Networknet.nl

Today I hit my first problem when creating a new JeOS virtual appliance based on Ubuntu. After I mount and extract the VMWare tools; I tried to run the ./vmware-install.pl script and got the warning below.

/tmp/vmware-tools-distrib/vmware-config.pl
The following VMware kernel modules have been found on your system that were
not installed by the VMware Installer.  Please remove them then run this
installer again.

vmnet
vmhgfs
vmblock
vmmon

Execution aborted.

I did not install any VMWare tools before, but somehow these VMWare modules were included in the ubuntu-7.10-jeos-i386.iso file. Go to following directory and remove the mentioned directories.

cd /lib/modules/2.6.22-14-virtual/ubuntu
ls
sudo rm -r vmnet/
sudo rm -r vmhgfs
sudo rm -r vmblock
sudo rm -r vmmon

That’s all. Rerun the install script.

I have two identical Cacti (0.8.6h) virtual machines and both based on CactiEZ version 1 installation. I have been migrating the virtual machines between different Vmware hosts  and at some point my second Cacti appliance did not update the RRD graphs and using the web interface was very slow. I tried rebooting; changing parameters in the Configuration Settings but noting helped to solve the issues. I even tried to bring the virtual machine on the original vmware host and I had same symptoms.

I logged in with a SSH session and went to my cacti database folder. I did a list and reviewed the file structure.

Shell> cd /var/lib/mysql/cacti
Shell> ls -all
-rw-rw—- 1 mysql mysql 303340164 Apr 15 09:32 poller_output.MYD
-rw-rw—- 1 mysql mysql 119115776 Apr 15 09:32 poller_output.MYI

As you can see the poller_output.MYD and poller_output.MYI are too huge and mysqld used a lot of resources. Total size of two files 366MB and my physical memory assigned to this virtual machine was only 384MB.

I read couple posts on the Cacti forms and decided to clean up my poller_output table from shell.

Shell> mysqld
mysql>use cacti mysql>
delete from poller_output;
mysql>exit

Read the rest of this entry »

Two hours ago I upgraded Networknet.nl with Wordpress version 2.5 and immediately after the php binaries and database were upgraded I hit my first problem. Within my Widgets Design Sidebar I have the Pages widget which shows all pages on the right side of my blog starting with “About me”.. VMware. For some topics I created pages instead of Wordpress posts and linked those pages to a parent page; these two pages were excluded from “Pages Widget”. After the upgrade my pages widget become really huge showing 30+ pages. I did a search on Wordpress support site and someone else already reported this problem; no response yet except me on this post. 

I didn’t want to downgrade my new Wordpress version and therefore I used this method to query the parent pages and exclude them manually without doing the PHP code research how to exclude them.

• Open phpmyadmin
• Select your wordpress database
• Open wp_posts and click SQL

• Paste this sql query

SELECT id
FROM ‘wp_posts’
WHERE ‘post_type’ = ‘page’
AND ‘post_parent’ NOT LIKE ‘0′

• Click GO button and click Export; I export to CSV to excel and copied all cells into Notepad.

• Paste these pages into the exclude for your pages widget.

Dirty solution but I solved my problem in less than 5 minutes for now…

At some point my lab Openfiler was assigned static ip address and the iSCSI target stopped to function; also it was not reachable by the web interface at tcp port id 446 anymore. Several reboots did not help to solve the problem.

What happened? The network interface configuration was changed. One of the eth config files was changed from DHCP to a static ip address.

vi /etc/sysconfig/network

[root@net-nas-01 sysconfig]# cat network
NETWORKING=yes
HOSTNAME=net-nas-01.networknet.nl
GATEWAY=192.168.100.254

vi /etc/sysconfig/network-scripts/ifcfg-eth0

[root@net-nas-01 network-scripts]# cat /etc/sysconfig/network-scripts/ifcfg-eth0
DEVICE=eth0
IPADDR=192.168.100.12
BOOTPROTO=static
NETMASK=255.255.255.0
NETWORK=192.168.100.0
GATEWAY=192.168.100.254
BROADCAST=192.168.100.255
ONBOOT=yes

Openfiler service failed if you checked the services (# service –status-all). The problem is Openfiler service checks the FQDN in the hosts file located in the etc directory.

