5.0/ks.cfg rootpw password reboot network --bootproto=static --ip=10.0.2.100 --gateway=10.0.1.254 --nameserver=10.0.2.11 --netmask=255.0.0.0 --hostname=esxi5.r71.nl --addvmportgroup=1 %firstboot --interpreter=busybox # enable HV (Hardware Virtualization to run nested 64bit Guests + Hyper-V VM) #grep -i "vhv.allow" /etc/vmware/config || echo "vhv.allow = \"TRUE\"" >> /etc/vmware/config # enable & start remote ESXi Shell (SSH) vim-cmd hostsvc/enable_ssh vim-cmd hostsvc/start_ssh # enable & start ESXi Shell (TSM) vim-cmd hostsvc/enable_esx_shell vim-cmd hostsvc/start_esx_shell
Make the script available on a webserver. Use the URL to the file in step 5.
vi /tftpboot/pxelinux.cfg/default LABEL ESXi5 KERNEL images/vsphere/esxi5/mboot.c32 APPEND -c images/vsphere/esxi5/boot.cfg pxebooting ks=http://10.0.2.14:8080/vSphere/ESXi_5.0/ks.cfg +++ IPAPPEND 1
If you got an ESXi hosts running and you want to run an ESXi server as in virtual machine and add VM's to the virtualized ESXi host, these nested VM's won't start. Error message:
ESX Running VMware ESX in a virtual machine requires the outer virtual machine to be configured for running a VMware ESX guest operating system. You cannot power on a virtual machine until the outer virtual machine is reconfigured.
The solution is simple, not supported by VMware:
* Shut down the virtual ESXi5 VM
* Click Edit Settings (of that VM)
* Click the Options tab
* Click Advanced / General / Configuration Parameters…
* Click Add Row
* For the Name/Value enter: monitor_control.restrict_b