GEO offers the ability to move past the single data center, allowing for multi data center High Availability (HA). Even when a primary site is down, traffic is diverted to the disaster recovery site. Also included in GEO is the ability to ensure clients connect to their fastest performing and geographically closest data center.
The GEO product is available in two forms:
A standalone GEO product
A Global Server Load Balancing (GSLB) Feature Pack that is part of the KEMP load balancer (LoadMaster) product
Throughout this document, when we refer to the â€œGEO LoadMasterâ€ we are referring to either the GEO LoadMaster or the LoadMaster with the GSLB Feature Pack enabled.
GEO has the same management interfaces as KEMPâ€™s Server LoadMaster hardware appliances, including all the foundation technology such as syslog logging, email notifications, interface bonding, and Gigabit support. GEO provides advanced application health checking, to ensure that unavailable services or data centers are not visible to clients. Health checking occurs at the site level, allowing for flexible decision making on when traffic should be diverted per Fully Qualified Domain Name (FQDN).
GEO can be deployed in a distributed (active/active) high availability configuration, with multiple GEO LoadMasters securely synchronizing information. Introducing GEO into existing Authoritative Domain Name Services (DNS) requires minimal integration work and risk, allowing you to fully leverage the existing DNS investment.
GEO securely and seamlessly integrates with core LoadMaster functionality to offer Real Server Load balancing, in which GEO uses local data center metrics provided by the LoadMaster, allowing clients to connect to the most available target. This is supported in both the GSLB feature pack and the standalone GEO product.
Currently, GEO only handles A (IPv4) and AAAA (IPv6) records.
This section describes how the GEO functionality typically works. Please note that within this configuration we are depicting the GEO LoadMaster as being located outside the data centers. Though this can be the case, typically the GEO LoadMaster would be located within one or more of the data centers.
1. An external client tries to connect to FQDN test.domain.com.
2. The external client checks its local DNS.
3. The local DNS forwards the request to the public DNS.
4. The Public DNS forwards the request to GEO as it is the authoritative DNS for this record.
5. GEO checks persistence and scheduling and decides which site to return.
6. GEO returns the IP address for the client that made the request (in this case; the public DNS).
7. The public DNS returns the results to the Local DNS.
8. The local DNS returns the result to the client.
9. The client connects directly to the site.
The above diagram could depict a couple of scenarios:
Two active/active sites with round robin/location/proximity scheduling
Two sites â€“ one active and one disaster recovery with fixed weight scheduling
GEO LoadMasters set up as partners in active/active mode. Please refer to the Distributed LoadMaster Partners section for further information on GEO partners.
By default, if the client is internal they are given an internal address. If the client is external, they are given an external address.
The sections below refer to various aspects of GEO deployment.
When using the GSLB Feature Pack, GEO can be enabled or disabled by clicking either the Enable GSLB or Disable GSLB menu option under Global Balancing in the main menu of the LoadMaster WUI. When GSLB capabilities are enabled on a LoadMaster, the Packet Routing Filter is also enabled and required. When GEO is disabled, it is possible to either enable or disable the Packet Routing Filter in System Configuration > Access Control > Packet Filter.
Refer to the sections below for a description of the different parts of the home screen.
After initially logging in to the LoadMaster, if Session Management is enabled - some login information is displayed:
The last login time and IP address of the current user
The number of successful logins by the current user in the last 30 days
The total number of failed login attempts by any user (including unknown usernames) since the last successful login
IP address: The IP address of the LoadMaster.
LoadMaster Version: The firmware version of the LoadMaster.
If the Allow Update Checks feature is enabled - when a new version of the LoadMaster firmware becomes available, a message is displayed at the top of the Home screen to inform you. To enable the auto-check feature, go to Certificates & Security > Remote Access.
Serial Number: The Serial Number of the LoadMaster.
Boot Time: The time of the last server reboot.
The WAF Status section will only appear on WAF-capable LoadMasters. For further information on WAF, please refer to the KEMP Web Application Firewall, Feature Description.
CPU Load: The percentage of load to the CPU of the LoadMaster appliances and to the CPU running a Virtual LoadMaster (VLM).
TPS [conn/s]: The total number of Transactions Per Second and the number of Secure Sockets Layer (SSL) transactions per second.
WAF Stats: Application Firewall Pack (AFP) status - shows the total number of handled connections, and the total number of incidents.
Net Load: Network load in megabits per second, shown for each configured interface.
CPU Temp.: Displays the temperature of the CPU.
The CPU Load and Net Load data is updated every 5 seconds.
Clicking the View License link will display model, subscription, subscription expiry and subscription feature details, such as the activation date and end date of the LoadMaster license.
If the subscription has expired, a message is displayed in the License Information section. To renew the subscription, please contact KEMP.
Upgrade: Upgrade the LoadMaster by buying a license from the KEMP purchase portal.
