|
November 2010 |
Introduction:
The only RAID management capability that Openfiler has using the web interface is being able to manage logical RAID volumes created within Openfiler. However, if you are using Adaptec RAID you are still able to manage your RAID volumes by taking the following steps.
Setup:
The first part involves installing the ASM agent locally on the Openfiler server; the second part involves installing ASM on a remote system and connecting to the ASM agent on the Openfiler server. Some familiarity with Linux based operating systems is recommended as most of this involves the command line. In this example the 64Bit version of Openfiler is used, the procedure is
not that different if you are using the 32Bit version and I will point out when alterations need to be made.
1) Start by logging on locally to the Openfiler server; we will begin by updating Java to the most recent release available from the repository:
conary update sun-jre
If successful you should get an output similar to the following:
2) Next we need to make a note of the Java directory name; enter the following (substitute "lib64" for "lib" if you are using 32Bit Openfiler):
ls /usr/lib64/jvm
Make a note of the output, in this example the Java folder was “sun-java-5.0u17”
3) Next we will download the ASM installation package using the wget command:
wget http://download.adaptec.com/raid/storage_manager/asm_linux_x64_v6_50_18579.tgz
The address will be different depending on the version of ASM you are downloading, at the time of writing this guide build 18579 is the most recent release and is compatible with series 2,3 and 5 cards. If you have used the correct link you should see it begin downloading the file and the completed output should resemble the image below:
4) Next we will extract the .rpm installtion file from the archive and move it to the root directory; by default when you log in you will be in the root user home directory. Begin by extracting the archive:
tar –xf asm*tgz
Two folders should now be present, “cmdline” and “manager”; we are going to copy the .rpm file from the manager folder to the root directory:
cp manager/Stor*rpm /
You can now clean up the extracted files:
rm manager –R –f
rm cmdline –R –f
5) Next we need to extract the storage manager files from the rpm package into their proper directory. Begin by moving to the root directory; then we will convert the rpm package to a more accessible archive type and at the same time extract the files from the newly created archive into place:
cd /
rpm2cpio Stor*rpm | cpio –idumv
The verbose switch is used for the cpio command so a list should be printed showing the files that have been extracted similar to the following:
Clean up the rpm package after successfully extracting the ASM files:
rm Stor*rpm –f
6) Now we need to move around and edit a few of the ASM files; begin by moving to the Storage Manager directory and editing the Storage Agent config file:
cd /usr/StorMan
nano +135 StorAgnt.sh
You should appear on or near the line you need to edit in the config file, you want to point it to the recently updated Java directory you made a note of earlier. Edit the following line (line 135 with build 18579):
JAVA_HOME="."
to
JAVA_HOME="/usr/lib64/jvm/sun-java-5.0u17"
or if you are using 32Bit
JAVA_HOME="/usr/lib/jvm/sun-java-5.0u17"
See image below for reference:
To save your changes and exit from nano press CTRL+X , then Y to confirm saving and then hit ENTER to confirm the file name.
7) The final step before starting up the agent is to move the ASM executable into place and to change access privileges to it and another file. It is assumed for the following commands that you are still in the ‘/usr/StorMan’ directory.
cp stor_agent /etc/init.d/
chmod +x /etc/init.d/stor_agent
chmod 0774 libSTORUTIL.so
8) Now if everything has been done correctly the ASM agent should start successfully:
/etc/init.d/stor_agent start
You should see the following:
This means that the agent has started, however, if something was not configured correctly the agent will stop shortly after starting. To check that it is still running we will filter the output from the processes command and you should see an entry for both the storage manager and the instance of java it is using:
ps a | grep –i stor
If the output is not similar to the following something has not been configured correctly:
9) Now we are going to move to a remote client system to download and install ASM. In this example a Windows system is used, the installation is very straight forward. After starting up ASM and logging in go to the ‘Action’ menu at the top and select ‘Add Managed System’. In the new window enter the IP address of the Openfiler server as well as the root username and password and click ‘Connect’.
A new system should appear on the list in the left pane; select it and you should now able to manage the RAID volumes as if they were local.
