Cisco Application :: GSS 4492R Configuration With ACE4710?
Dec 8, 2010
I have to deploy the Cisco GSS in our 2 dataceters globally seprate IP ranges to loadbalance the exchange 2010 environment with Cisco ACE 4710 series SLBs. The scenario is to deploy one GSS + ACE on each datacenters and our nameserver will point to both GSS's IP addresses to get through. Incase primary site "site A" goes down name server will point the client's request to "site B".
What will be the physical setup of the GSS here and what configuration should on SLB ACE will make it work? Do GSS and ACE need to be in the same vlan? is this necessary to use Both interface of the GSS to get things working? How the GSS will check the health check on ACE if they both are on different vlans/ip range? Our ACE will be in routed mode do we need to assign the Real server default gateway as ACE inside interface with the server farm or just do the SNAT of the client IPs so the request can come back to ACE?
My customer they do not want change their real server IPs. So I need setup one interace (one armed) for them on ACE4710. Who had this sample configuration? (CSS has this but it seems to be not compitable with ACE)
I am attempting to configure an ACE4710 to perform SSL end-to-end confguration. i.e. SSL termination - load balance - SSL initiate to backend server.The configuration appears to work fine in a test lab using any old web server, however when I peform the same configuration in the production environment it does not work. It appeatrs from a capture run on the ace that the ace is reseting the tcp connections after communicating with the back end server. The main difference I can think of in this environment is that the cert and key pair the ace is using where exported from the backend server, i.e. both the ace and the backend server have the same certificates and keys. Is this allowed? how to troubleshoot why the ace resets the connection.
I have a problem, recently I can not telnet to the ACE 4710, the ACE version is A4(2.0). I can enter by web and console but not for telnet. I try to give more resource to the admin context but it doesnt work.
We have a rather strange issue, and I'm not sure it's really a problem with the ACE or not. We created an HTTP parameter map called "TCPreuse" and applied it to a virtual server. A show conn detail displays "[ conn in reuse pool : FALSE]" for all connections pertaining to this virtual server. The rserver in question is Linux(Ubuntu) + Apache.Next we applied the HTTP parameter map to another virtual server - this time IIS7 + Windows Server 2008. There are plenty of entries "[ conn in reuse pool : TRUE ]" when I do a "show conn detail".What could the web servers be doing differently that would cause the connection reuse to work on one virtual server and not the other?
I need to upgrade 2 active-standby cisco ACE4710, the issue is I cannot access FTP/TFTP/SFTP server via Admin context but can be accessible via other contexts.
Can I copy the ACE system software file from FTP/SFTP/TFTP server to image: directory durectly or need some other way around ? I could see the option is available to copy ftp: to image: via other context.
All clients point to unique VIP and Port configured, I need to know if I can apply any filter or rule that allows me to distinguish when a customer goes to the URL1 or URL2.If any client try to access to URL-2, your traffic must be deny.In summary, from Internet I should be able to go only to URL-1.
I have HA configuration for two ACE4710. FT between Ace's is configured as L2 (V LAN). Active ACE is sending heartbeats, but switch shows lot of 'input errors' on ingress and this is a major problem. FT is logically not working (there is no connection between these two Ace's over V LAN). There is only L2 configuration, with speed and duplex auto, no other special configuration. When I connect Ace's directly, FT is working without problem.
I can see lot of errors on input direction (from ACE) to switch port, that means, L1, or L2 problem, but direct connection (using the same Ethernet cable) is working. I tried 'shut/no shut' on both sides, set duplex/speed,... without success.
Do you know if it is possible in ACE 4710 appliance to configure a SIP TLS ?The SIP probe we have in the configuration guide it is only for clear text. for Lync 2013 we need to establish first a TLS session and then within it, send an SIP request..IS it possible in any version? I tried also to configure a HTTPS probe but it fails as it sends a GET which the Lync SIP server doesn't understand.
ACE 4710 deployment model. We'll be doing an eval later in the year, but I'm just looking to understand the architecture.We have a stack of 3750 switches with a single VLAN (10.1.1.0/24). Connected to that stack is a pair of web servers (10.1.1.5 and 6) that we want to provide load balancing/failover for. Some of the clients are located right there on that same VLAN. Other clients may be coming from other spots in the infrastructure.It sounds like I could put a pair of 4710s connected to that stack of switches, in a single arm deployment? And then the virtual IP and the real servers would all be 10.1.1.0/24. Maybe use an etherchanel to connect each 4710 to two 3750s?
I'm implementing and found out some issues are unresolvable on ACE4710. This network have been running on a server without LB. Now the second server comes up. We choosed to implement with Routed Mode.This network Peak @ 300Mbps. Now on we're doing the first context which is function as content web-farm. In near future, 2nd context which takes care of indexing web-farm when they buy more server.