[root@net-nas-01 ~]# service openfiler –full-restart
Stopping openfiler:                                        [FAILED]
Starting openfiler:                                        [FAILED]
[root@net-nas-01 ~]# service openfiler -configtest
Usage: openfiler {start|stop|restart|condrestart|reload|status|fullstatus|graceful|help|configtest}
[root@net-nas-01 ~]# service openfiler configtest
httpd: Could not determine the server’s fully qualified domain name, using 127.0.0.1 for ServerName
Syntax OK
[root@net-nas-01 ~]# cd /etc
[root@net-nas-01 etc]# cat hosts
# Do not remove the following line, or various programs
# that require network functionality will fail.
127.0.0.1               localhost.localdomain localhost

Go and modify the hosts file. To modify the hosts file run this command –> vi /etc/hosts and make sure the FQDN names for 127.0.0.1 and the static ip address are set correct.

[root@net-nas-01 etc]# service openfiler –full-restart
Stopping openfiler:                                        [FAILED]
Starting openfiler:                                        [  OK  ]
[root@net-nas-01 etc]# cat hosts
# Do not remove the following line, or various programs
# that require network functionality will fail.
127.0.0.1                 net-nas-01.networknet.nl
192.168.100.12            net-nas-01.networknet.nl
[root@net-nas-01 etc]#

For more information go to Openfiler website.

If you are managing and monitoring a network you will probably be interested to keep an eye on the latency of your network links. Especially for those links which are connected with a dedicated internet connection and a IPSec VPN tunnel to the datacenter. Latency is predictable but for non private IP VPN links without reserved bandwidth and QoS/CoS it sometimes may help solve some problems or rethink and discuss high latency with the service provider. With dedicated Framerelay/Leased Line/IP VPN network links you can agree with the service provider on the different latency values and have SLA’s in place.

Latency in a packet-switched network is measured either one-way (the time from the source sending a packet to the destination receiving it), or round-trip (the one-way latency from source to destination plus the one-way latency from the destination back to the source). Round-trip latency is more often quoted, because it can be measured from a single point. Note that round trip latency excludes the amount of time that a destination system spends processing the packet. Many software platforms provide a service called ping that can be used to measure round-trip latency. Ping performs no packet processing; it merely sends a response back when it receives a packet (i.e. performs a no-op), thus it is a relatively accurate way of measuring latency.

http://en.wikipedia.org/wiki/Latency_%28engineering%29

I am using CactiEZ v0.3 to address this issue and monitor the ping latency. Those hosts may or may not be a SNMP enabled device. If the router is service provider managed than you probably won’t be able to get any access to SNMP, but for this we don’t need it.

a) Create new device in Cacti Management console. See example for www.networknet.nl

b) Choose for Availability “Ping” and use “ICMP” ping as method. Click Add and reopen the device.

c) Locate Associated Graph Templates and choose “Unix-Ping Latency” in the Graph Templates. Click Add.

d) Click “*Create Graphs for this Host”

e) Select “Create: Unix - Ping Latency” graph template and click Create.

f) Choose Red as legend color and click Create. + Created graph: www.networknet.nl - Ping Latency is now created.

i) Create new Graph Tree and add the new created graph for the network latency. See my example.

The result:

After installing CactiEZ 0.30 and you start using the Device Tracking Management (MacTrack plugin) area you may receive the following error when creating new site or device.

Error: save Failed.

 

This error is being generated by the Cacti system because there are no tables in the Cacti database.

Login on the console or with a ssh shell and execute the following commands

cd /var/www/html/plugins/mactrack

php database_upgrade.php

I have couple other issues after scanning a cisco switch, which I will need to follow up on the cactiusers.org forum. The device type list is empty in the newest version of CactiEZ and you will need to populate this yourself now.