As you can see on the left side of the picture, the hostname for the server is “openfiler” as I had previousely changed it to this. By default it is “localhost” and this causes ASM to not send email notifications. To change the hostname of the server if you have not already, log into the web interface and under the ‘System’ tab change the hostname and click ‘Apply’.
You may need to restart the server, but hold off on doing this until completing the next step as it requires a restart aswell. You may also need to remove and re-add the managed system from the remote client ASM for it to see that the hostname has been changed.
10) The final step is to set the ASM agent to run automatically on startup, we are going to use chkconfig to do this:
chkconfig --level 2345 stor_agent on
Restart the system and if everything has been done correctly you should still be able to connect ASM as well as setup and send test email notifications (follow this guide to set up email notifications).
Email notifications are handy as you are able to receive email alerts whenever an important RAID event occurs so you can bring the system back up to full data redundancy as quickly as possible.
|
|
|
November 2010 |
Introduction:
Milestone are one of the leaders of the IPVS software market, their flagship video surveillance software being XProtect. The corporate edition is their most powerful suite as it adds a central management server to more easily manage an unlimited amount of cameras and recording servers.
Setup:
This guide will follow the setup of a very basic IPVS layout, the procedures can be applied to a scaled up network, however extra configuration would be needed. You will need to go through the Administrator’s Manual as there are many settings and features not mentioned in this guide. A basic layout diagram is below:
There are 4 main parts to this procedure: Installing the Management and Map services on the Main server; installing the Recording service on the secondary server, installing the Management and Smart clients on the client system and then basic configuration.
1) Firstly make sure all the systems being used have at least .NET framework 3.5 SP1 installed as it is required for most of the Milestone installations.
2) On the main server run the Management Server installer; the install window will show you the three steps to installing the management server: IIS install, management database install and management service install. If you already have IIS or SQL server installed on the system select the ‘Advanced’ installation as you will need to set it up in a way that doesn't interfere with existing services/applications using them. If you have neither installed select 'Typical'.
3) After the main installation steps have finished it will prompt you to install the Map server and the Service channel; in this example I will install both on the main server but you can choose to install them on a different system if you wish to. Their installation requires no input.
4) Once this has completed a webpage will show up; this is the Admin application page, published from the main server, that other systems can connect to it to download various installers. On your secondary server open a web browser and enter the address “http://[server_IP
_address]/installation/admin” to get to this page. Select and download the recording server, run the installer once downloaded.
5) You can again select the ‘Typical’ installation option, put in the management server IP address and video storage location when asked.
6) Finally from the client system go to the main server’s installer page and download the Management Client. There are two types of clients: the Management Client controls and configures the whole IPVS setup, this is what the administrator will use; the smart client is for IPVS monitoring, this will be what the video surveillance operators will be using.
7) Open up the management client and login using valid Administrator credentials for the management server. The first step is to set up the Recording Server and camera. Go to ‘Servers>Recording Servers’ on the left pane; if the IP addresses are set correctly and the servers can communicate properly you should see the hostname of the recording server in the middle pane. Right click it and select “Authorize Server”; you should now be able to select the different setting tabs from the right pane to configure the Recording Server such as storage, archiving and failover servers.
8) The next important step is to add cameras to the recording server; right click the server’s name and select “Hardware Detection Wizard”. Follow through the wizard, you can either specify the IP addresses of each camera to add (Assisted) or choose to automatically scan IP address ranges for supported cameras (Automatic). Once it has found and listed the available cameras you can set the username/password, IP address/port number and the hardware driver that will be used for each camera.
9) You can now expand a list from the Recording server’s name in the middle pane and select the cameras you have added. You can then expand each camera and select each sub-device to enable and configure (one camera device may have multiple sub-devices such as the actual camera feed, the microphone, speaker, etc). A number of sub-devices are listed by default regardless of whether they are present on the camera.
10) Select the 'Camera' sub-device and you will see some configuration tabs appear in the right pane; from here you can configure most of the camera functions such as stream quality, recording quality and motion detection. Enabling recording on a camera does not by itself mean that the video feed will be stored at all times, you also need to create Rules to determine when to store video feeds (this will be explained later on.)