From following diagrams.I browsed from internet into this service. "show service-policy" shown as '0' (counter was not running). I guessed that there is something wrong in FW configuration. So I isolated out FW. Then I plugged-in my PC into network 30 (192.168.30.X) in front of this LB, then browsed into LB's VIP (192.168.30.1). LB "show service-policy" came up BUT there is nothing return to my PC (client). "show conn" on LB as "SYNSEEN". What's SYNSEEN?! Some meaningful.
Then I tried to figure out with a PC running 'apache' and took the place of real server. "It works!" returned from LB/Server. "show conn" became 'Establish' Programmer guy said if I browse into web-farm (i.e. content web-farm) directly pkt will be redirected to indexing server. But they said it will be L7 redirection. Not LB/Network level.
We recently configured a setup to loadbalance 2 application server using ACE4710. Initially the configuration was to link two app servers directly to ACE4710 without connecting to a Switch, but later, it was advised that ACE4710 is not able to work without connecting to a switch.
1. ACE4710 is not able to link directly to APP/WEB server, but it must go through a network Switch.
2. If item-1 above is true. We used to have a older Cisco Loadbalancer which is able to link directly to WEB/APP servers. What is the reason or advantage of removing this feature?
On my ACE4710s I'm using least-loaded predictors monitoring Microsoft Windows CPU usage. There are times when the MS Windows CPU OIDs can change between reboots. Any way for the ACE to automatically adjust to the new CPU OIDs and continue to get accurate CPU usage values?
one of our 2 ACE boxes in FT group suddenly reboot-ed from active state and now it continuously reboots with the following error:
insmod: error inserting '/isan/bin/klm_octeon_device.klm': -1 No such deviceerror inserting /isan/bin/klm_octeon_device.klmDaughter Card Not Found. Rebooting..INIT: Sending all processes the TERM signal...
We have a CSS11503 that is currently being used to accept incoming HTTPS and SSH connections on a specific VIP and then PAT those client connections. I understand that it also PATs the server initiated connections. [code]
For server load balancing, does the ACE4710 support custom protocols? We'll be using HTTP for server health monitoring, and to determine if a server is up or down. But the client/server application is custom, and includes a lot of non-standard ports. Can the server VIP handle generic TCP connections? For example client1 connects to the VIP on http, but then later client1 switches to using tcp842 (a custom protocol, not http).
I have a ACE4710 setup to load balance a couple of web servers. The real servers all show as inservice as do the propbes and serverfarms/virtual servers. If I ping the Virtuual server ip address I get a reply but it I try to access VIP via telnet or web browser. I get a connection could not be open error on the client.The question is how do i determine where the error is comming from so far I can not tell if the client is getting through the acl or not.I have used the trouble shooting guide and nothing has worked to determine the cause so far. show service-policy int479 detail does not show an increase in the hit count when I try to connect.show stats conn does not show an increase in failed or timed out connections when i try to connect. [code]
My customer wants each ACE4710 (of a highly available cluster) to have its own, dedicated port for management purposes.
According to documentation, IP addressing can be applied to VLAN interfaces, so in order to satisfy the requirement, I should make one port belong to an "access VLAN X", and then apply IP addressing to the corresponding "interface VLAN X". This should satisfy my customer´s requirement in an indirect way.
But... ¿ Can´t I just configure IP address on one of the 4 ethernet ports in order to save the work of building the aforementioned VLAN? I am asking this since I do not have access to a real box in order to verify.
I was trying to implement stickiness based on cookie. Server inserts a cookie and sends it to the browser. I learned from app team that this cookie is changing dynamically during the session, so stickiness based on server’s cookie doesn’t work.
Now I want to investigate into possibility of ACE to insert a cookie. My question is: ACE feature of “cookie insert”: does it add additional cookie into http header without removing server cookies or it deletes the cookie(s) that server put into http header and replaces them with its own cookie?
We have an ACE Appliance in a DMZ and the ACE Appliance's Admin Context IP is translated between ACE and ANM. The ANM Server does not get translated. It is just the opposite then in another Community discussion.
Our Problem: When adding the ACE4710 Appliance to the ANM imported Device List, we use the ACE's NATed Admin Context IP. Import works well, but ANM reflects the Admin Context IP with it's real configured IP. Polling the ACE Appliance does not work therefore.
Is there a possibility of telling the ANM, that the ACE has to be polled through a NATed IP? I could not find a field to set a NATed Mgmt IP.
