a strange VMWARE Error

Author: Stephan  |  Category: Linux, Netzwerk, vmware

this morning websphere client of one of the cluster I manage stopped working with the following error.
Cannot connect to vCenter Single Sign On server. https://(FQDN or IP address of SSO):7444/ims/STSService.

The Reason was simple and curious.
The SSO Service was running a year an needs a restart because of a expired temporary certifcate.

Here is the KB article
http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=2122788

shellshockfix for old Linuxsystems

Author: Stephan  |  Category: Linux, Netzwerk

Old Systems which are vunerable because wikipedia
shellshock aren’t that easy to update. One method is compilling bash manually ( stolen from ubuntu )

mkdir src
cd src
wget http://ftp.gnu.org/gnu/bash/bash-4.3.tar.gz
#download all patches
for i in $(seq -f "%03g" 0 25); do wget http://ftp.gnu.org/gnu/bash/bash-4.3-patches/bash43-$i; done
tar zxvf bash-4.3.tar.gz
cd bash-4.3
#apply all patches
for i in $(seq -f "%03g" 0 25);do patch -p0 < ../bash43-$i; done #build and install ./configure --prefix=/ && make && make install cd .. cd .. rm -r src

Mysql InnoDB CrashRecovery

Author: Stephan  |  Category: Linux, Netzwerk

Symptom

Mysqlserver is suddenly stopping (crashing) . Even after a restart the server ist stopping after a few seconds again.

In /var/log/daemon.log there are large mysql Hexdumps as well as the assumtion that the database is physically corrupt.

prerequirement for a successfull repair

In case of such an emergncy you need an uptodate mysqldump of all databases. If not, you have no change to fix the database without the loss of data

solution

restart Mysql with the Parameter

innodb_force_recovery = 4

drop your databases ( with drop Database ) as long till mysql starts normaly

If all Databases are part oft the problem, then you have to deinstall mysql , purge /var/lib/mysql , an then install mysql again.

At the latest mysql should run normal now. Time to use the dump to restore your databases.

NFS mounting incorrect NFS export

Author: Stephan  |  Category: Linux, Netzwerk

Symptom

There are two NFS exports in the /etc/exports file on the NFS server. Regardless of which export is mounted, only the first export actually gets mounted.

The /etc/exports looks like this

#cat /etc/exports
/exports/export1 *(rw,sync,insecure,root_squash,no_subtree_check,fsid=0)
/exports/export2 *rw,sync,insecure,root_squash,no_subtree_check,fsid=0)

Solution

Reason for this stange behaviour is the Duplicate fsid setting. You can only have one fsid=0 per NFS Server. Change the second fsid to a small integer and everything is working as designed

# cat /etc/exports
/exports/export1 *(rw,sync,insecure,root_squash,no_subtree_check,fsid=0)
/exports/export2 *rw,sync,insecure,root_squash,no_subtree_check,fsid=1)

Die Wichtigsten 5 Paradigmen zur Fehlersuche in technischen Systemen

Author: Stephan  |  Category: Linux, Netzwerk, Sonstiges

In einer Fortbildung mit Kollegen haben wir heute die in unseren Augen die 5 wichtigsten Paradigmen bei der Fehlersiche in technischen Systemen zusammengeschrieben.

1.) Welche Komponenten sind alles beteiligt ?
2.) Kann ich den Fehler reproduzieren, wenn nein , warum ?
3.) Welche Komponenten kann ich systematisch ausschliessen?
4.) Wenn ich nicht zum Ziel komme müssen alle Prämissen geprüft werden.

und in einem Systemn, an dem mehr wie ein Techniker beteilgt ist sollte folgende Frage nicht fehlen:

Wer kann mich unterstützen ?

adblock-plus undercover – Eine Linkempfehlung

Author: Stephan  |  Category: Netzwerk, Sonstiges

Einen spannenden Fall, den ich vor kurzem via Twitter gelesen hatte , möchte ich hier weiterreichen.
Es geht um die riesigen Wirtschaftsinteressen, die hinter dem kostenlosen Browserplugin ad-block plus stehen.

http://www.mobilegeeks.de/adblock-plus-undercover-einblicke-in-ein-mafioeses-werbenetzwerk/

agent forwarding and sudores

Author: Stephan  |  Category: Linux, Netzwerk

for a correct agent forwarding in a sudoers environment you need two prerequesites

1.) In /etc/sudoers you need
Defaults env_keep = “SSH_AUTH_SOCK”

2.) DON’T call sudoers this way
“sudo su -“

Heartbeat like IP-Takeover

Author: Stephan  |  Category: Linux, Netzwerk, Sonstiges

Heartbeat Based Clusters are a fine thing but even clusters can crash. In this case you may have to bring up the ressources manually. With

ip -f inet addr add 192.168.1.1/24 brd 192.168.1.255 dev eth1

you can add the VIP manually but you have to clean the ARP Caches manually to have sucess. You clean the ap caches with this command:

arping -c 4 -A -I eth1 192.168.1.1

A strange rdiff_backup error and its solution

Author: Stephan  |  Category: Linux, Netzwerk, Sonstiges

After a longer migration of rdiff_backup repository (4 Weeks) I got this strange error

[03/20/13 13:31:44] Exception 'Found too many current_mirror incs!' raised of class
'':
[03/20/13 13:31:44] File "/usr/lib/pymodules/python2.6/rdiff_backup/Main.py", line 304,
in error_check_Main
[03/20/13 13:31:44] try: Main(arglist)
[03/20/13 13:31:44] File "/usr/lib/pymodules/python2.6/rdiff_backup/Main.py", line 324,
in Main
[03/20/13 13:31:44] take_action(rps)
[03/20/13 13:31:44] File "/usr/lib/pymodules/python2.6/rdiff_backup/Main.py", line 280,
in take_action
[03/20/13 13:31:44] elif action == "backup": Backup(rps[0], rps[1])
[03/20/13 13:31:44] File "/usr/lib/pymodules/python2.6/rdiff_backup/Main.py", line 337,
in Backup
[03/20/13 13:31:44] backup_final_init(rpout)
[03/20/13 13:31:44] File "/usr/lib/pymodules/python2.6/rdiff_backup/Main.py", line 501,
in backup_final_init
[03/20/13 13:31:44] checkdest_if_necessary(rpout)
[03/20/13 13:31:44] File "/usr/lib/pymodules/python2.6/rdiff_backup/Main.py", line 916,
in checkdest_if_necessary
[03/20/13 13:31:44] need_check = checkdest_need_check(dest_rp)
[03/20/13 13:31:44] File "/usr/lib/pymodules/python2.6/rdiff_backup/Main.py", line 907,
in checkdest_need_check
[03/20/13 13:31:44] assert len(curmir_incs) == 2, "Found too many current_mirror incs!"

What happened ? Because of the very long Migration of the repository more than one current_mirror Statusfiles had been synced.
In the errormessage above it had been two files .
how to solve ?
Simply delete in all rdiff_datadirs all current_mirror* files except the newest one and your rdiff_backup is running again

Apache solr search for my blog

Author: Stephan  |  Category: Linux, Netzwerk, Sonstiges

After a long time I finished Integrating my blog into an Apache Solr Search Server