11) In order to manage a large number of devices you will need to use Device Groups. Using device groups you can configure multiple cameras/devices at once and this can also be used to apply things such as recording rules to the whole group instead of each camera. Select ‘Devices>Cameras’ from the left pane and then in the middle pane right click ‘Cameras’ and select ’Create New Device Group’.
12) Right click the newly created device group and select ‘Edit Device Group Members’. On the left side select the Recording Servers tab, select the Camera from the list and click add and accept when finished. To configure a group, select it on the middle pane and use the configuration tabs on the right, you will notice that things such as motion detection are only available when configuring individual cameras.
13) By default a motion detection Rule is present, this means if you enable recording and motion detection on a camera it will begin recording whenever it detects something. If you want to always be recording you will need to create a new rule. Go to ‘Rules and Alerts>Rules’ on the left pane, right click the middle and select ‘Add New Rule’.
14) The rule creation wizard is fairly straight forward; below is a rule list which will have every camera in the ‘Cameras’ device group recording all the time. Each line represents an option selected in the wizard and in order to specify which device/group to use in the rule click the underlined blue text in the rule list to bring up a selection window. You can also specify certain time frames to always record in (a certain time period each day, certain day/s each week or custom timetable profiles you can create in another section called Time Profiles).
15) Browse to "http://[server_IP_address]/installation" ; download and install the Smart client. Open up the smart client, log in with valid Administrator credentials. In this example we are using the Administrator account but you are able to set mulitple View Groups in the management client in order to limit each user or groups access to the cameras/camera groups you allow them to.
16) Select the Setup tab; by default there is the ‘Private’ view group. Right click the Private group and select ‘New Group’; right click the newly created sub-group and click the ‘Create New View’. The view grids will be the camera layouts that the IPVS monitoring staff will see; as there is only one camera in this example we can select the 4:3> 1x1 layout.
17) To add a camera to the layout, select and drag it from the menu on the left side to the empty space on the right, you should now see the live feed from the camera. From here you can also setup layout maps; create a new view and drag the map icon from the left menu onto it. Select the picture to use and then use the toolbar that appears to add devices and other things to the map (see the Administrator Guide for more configuration details).
18) You will now be able to watch live and recorded video feeds from the Live and Playback tabs. The controls are very user friendly and do not take long to pick up at all.
Notes) Make sure all servers are time synchronized; a time sensitive token system is used between the management and recording servers and out of synch times can affect usability (different time zones are OK).
If you are having communication problems between your servers this may be due to certain needed ports being closed. There is a list of the ports needed to be opened on page 26 of the Administrator’s Manual .
This is a very brief overview to give you an idea of how the corporate edition is set up. There are many other features and settings to have a look at and you will need to read through the Administrator Manual to get the most out the software.
|
|
|
October 2010 |
Introduction:
The installation for ESXi is much simpler and faster than ESX as there is no need to configure things such as partitions for the service console. This guide will go through the installation from an ESXi installation disc; there is only a text based installation available on the disc.
Setup:
While there is not an option to set the datastore’s block size during the installation you are able to delete and recreate the datastore from the remote client after the installation where you are able to select which block size you want.
1) Choose the installer from the boot menu; on the first page press enter to choose to install ESXi. There is also a repair option on this page for existing installations.
2) Press F11 to accept the EULA
3) Select the storage device to install ESXi on and confirm that you want to overwrite existing installations if there are any. You can press F1 when highlighting a storage device to see more information about it. You will also get a warning that any existing data will be overwritten ont he selected storage device, press F11 to accept.
4) The installation will only take a few minutes, press enter to reboot the system once it is completed.
5) The screen shown after restarting will have the server's IP address that you can use to access it through the remote client. You are also able to press F2 when accessing the server locally to be
able to perform limited configuration such as network settings.
6) Most configuration will be done through the remote client. Go to the server's address in a web browser and it will have links to download the remote client if you don't already have it installed.