On the About LoadMaster page, you can view licenses for third party software that is used in the LoadMaster.
To view a license, click the View button next to the relevant item.
Other links are provided at the bottom of the home page:
Support & FAQ: A link to the KEMP Support site
Find Online Documentation: A link to the KEMP documentation page
When using the GSLB Feature Pack on a LoadMaster, the GEO-related options can be found by selecting the Global Balancing option in the main menu on the left of the WUI.
There is another GEO option which is not contained in the Global Balancing main menu option â€“ Use for GEO Responses and Requests. You can get to this setting by going to System Configuration > Network Setup and selecting the relevant interface.
By default, only the default gateway interface is used to listen for and respond to DNS requests. This field gives you the option to listen on additional interfaces. When this option is enabled, GEO also listens on any Additional addresses that are configured for the interface.
This option cannot be disabled on the interface containing the default gateway. By default, this is eth0.
LoadMaster High Availability (HA) pairs only listen on the shared interface.
LoadMaster connects to one or more networks. By default, a single interface (eth0) is used for DNS responses. In this documentation, eth0 is assumed to be used as the sole interface used for DNS responses.
In a one-armed configuration, the DNS responder service can be configured for any subnet. The LoadMaster connects to a Layer 2 network through a single interface, eth0.
If a firewall is already in place performing PAT to a DMZ in a non-routable (RFC1918) IP space (for example, 192.168.x.x or 10.x.x.x), please make sure a 1-to-1 PAT for port 53 UDP/TCP exists to the LoadMaster.
KEMP does not recommend a Layer 3 source IP NAT to the LoadMaster as it will mask source IP visibility during geographical coding operations, all devices before the LoadMaster should be transparent.
The LoadMaster(s) can be located on the DMZ with no large-scale network changes required. As shown in the diagram above, the default gateway of LoadMaster should point to the firewall.
When referring to client source IP, we are taking about the client resolver of the workstation, not the source IP of the workstation or its corresponding NAT to the Internet. This is an important concept to understand; client IPs are of the corresponding DNS resolver. The LoadMasterâ€™s geographical encoding operations are based on this client IP. A common deployment for client DNS resolvers is depicted in the diagram below.
In the illustration above, the following steps occur:
1. The client workstation asks the local DNS server for a translation of www.web.example.com.
2. The local DNS server forwards the request to an ISP or Internet DNS server.
3. The ISP/Internet server has the relevant A records and NS records pointing to the LoadMaster.
4. The GEO LoadMaster responds to the DNS query with an appropriate answer.
It is important to understand that it is step 3, within the described configuration, which defines the client IP address as presented to the LoadMaster, not step 1 or step 2.
If the firewall is transparent, the GEO LoadMaster will see the client as the ISP. If the firewall is NATing the traffic, the GEO will see the client IP address as the firewall.
The above diagram illustrates the difference between recursive and iterative DNS.
With recursive DNS:
5. An external client checks the local DNS server for the IP address of the FQDN.
6. If the Local DNS Server cannot provide the IP address, the local DNS requests the address from the ISP/internet DNS server.
7. If the ISP/internet DNS server cannot provide the IP address, it requests the address from the firewall.
8. If the firewall cannot provide the IP address, it requests the address from the GEO LoadMaster.
9. The return traffic sends answers back to each device along the chain in the network.
In recursive DNS, the GEO LoadMaster sees the client as the ISP server. Please bare this in mind when using location-based or proximity scheduling.
With iterative DNS:
10. The client checks the local DNS server for the IP address of the FQDN.
11. The local DNS server tells the client to contact the ISP/internet DNS server.
12. The client checks the ISP/internet DNS server for the IP address of the FQDN.
13. The ISP/internet DNS server tells the client to contact the firewall.
14. The client checks the firewall for the IP address of the FQDN.
15. The firewall tells the client to contact the GEO LoadMaster.
16. The client checks the GEO LoadMaster for the IP address of the FQDN.
17. The GEO LoadMaster answers the DNS query.
These are all separate connections.
Integrating the LoadMaster with your Authoritative DNS can be completed with only a few DNS records:
1. Create a new A record which is pointed to the LoadMaster, for example lm1.example.com. Create the corresponding PTR record for the reverse lookup by IP. Forward-confirmed reverse DNS support is required.
2. For each hostname that needs to be delegated to the LoadMaster, create a NS record and set the value to the A record created for the LoadMaster in the previous step, for example www.web.example.com to lm1.example.com.
3. When using GEO active/active configuration, repeat step 1 for the second LoadMaster using a unique hostname, for example lm2.example.com. Repeat step 2 using the second LoadMaster. This results in 2 NS records for www.example.com; one pointing to lm1.example.com and one to lm2.example.com.
Miscellaneous GEO parameters can be configured by going to Global Balancing > Miscellaneous Params in the LoadMaster WUI.
