jump to navigation

SQL [sqlservr.exe] Memory Consumption on Small Business Server July 7, 2009

Posted by Gomez in IT, Microsoft, Performance & Alerts.
Tags: , , , ,

I started receiving alarms that pages/sec on my client’s SBS Server was 6952.92 pages/sec. Please note that Microsoft Best Practices recommends no more than 20 pages/sec.
I found that sqlservr.exe was consuming in excess of 1.5GB of memory. This is been a known issues for Servers running Small Business Server, I will walk you through the process of reducing usage for this memory thirsty process.

1. Open Task Manager > Processes Tab, View Menu > Columns, select PID check box.
2. In the list of processes locate the sqlservr.exe process that is consuming the most memory (there might be multiple instances of sqlservr.exe) take note of the PID number next to the memory thirsty process.
3. Open a command prompt and issue the following command without the quotes,
    “tasklist /svc” (this is used to display running processes (PIDs) and applications)
     next to the PID number previously identified in taskmanager take note of the service name.
   The service would be MSSQL$SBSMONITORING.

4. At the command prompt issue the following commands, after each command press enter;
 Command 1: osql -E -S SERVERNAME\sbsmonitoring  
(replace SERVERNAME with the hostname of your server, sbsmonitoring is the service name )
 Command 2: sp_configure ‘show advanced options’,1
 Command 3: reconfigure with override
 Command 4: go

Command 1: sp_configure ‘max server memory’, 200
(200 represent memory size in MB, replace 200 with any memory size you wish, default is 2147483648)
 Command 2: reconfigure with override
 Command 3: go

 After the last command “Go” you should see a confirmation that DBCC execution has completed. You can just type exit at the command prompt to exit.Open Task Manager > Process Tab and review the sqlservr.exe and confirm that memory usage has decreased.
Best of luck..

Guyana Linux Users Group Snort installation July 7, 2009

Posted by daakeung in IT, Linux, Uncategorized.
Tags: , , , , , ,
add a comment

Every month now, GLUG meets up to discuss and share ideas.  Last GLUG meeting I volunteered to do an installation of SNORT on CENTOS and live demonstration.  It was a privilege to contribute to such a growing community in Guyana. Also Mike Harrison of Utiliflex took part to offer us his valuable experience and motivation.

Download: snort_installation_presentation

Quick guide how to Setup a linux dhcp server (Centos 5.2) July 3, 2009

Posted by daakeung in Uncategorized.
Tags: , ,

Quick guide to setup a linux DHCP server on a network, you can modify this to work with your own design.

[root@dhcp ~]# yum install dhcp

[root@dhcp ~]cp /usr/share/doc/dhcp-3.0.5/dhcpd.conf.sample /etc/dhcpd.conf
cp: overwrite `/etc/dhcpd.conf’? y
[root@dhcp ~]vi /etc/dhcpd.conf

ddns-update-style interim;
ignore client-updates;

subnet netmask {

# — default gateway
option routers        ;
option subnet-mask    ;

#option nis-domain               “akeung.com”;
option domain-name              “akeung.com”;
option domain-name-servers;

option time-offset              -18000; # Eastern Standard Time
#       option ntp-servers    ;
#       option netbios-name-servers;
# — Selects point-to-point node (default is hybrid). Don’t change this unless
# — you understand Netbios very well
#       option netbios-node-type 2;

range dynamic-bootp;
default-lease-time 21600;
max-lease-time 43200;

# we want the nameserver to appear at a fixed address
#        host ns {
#                next-server marvin.redhat.com;
#                hardware ethernet 12:34:56:78:AB:CD;
#                fixed-address;
#        }

Then esc, press


To start the service :

service dhcpd start

To enable it at bootup :

chkconfig dhcpd on

You can locate the leases at :


Also, if a reservation is needed. Simply add the following to /etc/dhcpd.conf

host daakeung {
option host-name “daakeung.akeung.com”;
hardware ethernet 00:00:00:00:00:00;      #<—-Replace with Client’s MAC address

Cisco Dual Wan Fail-Over using SLA Tracking July 3, 2009

Posted by daakeung in Cisco, IT.
Tags: , , , , ,
Maintaining a high availability connection is critical in any modern network infrastructure.  Some would have access to the internet via multiple ISPs.
In most cases where the ISPs would not peer via any routing protocol, customers are left to setup fail-over with static routes.
There are two types of failures:
  1. Physical : the router would detect one of its interface not connected, invalidated any next hop routes through it. In turn move to another valid route with a higher metric.
  2. Logical : the router has it’s interface online, but somewhere along the path it has to take to reach the destination is unavailable.
Basic router commands with higher metric will not fail-over with a logical failure.
Using SLA Tracking, we can accomplish this.
Consider the following diagram and configuration:
Cisco Dual WAN

Cisco Dual WAN

Cisco IOS used: c181x-adventerprisek9-mz.124-24.T

! Identify the SLA.
ip sla 10

! Set up the IP address to ping, in this case yahoo.com ip.

! Set how long in milliseconds to wait for a reply.
timeout 1000

! Repeat Rate.
frequency 3

! Start SLA 10 from now to forever.
ip sla schedule 10 life forever start-time now

! Setup track 10 to sla 10 for reachabilty.
track 10 ip sla 10 reachability

! Setup route for testing reach ability to the internet.
ip route

! Setup primary link with track 10 parameter.
ip route track 10

! Setup fail-over link with higher metric.
ip route 20
Depending on your configuration, you will need to setup NAT on both interfaces.