You have a 60 Day evaluation period to begin with, you will need to enter the license given to you by VMware in the configuration section before this period is up or a lot of its features will become disabled. To request a license you need to go to VMware's website and register or login request one.
|
|
|
October 2010 |
Introduction:
This guide will follow the installation of ESX4.1 using the graphical installater; while it will be a basic setup with a single storage device it can be used for a more complex configuration with a few changes. Each step in the guide will usually represent one page of the installer.
Setup:
For some of the steps, such as changing the partitioning from the default, it is recommended you look into what would be best for the specific server’s role as these changes may not be necessary in all cases.
Before starting the installation you need to consider how you want your datastore to be set up. The graphical or text based ESX installation creates a VMFS partition (datastore) with a 1MB block size using all the available space on the selected storage device by default
(you cannot change this from the interface). This means file (i.e. virtual disk) size in the default datastore is limited to 256GB. The problem is the service console is itself on a virtual disk stored on this partition making it difficult to delete and recreate the datastore with the desired size and block size after installation.
You can get around this by installing ESX 3.5 with a larger block size and upgrading to 4.1. Alternatively you can make some changes to the installer configuration files; this guide tells you how to do both.
This would be less of a problem if you are using shared storage for virtual disks as the locally created datastore would most likely only be for the ESX installation anyway, meaning you wouldn’t need larger file size limits. If you are using local storage only, another solution could be to create two RAID volumes: a smaller one for the datastore which only contains the service console virtual disk and the second larger volume for virtual disks can be partitioned after installation where it allows you to select a larger block size.
1) After booting into the installation disc, select “Install ESX in graphical mode”
2) If you are changing the installer configuration files to increase the datastore block size this should be done straight after the GUI appears.
3) The first part of the installation is straight forward: accept the EULA; select US English for the keyboard; load custom drivers if necessary (ESX supports a variety of hardware by default so this may not be needed); enter your key or skip it to enter it after installation.
4) Select the network adapter to use for the service console; enter VLAN settings if necessary.
5) Enter static network settings if necessary, you will need to do this if you want to set the system’s hostname at this point (the host name can be set later from the remote client if you select DHCP).
6) Select ‘Advanced Setup’ to be able to change the partitioning or set up datastores on multiple volumes.
7) Select the storage device to use; if it detects an existing datastore it will ask you if you want to keep or overwrite it. If it does not detect one it will give you a warning message that any existing data on the storage device will be wiped.
8) The default datastore name is “datastore1”, at this stage you can also choose to setup other datastores on other storage devices.
9) The next step is where you can change the partitioning of the service console. Below is a picture of the changes made and below that is an explanation for each change :
swap partition - this can range from 600MB to 1600MB, you can increase this to the max if desired.
The following are to prevent the root partition from becoming full which can cause problems with the service console
/var -changed from just /var/log as there are also system logs stored in /var.
/tmp - temporary file storage for things such as decompressed archives during installations.
/home - users home directories stored here.
10) Select the correct timezone from the map, on the next page set the time or specify an NTP server to use.
11) Enter the password for the root account, this needs to be at least 6 characters long and there is no complexity requirement. You can also add other management accounts at this time.
12) The final page before beginning the installation is the verification page; here you will now see two additional service console partitions: /boot (self explanatory) and the vmkcore partition that temporarily stores log and error information if the kernel crashes. All up the service console virtual disk will be around 11GB.
13) Restart the system after the installation is completed, the page that appears during start up will tell you what IP address to access if you have chosen to use DHCP. You can also access the service console command line locally from here.
14) Go to the server address in a web browser where you will be able to download the vSphere remote client.
If you are using Adaptec RAID it is recommended that you follow this article to set up the Adaptec Storage Manager agent for RAID monitoring purposes.
|
|
|
October 2010 |
Introduction:
This article was written as a result of a customer requirement to install an Intel SSD Drive into an older Compucon System. It represents a discovery process including pre-requesites, confirmation, problems that were encountered and resolutions to those problems.
1) What are the pre-requesites to support an SSD?
An SSD appears to the system as a normal SATA HDD however to get the full performance benefits and the full life expectancy of the SSD you should be using an OS that supports TRIM such as Windows 7 or Server 2008R2. TRIM support itself requires a SATA Controller running in AHCI Mode.