Configuration of global parameters controls the behavior of the entire LoadMaster. The Source of Authority (otherwise known as Start of Authority) information is not required for basic functionality; however, it is recommended to populate this metadata to accurately represent the LoadMaster DNS server.
Example Source of Authority values, and descriptions of each of the fields, are provided below:
The name of the zone when using DNSSEC
Source of Authority
The name of the domain owner
The name of the DNS server
Email address of the DNS administrator (who to contact if there is any issue with the DNS authoritative record)
TTL (Time to Live), which is measured in seconds, defines how long a DNS answer is valid for. This can be configured globally in the Miscellaneous Params screen, or on a per-FQDN basis.
Resource Check Parameters define the GEO health checking that occurs from LoadMaster to GEO Clusters and Real Servers. For more information on clusters, refer to the Scheduling Methods section. The minimum values for the Resource Check Parameters are as follows:
Check Interval: 9
Connection Timeout: 4
Retry attempts: 2
Depending on the behaviour, the wait can take up to (<Retries>+1)*<Timeout>:
If there is no response at all, it will wait the maximum duration as stated above
If the service returns a rejection of some form, it may take significantly less time to fail
Global Server Load Balancing (GSLB) Persistence, also known as â€˜Stickinessâ€™, is the property that enables all name resolution requests from an individual client to be sent to the same set of resources until a specified period of time has elapsed. This ensures that users are able to retrieve and interact with session-specific data. Stickiness can be set globally in the Miscellaneous Params section, or for an individual FQDN.
If connecting from a client to the GEO LoadMaster directly, the GEO LoadMaster keeps a persistence entry for the request. If connecting from a DNS server to the GEO LoadMaster directly, the GEO will keep a persistence entry for the request, as will the DNS server. When troubleshooting, ensure to set Stickiness to 0 and clear the DNS cache on the DNS server (Dnscmd /clearcache on Windows Server).
For further information, refer to the GEO Sticky DNS, Feature Description.
This section allows you to upload a GEO location database file and install an update to the geographical database.
A Fully Qualified Domain Name (FQDN) is the hostname in which you need to perform load balancing. The FQDN can be any hostname in the top-level domain or a hostname that is nested as a sub-domain. Each FQDN can be an A (IPv4) or AAAA (IPv6) record.
Each distinct hostname must be configured in the LoadMaster individually.
You can create an FQDN for www.example.com and also www.kemptechnologies.com.
To add an FQDN, follow the steps below:
1. In the main menu, select Global Balancing and Manage FQDNs.
2. Click the Add FQDN button.
3. Enter an FQDN name, for example www.example.com in the New Fully Qualified Domain Name textbox.
Wildcards are supported here, for example *.example.com matches anything with .example.com ending.
4. Click the Add FQDN button.
5. Click OK on the message that appears.
6. Select the relevant load balancing algorithm from the Selection Criteria drop-down list. For more information on the selection criteria, refer to the GEO, Product Overview.
7. If the Selection Criteria is set to Location Based, you can specify whether or not to allow Fail Over.
When the Fail Over option is enabled, if a request comes from a specific region and the target is down, the connection will fail over and be answered with the next level in the hierarchy. For example, if the Selection Criteria is Location Based - the country comes first in the hierarchy, then continent. If this is not available, the connection is answered by the nearest (by proximity) target. If this is not possible, the target with the lowest requests is picked. The Fail Over setting affects all targets.
Select the relevant options for the Public Requests and Private Requests drop-down lists.
8. The Isolate Public/Private Sites setting has been enhanced as of version 7.1-30. The checkbox has been migrated to two separate dropdown menus to allow more granular control of DNS responses. Existing behavior has been preserved and is migrated from your current setting, ensuring that no change in DNS responses is experienced. These new settings allow administrators finer control of DNS responses to configured FQDNs. Administrators may selectively respond with public or private sites based on whether the client is from a public or private IP. For example, administrators may wish to allow only private clients to be sent to private sites.
The following table outlines settings and their configurable values:
Site Types Allowed
Public, Private if no public
Private, Public if no private
Private and Public
Private, Public if no private
Public, Private if no public
Private and Public
9. A Failure Delay (minutes) can be set if needed. If a Failure Delay is set, another option called Site Recovery Mode becomes available. Refer to the Enabling Fail Over section for further information on these options.
Following a completely failed health check, the GEO LoadMaster waits for the specified number of minutes before taking the site out of rotation.
10. Enable/disable the Enable Local Settings check box, as needed. If enabled, configure the TTL and Stickiness options.
11. Enable or disable Unanimous Cluster Health Checks. If this option is enabled, if any IP addresses fail health checking â€“ the other FQDN IP addresses in the same cluster is forced down. For further information, please refer to the Scheduling Methods section.
12. Enter the IP address of the domain in the IP address text box.
13. If needed, select the Cluster name.
14. Click Add Address.
15. Select the type of health checking to be performed from the Checker drop-down list. For further information regarding health checking options, refer to the GEO, Product Overview document.
The GEO LoadMaster load balances DNS requests â€“ it does not load balance traffic. GEO offers many load balancing algorithms including round robin, weighted round robin, fixed weighting, real server load, location based, proximity and all available.
Each of these scheduling methods are described below.
â€œRound Robinâ€ load balancing can be used for all active data centers, which includes support for weights and a chained failover option for disaster recovery. Round robin scheduling in GEO LoadMasters works the same way as the normal LoadMaster with one exception; when using nslookup, by default it will check for both IPv4 (A) records and IPv6 (AAAA) records which actually sends out two requests.
For example, if you have two sites:
Request 1 IPv4 A record hits Site 1,
Request 2 IPv6 record AAAA hits Site 2,
Request 3 IPv4 A record hits Site 1,
Request 4 IPv6 AAAA record hits Site 2
Clients looking for IPv4 will always connect to Site 1.
Clients looking for IPv6 will always connect to Site 2.
To help prevent this during testing â€“ add an odd number of sites.
This method balances out the weakness of the simple round robin: incoming requests are distributed across the cluster in a sequential manner, while taking account of a static â€œweightingâ€ that can be pre-assigned per server.
The administrator simply defines the capacities of the servers available by weighting the servers. The most efficient server, A for example, is given the weighting 100, whilst a much less powerful server (B) is weighted at 50. This means that Server A would always receive two consecutive requests before Server B receives its first one, and so on.
Fixed weighted scheduling is usually used in Disaster Recovery (DR) sites. The highest weight Real Server is only used when other Real Server(s) are given lower weight values. However, if the highest weighted server fails, the Real Server with the next highest priority number is available to serve clients. The weight for each Real Server should be assigned based on the priority among the remaining Real Servers.
GEO uses local data center metrics provided by LoadMaster, allowing clients to connect to the least busy data center.
Location Based load balancing allows GEO to direct a client to a data center based on the client's country, continent or IP address range as defined by the created policies. If there is more than one site with the same country code, requests are distributed in a round robin fashion to each of the sites. Location Based load balancing can be used for granularity, for example, if a site in Germany fails â€“ send traffic to the next site in Europe (not the next closest proximity site).
Proximity takes Location Based one step further and allows for longitude and latitude granularity for definition of proximity. When using Proximity scheduling, new public sites are automatically mapped to geographic coordinates based on the GEO database. New private sites are mapped to 0Âº0'0" and function as expected. This coordinate should be overridden with accurate values to ensure correct balancing.
In order to use the Proximity selection criteria with private IP addresses, the IP Range Selection Criteria must be completed for all private subnets. In addition to this, both the coordinates and country must be configured. If they are not configured, requests from private IP addresses are rejected.
For more information on the IP Range Selection Criteria, refer to the IP Range Selection Criteria section.
In the example above, either proximity or location-based scheduling could be used. If proximity scheduling is used, the client is directed towards the German domain because it is geographically closer to it than the French domain. If location-based scheduling is used, the client is directed towards the French domain because it is in the same country. Using proximity scheduling here may result in a faster connection. However, if users need to be directed towards the French version of a website â€“ it may be better to use location-based scheduling.
The All Available selection criteria returns all possible healthy targets for an A (IPv4) or AAAA (IPv6) query request. The GEO LoadMaster will still refuse other records, for example MX. The contents of the returned list is also controlled by the Public Requests and Private Requests settings:
For Public Sites Only the list can only contain public addresses. Likewise, for Private Sites Only the list can only contain private addresses.
For Prefer Public the list only contains public addresses, unless no public addresses are available â€“ in which case the list contains private addresses (if any are available). Likewise, for Prefer Private the list only contains private addresses, unless no private addresses are available â€“ in which case the list contains public addresses (if any are available).
For All Sites the list contains all available addresses
The purpose of this is to provide a list of preferred addresses, if they are available. Otherwise, provide a list of non-preferred addresses as a failback measure for improved availability.
In the IP Range Selection Criteria menu option you can specify a location, country or custom location that apply to an IP address or range.
Custom locations can also be added on the IP Range Selection Criteria screen.
To do this, follow the steps below:
1. In the main menu of the LoadMaster WUI, select Global Balancing.
2. Select IP Range Selection Criteria.
3. Enter the IP Address or network. Valid entries here are either a single IP, for example 10.154.11.10, or a network in Classless Inter-Domain Routing (CIDR) format, for example 10.154.11.10/32.
4. Click Add Address.
5. Click Modify.
6. Specify the coordinates click Save.
7. Alternatively, select the country from the Location drop-down list.
The existing IP ranges can be modified or deleted using the buttons provided on the IP Range Selection Criteria screen.