2) What is TRIM and why do we need it?
Read the following Windows 7 Support Q&A on Solid State Drives Read for explanation of TRIM function and it's benefits.
http://blogs.msdn.com/b/e7/archive/2009/05/05/support-and-q-a-for-solid-state-drives-and.aspx
IMPORTANT: With TRIM support undeleting a file may in fact be impossible.
3) How can I tell if TRIM is enabled under Windows 7?
Run the following in a command prompt (cmd.exe) that has Administrator rights.
fsutil behavior query DisableDeleteNotify
If the result is '0' TRIM is enabled.
4) How do I achieve TRIM functionality on OS's that don't support TRIM?
This is manufactuer dependent. In the case of Intel SSD's use Intel Solid State Drive Toolbox.
http://downloadcenter.intel.com/Detail_Desc.aspx?agr=Y&DwnldID=18455
This tool allows you to force a TRIM function on an OS, such as Windows XP, that does not natively support and automatically perform the TRIM function.
Please read the following Q&A for additional information on the Intel Solid State Drive Toolbox.
http://www.intel.com/support/ssdc/hpssd/sb/CS-031221.htm
5) Is TRIM supported with Intel Rapid Storage in RAID Mode?
Yes, but only as a single drive & not in a raid volume, refer:
http://www.intel.com/support/chipsets/imsm/sb/CS-031491.htm
I think this is an issue we need to monitor as you lose some of the benefit of the SSD without TRIM & there may be compelling reasons to have a RAID-0 or RAID-1 using SSD's. Future versions of the Intel Rapid Storage Technology may add support for this.
6) Do we defrag SSD's?
7) How do I Reset or Factory Erase an Intel SSD?
HDDErase v3.3 uses the built in SE (Secure Erase) feature of ATA Firmware. On an Intel SSD this causes the Flash Block to be completely erased. This was the tool Intel supplied with all their Intel SSD Evaluation Kits to allow reviewers to erase all blocks to restore full performance prior to TRIM support being available. The MB BIOS MUST be set to IDE Mode. I found this did not work on a P35 based MB (P5K-E) but on an older 865 based MB where the BIOS allowed me to set the SATA Port 0 as Primary Master worked.
Download HDDErase v3.3 from here:
http://www.iishacks.com/2009/06/30/how-to-secure-erase-reset-an-intel-solid-state-drive-ssd/
Please note that the latest Official HDDErase v4.0 is reported as not supporting SSD's any more, this has not been confirmed.
HDDErase v4.0 can be downloaded from here:
http://cmrr.ucsd.edu/people/Hughes/SecureErase.shtml
Update: v2 of the Intel SSD Toolbox referenced in Question 4 now has the ability to perform Secure Erases
8) What do I do if the BIOS Hangs when detecting the Intel SSD when in AHCI or RAID Mode?
This only seems to affect older MB's such as the P35 (P5K-E) and the 965 (P5B-E) with an earlier Intel Option ROM. Later model MB's like the P7P55D-E LX were not affected. The problem was traced to the fact the Intel SSD geometry was being Automatically detected as LARGE and not LBA. This was only visible when the SSD was connected to an older MB, like an 865, which has a BIOS that reports the detected Geometry & Mode.
IMPORTANT: This problem did not occur when set to IDE mode however in IDE Mode there is no TRIM support because you cannot use either the MS AHCI Driver or the Intel AHCI or RAID Driver which are prequesites for TRIM support.
Histroically we have had similar issues with IDE/ATA HDD's where the Boot Sector or Partition have been corrupted & the BIOS reports incorrect Geometry & Mode when the Device is set to Auto Detect.
The Soluton was to Factory Erase the SSD using HDDErase v3.3 after which the SSD correctly reported as LBA on the 865 MB and subsequently worked on the P5V-E & P5B-E MB's.
Fin.
|
|
|
|
<< Start < Prev 191 192 193 194 195 196 197 198 199 200 Next > End >>
|
| Results 1720 - 1728 of 2511 |