8. In the main menu of the LoadMaster WUI, select Manage FQDNs.
9. Click Modify on the relevant FQDN.
10. If you entered a Proximity using the coordinates in the IP Range Selection Criteria screen, select Proximity in the Selection Criteria drop-down list.
11. If you selected a location, select Location Based.
12. Fill out the remaining details as needed.
When configuring an FQDN, one of the options which can be configured is Unanimous Cluster Health Checks. If this option is enabled, if any IP addresses fail health checking - other FQDN IP addresses which belong to the same cluster is marked as down. When Unanimous Cluster Health Checks is enabled, the IP addresses which belong to the same cluster within a specific FQDN are either all up or all down. For example, example.com has addresses 172.21.58.101, 172.21.58.102 and 172.21.58.103 which all belong to cluster cl58:
If 172.21.58.101 fails, the unanimous policy forces 172.21.58.102 and 172.21.58.103 down as well.
When 172.21.58.101 comes back, the unanimous policy brings back 172.21.58.102 and 172.21.58.103 along with it.
At any given time â€“ either all three addresses are available or all three addresses are down.
The same approach applies for site failure mode with manual recovery. Manual recovery causes a failed address to be disabled, so the administrator can re-enable it after fixing the problem. When Unanimous Cluster Health Checks is enabled, all three addresses are disabled.
The unanimous policy ignores disabled addresses. So, if you know that an address is down, and for whatever reason you want to continue using the other addresses that belong to the same cluster, you can disable the failed address and the unanimous policy will not force down the other addresses with it.
When Unanimous Cluster Health Checks are enabled, some configuration changes may cause FQDN addresses to be forced down or brought back up. For example, if an address is forced down and you remove it from the cluster while the unanimous policy is in effect, the address should come back up. Similarly, if you add an address to a cluster where the unanimous policy is in effect and one of the addresses is down, the new address should be forced down. This change may not occur immediately, but it should happen the next time health checking occurs.
If there are addresses with the Checker set to None combined with addresses that have health checking configured â€“ addresses with no health checking will not be forced down, but they can be forcibly disabled if the Site Recovery Mode is set to Manual. For example, say there are three addresses:
172.21.58.101 with a Checker of Cluster Checks
172.21.58.102 with a Checker of Cluster Checks
172.21.58.103 with a Checker of None
If site failure handling is off or automatic, the failure of 172.21.58.101 causes 172.21.58.102 to be forced down, but 172.21.58.103 remains up. The rationale is that if you do not want health checking on 172.21.58.103 then it should remain up.
However, if the Site Recovery Mode is set to Manual, failure of 172.21.58.101 causes both 172.21.58.102 and 172.21.58.103 to be disabled, along with 172.21.58.101. For site recovery â€“ all addresses are disabled, even the ones with no health checking configured. This is to keep traffic away from the problem data center until the system administrators fix it. This does not conflict with having addresses with no health checking because you can have an address that is up but disabled.
A cluster is a group of LoadMasters working in conjunction. Clusters can also be non-LoadMaster entities using TCP or ICMP health checks. GEO clustering is a feature mainly used inside data centers. Health checks are performed on a machine (IP address) associated to a specific FQDN, using the containing cluster server, rather than the machine itself.
You can add a maximum of 18 GEO clusters.
In terms of GEO, a cluster is a GEO LoadMaster polling another device for health checks. For example, a GEO LoadMaster can poll a firewall on a site in front of Real Servers. If the firewall is up, it is assumed that everything behind the firewall is up too. If the health check to the firewall fails, the Real Servers behind the firewall are marked as down.
GEO LoadMasters can be used with a normal LoadMaster as a cluster. There are two methods for clusters:
First, the LoadMaster can be polled for health checks (same as a firewall or any other device). If this LoadMaster health check fails, everything behind the LoadMaster is marked as down.
Second, you can use cluster checks. In this case, the GEO LoadMaster polls the LoadMaster and the LoadMaster informs the GEO LoadMaster which Virtual Services are up.
Clusters can be added, modified and deleted from the Global Balancing > Manage Clusters menu option.
In this example, here are the IP addresses being added:
GEO LoadMaster: 10.113.0.54
Regular LoadMaster: 10.113.0.28
Virtual Service IP address: 10.113.0.37
To add a cluster, follow the steps below on a GEO LoadMaster:
1. Enter the IP address of the cluster (in this example it is the LoadMaster with IP address 10.113.0.28).
2. Enter a Name for the cluster.
3. Click Add Cluster.
In the example, the LoadMaster (10.113.0.28) must be connected to the GEO LoadMaster (10.113.0.54). To do this, follow the steps below in the LoadMaster:
1. In the main menu of the LoadMaster WUI, go to Certificates & Security > Remote Access.
2. Type the IP address of the GEO LoadMaster (10.113.0.54 in this case) and click Set GEO LoadMaster access.
To add the FQDN and Virtual Service IP address, follow the steps below in the GEO LoadMaster:
1. In the main menu of the LoadMaster WUI, go to Global Balancing > Manage FQDNs.
2. Enter the FQDN and click Add FQDN.
3. Enter the IP address of the Virtual Service (10.113.0.37 in this example).
4. Select the relevant Cluster (LM28 in this example).
5. Click Add Address.
6. Configure any other settings as needed.
To modify an existing cluster, follow the steps below:
1. Click Modify on the relevant cluster.
2. Change the settings as needed.
The Type can be Default, Remote LM or Local LM:
Default: When the type of cluster is set to Default, the check is performed against the cluster using one of the following three available health checks:
- None: No health check is performed. Therefore, the machine always appears to be up.
- ICMP Ping: The health check is performed by pinging against the cluster IP address.
- TCP Connect: The health check is performed by connecting to the cluster IP address on the port specified.
Local LM: When Local LM is selected as the Type, the Checkers field is automatically set to Not Needed. This is because the health check is not necessary because the cluster is the local machine.
Remote LM: The health check for this type of cluster is Implicit (it is performed using SSH).
The only difference between Remote LM and Local LM is that Local LM saves a TCP connection because it gets the information locally and not over TCP. Otherwise, the functionality is the same. Default is a generic cluster type that does not communicate with a LoadMaster. It uses TCP or ICMP health checks.
Remote LM and Local LM are only used when the target is a LoadMaster as opposed to a server or another resource. Local LM is used by GEO to check the LoadMaster it is enabled on.
When Default is selected, either ICMP Ping or TCP Connect can be selected as the health check type in the Checkers drop-down list.
When Remote LM or Local LM is selected, no health check options are available. Remote LM health checks are performed using SSH on port 22.
If needed, click Show Locations to enter the latitude and longitude of the location of the IP address.
After a cluster is initially added, the health check marks it as Up by default. The status is updated after the next health check cycle.
To delete a cluster, click Delete in the Configured Clusters screen.
There is no â€œundoâ€ function here. Delete with care.
When setting up clustering in a multi-GEO environment, all LoadMaster clusters must be of the type Remote LM.
Each LoadMaster which is to be used as a cluster must have all GEO IP addresses listed in the Remote GEO LoadMaster Access field, which is in Certificates & Security > Remote Access in the WUI.
When LoadMaster and GEO HA pairs are used, the shared IP address must be listed in the Remote GEO LoadMaster Access field.
DNSSEC verification of signed responses was included in the DNS client in LoadMaster firmware version 7.1.34.
DNSSEC digital signing (2K key) support for DNS responses was added to the GSLB LoadMaster in firmware version 7.2.37.
DNSSEC helps protect against cache poisoning using a set of extensions that provide origin authentication of DNS data, data integrity and authenticated denial of existence. DNSSEC provides a mechanism to sign requests and prove the validity of records in a given zone and does this through a process called zone signing.
DNSSEC adds four new resource record types:
Resource Record Signature (RRSIG)
DNS Public Key (DNSKEY)
Delegation Signer (DS)
Next Secure (NSEC)
Before configuring DNSSEC, a zone must be defined. A zone is a single unique part of a DNS namespace hierarchy that serves as the authoritative source for information about a select set of DNS domain names.
To define a zone, go to Global Balancing > Miscellaneous Params and specify a Zone Name.
To enable DNSSEC in the LoadMaster, follow the steps below:
1. Go to Global Balancing > Configure DNSSEC to configure the DNSSEC options.
2. You can either import the Key Signing Keys (KSKs), or generate them. To import them, click Import and browse to and select the files. If generating, go to the next step.
A KSK is a type of DNSKEY that is used to sign the keys contained within a DNS zone and are leveraged to validate resolvers. The KSK also signs the Zone Signing Key (ZSK).
3. If generating the KSKs, click Generate. Select the Algorithm and Key Size and click Generate.
4. The KSK details are displayed.
5. Select the Enable DNSSEC check box.
There is no user interface for ZSK files. A ZSK is used to generate Resource Record Signatures (RRSIG) for each set of resource records in a zone and sign these records. GEO creates the ZSK files automatically when DNSSEC is enabled. The same algorithm is used as specified for the KSK files. A key size of 1024 is used. If DNSSEC is disabled, the KSK files are deleted.
It is possible to download blacklist rules from KEMP to block access from IP addresses that are on the blacklist. A whitelist can be manually specified that overrides the blacklist. These rules can be set to automatically download and install or they can be manually downloaded and installed.
This is a subscription-based feature. If you cannot see these options, or if any fields are grayed out, please contact KEMP to upgrade your subscription.
To configure the IP blacklist settings, follow the steps below:
1. In the main menu of the LoadMaster WUI, go to Global Balancing > IP Blacklist Settings.
2. Select whether or not to Enable Automated GEO IP Blacklist data Updates.
3. In the Last Updated section, you can manually download the rule updates now. You can also view changes, that is, what are the differences between the latest downloaded rules and the previously downloaded rules.
After downloading the rule updates, they must be installed to become active.
If the rules are more than 7 days old, a message appears.
4. Select whether or not to Enable Automated Installs. When enabled, you can specify what time to install the updates at.
5. You can Manually Install the GEO IP Blacklist data by clicking Install Now.
If the GEO blacklist data is not updated for more than 7 days, a message appears.
6. View the GEO IP Blacklist data file by clicking View. This displays a full list of the blacklisted IP addresses.
7. Addresses and networks can be added to the whitelist by entering them into the Address/Network text box and clicking Add. The whitelist overrides the blacklist.
The Certificates & Security option in the main menu of the LoadMaster WUI enables you to import and manage SSL certificates. It also gives the ability to generate a Certificate Signing Request (CSR). This option is only relevant for the GEO LoadMaster WUI access.
When there are multiple LoadMaster boxes, where each box could be a single LoadMaster or a HA pair, they can be linked together to act as a single resource.
When a HA LoadMaster pair is configured to do GEO synchronization, all of the shared IP addresses must be added to each partner configuration correctly.
All of the boxes remain synchronized with each other and share their DNS Configurations, FQDN information, â€˜Stickinessâ€™ information and health checking updates. Any updates are automatically shared with all the other Distributed Partners.
The Geographical IP Database used for the Proximity and Location Based load balancing methods is not distributed between the LoadMaster partners. Any updates to the Geographical IP Database must be configured on each LoadMaster individually.
HA is the same for GEO LoadMasters as it is for regular LoadMasters â€“ it is an active-passive pair of units.
Partners are two or more GEO units in an active-active mode.
It is possible to have both HA and partners.
In the example diagram above, public and private addresses are used. The partner address should be the NATed addresses if the GEOs are connecting externally.
Before partnering GEO LoadMasters, backup the relevant GEO LoadMaster that has the correct/preferred configuration. This backup should then be restored to the other LoadMasters that are partnered with the original LoadMaster. This is a prerequisite because GEO partners are active-active and share the same configuration. As a result of this, you mustensure that the settings are in sync before adding a partner.
To perform a backup, follow the steps below in the WUI of the GEO LoadMaster that has the correct/preferred configuration settings:
1. In the main menu, go to System Configuration > System Administration > Backup/Restore.
2. Click Create Backup File. The file downloads.
Then, on the GEO LoadMaster(s) that are partnered - follow the steps below to restore the configuration:
3. In the main menu, go to System Configuration > System Administration > Backup/Restore.
4. Click Choose File.
5. Browse to and select the backup file.
6. Select the Geo Configuration check box.
7. Click Restore Configuration.
Now that the correct configuration is applied, the GEO LoadMasters can be partnered.
To partner the GEO LoadMasters, follow the steps below:
1. Select the Certificates & Security > Remote Access option from the main menu.
2. Ensure the Enable Remote SSH Access check box is selected.
3. In the GEO LoadMaster Partners text box, enter the IP address of the LoadMaster to partner with. If there is more than one box, enter the IP addresses but separate them with a space.
4. Click Set GEO LoadMaster Partners.
5. Enter the port number that the LoadMasters use to communicate in the GEO LoadMaster Port text box.
6. Click Set GEO LoadMaster Port.
7. In the GEO update interface drop-down list, select the GEO interface that the GEO partners will communicate through.
8. Repeat the above steps for all other GEO LoadMasters to be partnered.
After a GEO LoadMaster is partnered, its status is indicated in the GEO Partners section of the Remote Access screen.
Figure 3-22 shows GEO Partner status of Green indicating the two partners can see each other.
Figure 3-22 also shows GEO Partner status of Red indicating the LoadMasters cannot communicate. The reasons for this include (among other possibilities); one of the partners is powered down, there may be a power outage or a cable disconnected.
A red status could also appear because when initially set, the status of the GEO Partner on the first GEO LoadMaster displays as red. When you set the GEO Partner on the second GEO LoadMaster, the status displays as green on the second LoadMaster but remains red on the first GEO LoadMaster for approximately three minutes until the first GEO LoadMaster updates. As a workaround, click Set GEO LoadMaster Partners on the first LoadMaster again and the status changes to green.
If there is a failure to update the GEO partner, the logs display an error message saying the GEO update to the partner failed. The message displays the IP address of the partner.
If you experience any issues with partnering the GEO LoadMasters, check the SSH settings and do a TCP dump to confirm that SSH connections are being sent and received by all parties.
When upgrading GEO partners, it is strongly recommended that all nodes are upgraded simultaneously. Because GEO partners operate in active-active mode, upgrading at the same time ensures that consistent behavior is experienced across all nodes.
If you must operate a GEO cluster with mixed versions, be sure to make all changes from the most recent version. This prevents configuration loss due to incompatible configurations. Additionally, changing configuration options not present in older versions results in disparate behavior.
Microsoft Exchange data center or site failures require a combination of both automatic and manual steps to be completed before client service is fully restored and for the outage to end. The manual steps, mainly centred on the administration of the mailbox databases, pose some unique issues regarding an Exchange data center failover which are not found in other types of site failures.
GEO functionality provides options to Exchange administrators to assist in dealing with the unique issues posed by an Exchange data center failover.
KEMP recommends enabling the Fail Over option in Exchange environments.
Enabling the Fail Over option enables Location Based FQDNs to select the most appropriate site for requests in the event that the best match is not available. When the Fail Over option is enabled, if a request comes from a specific region and the target is down, the connection will fail over and be answered with the next level in the hierarchy. If this is not available, the connection is answered by the nearest (by proximity) target. If this is not possible, the target with the lowest requests are picked. For example, if a request from Ireland is received, but the site assigned to Ireland is unavailable, a site assigned to Europe is selected. If the site assigned to Europe is also unavailable, a site assigned to Everywhere is selected. If this too is unavailable, the least requested of the available sites is selected. The Fail Over setting affects all targets. The Fail Over option is only available when the Selection Criteria is set to Location Based.
Implementing an Exchange data center failover is not a trivial event - it is not recommended to automatically failover upon detection of a site failure. Delaying the failover for a short period ensures that failovers do not occur because of trivial and temporary failures.
Delaying the failover can also provide the Exchange administrators time to ensure that the secondary site is ready to provide the requisite levels of service.
The LoadMaster provides a Failure Delay option which, when enabled, delays a failover occurring for a configurable period of time after a site failure is detected. If, after the delay, the site recovers, the failover is not initiated. If the site has not recovered, the failover is initiated as per normal.
If a Failure Delay is set, another option becomes available underneath it â€“ Site Recovery Mode. Two modes are available:
Automatic: The site is brought back into operation immediately upon site recovery
Manual: After the site has failed, disable the site. Manual intervention is required to restore normal operation.
It is recommended that when a failed data center recovers, attempting to restore services to the recovered data center is not initiated (a failback) before the application failover process is complete and until the recovered data center is deemed to be healthy. In addition, the mailbox databases should beready for use.
If the failback is initiated before these are complete, then issues may arise with the mailbox databases and prolong the outage.
KEMP recommends not to allow automatic failbacks when a failed data center recovers in the case of Microsoft Exchange. Requiring some manual intervention before a recovered data center is deemed available for a failback means that administrators can ensure that the optimal conditions exist before a failback occurs.
The LoadMaster provides Site Recovery Mode options that determine how a failed data center becomes available for failback upon recovery. Selecting the manual option configures the LoadMaster to administratively disable the failed data center upon the initiation of a failover. This ensures that, even if the failed data center recovers, administrator intervention would be required before the data center is available for a failback to occur.
GEO functionality provides options to Exchange administrators to assist in dealing with the unique issues posed by an Exchange data center failover. These options can be found within the Site Failure Handling section of the FQDN configuration page, which is accessed as follows:
1. In the main menu, go to Global Balancing > Manage FQDNs.
2. Click Modify for the relevant FQDN.
This setting determines how long (in minutes) the LoadMaster waits until it designates a site as having failed, thereby initiating a site failover.
It is recommended that this setting is enabled when configuring multi-site Exchange deployments. The optimal value given to the delay depends on the configuration of the Exchange deployment.
Site Recovery Mode
This determines what recovery options are implemented when a failed site recovers.
Automatic - when the data center recovers, the LoadMaster automatically performs a failback (restores services to the recovered data center)
Manual - upon failure, the data center is administratively disabled and is not available for a failback until the admin clicks the Enable button for the relevant data center
The Manual option is recommended for multi-site Exchange deployments to ensure that failbacks do not occur until the administrators ensure that the optimal conditions for a failback exist.
The Lockdown option has been deprecated and will only be available to people who upgrade with this configuration enabled. If you do not have the configuration enabled or if you change any of the Site Failure Handling options, then it will not be available. The Lockdown option is not recommended as a Site Recovery Mode option.
Web User Interface (WUI), Configuration Guide
SSL Accelerated Services, Feature Description
Reason for Change
Updates for the 7.2.36 release
Updates for the 7.2.37 release
|Mar 2017||Release updates||Updates for the 7.2.38 release||11.0||LB|
|July 2017||Release updates||Updates for the 7.2.39 release||12.0||LB|