Configured IP on ACE Admin Context: 192.168.0.10 NATed ACE Admin Context IP: 172.16.0.10
Imported ACE with IP 172.16.0.10 into ANM, but ANM polls for Rserver, Vserver, Probes, etc. via 192.168.0.10 - which is not reachable from the ANM.
i am trying to use teh ACE to replace an apache based load balancer in an jboss application cluster. I am using L7 loadbalancing to load balance between multiple components. the way these jboss application servers work with apache is that.When the jboss application starts up on the on application cluster, it issues a GET opencase/webservices/config-service?wsdl to the loadbalancer IP.The apache based LB in turn talks to the same box on port 8009 via ajp retrieves the configuration file and provides it back to the application on port 80.And after 2 has completed the Jboss application comes up. Basically to start the application the Apache loadbalacer will accept requests from the its target list and load balance the request back to them itself.Not sure how i can use the ACE to accomplish this. attached are my topology (logical) and the ace configuration. from my topology file -- the net-cms-1 will issue a get request to teh VIP (on the ace), the ACE accepts the connection but soon resets it.
I am currently stuck to setup an automated configuration backup for ACE Blades. I found a script to backup the ACE from the Cisco ANM box but unfortunately I am not very familiar with Linux. (script) in place, to "pull" the ACE config from a Microsoft system ?
I have Ace 4710 version A4.1.1 and I am experiencing interesting problem with GUI and SSH reachability. I am unable to connect to management vlan3000. Interesting is that I can ping from ACE to network but unable to ping or SSH or HTTPS to ACE. Everything seems good. ARP is ok, switch is OK, line is up, protocol is up. Management is enabled for icmp, https, ssh to any.
When I do show interface I noticed line Config download failures : 9.
Hardware type is VLAN MAC address is 00:1e:68:1e:bc:db Virtual MAC address is 00:0b:fc:fe:1b:01 Mode : routed IP address is 10.168.0.18 netmask is 255.255.255.128 FT status is active Description:Management VLAN MTU: 1500 bytes Last cleared: never Last Changed: Mon Jan 21 16:48:54 2013 No of transitions: 5 Alias IP address not set Peer IP address is 10.168.0.19 Peer IP netmask is 255.255.255.128 Assigned on the physical port, up on the [Code] ....
I only have configured load balancing on apache with a very simple setup. I have to deploy 2 applications on my clients environment that run inside jboss. One of these applications needs session to be sticky to work properly. The other does not.
In apache I can configure is the sticky parameter is true or false, based on the url, like /appA/* is sticky and /appB/* is not sticky. Can I do that in a CSS 11503? My client insists that it is impossible. That the CSS is only ip based.
I copied the configuration below from the manual: owner arrowpoint # content ruleWapSticky
I am configuring a load balancer from cisco, a ACE 4710.Load blancing is completely new to me, and i am unexpereinced in this field. It has to be configured for a customer that want to load balance HTTP and RTSP traffic over 4 application servers (Back-end),I searched alot on google for possible solutions, and got RTSP in some way to work, but http wont work says my customer.
We are running ACE 20 modules in highly available active / standby (all active contexts on one module) mode. Currently they are on A2 (2.4) version. We are going to upgrade them to A2 (3.6a). The question is that how ling can we run them in two different SW levels? In otherwords can we have few days between upgrading both modules?
I have two ACE-4710 in active/standby mode, running code A3(2.2). Four contexts are configured. Both devices were functional without problem, until I reload the standby unit. After reload, the standby unit completely lost its configuration with exception of the FT vlan and the FT peer configuration in the Admin context... Both units recognized each-other and I can still ping the primary unit on the FT vlan, but nothing else. Contexts are lost and interfaces are shutdown! Nothing changed at the software level, both devices run exactly the same image and the same licences are installed (it worked well before the reload).
So, I decided to reconfigure the basics on the standby unit in order to trigger a config sync from the primary. And here arrives the problem : I reconfigure the FT vlan, the FT peer, I check the peer state and everything is OK.
Then, I try to ping the primary unit from the standby unit with success :
switch/Admin# ping 192.168.16.1 Pinging 192.168.16.1 with timeout = 2, count = 5, size = 100 .... Response from 192.168.16.1 : seq 1 time 0.000 ms
I am looking at management (backup of the configuration) of the ACE 4710 running A4.1, the management software is Cisco Cirrus. The question I have is around the management of the context's, I have a backup of the Admin but would like the user context's also, how this is completed.
Is it possible to deploy the CSS11501 in one arm design to loadbalance the authentication traffic Radius across CSACS servers which is on UDP 1645 or 1812 port, is it required to configure the NAT or not, if yes how can define the shared secret in the CSS. also tell me how to configure the keepalive for udp traffic in this scenario other then default icmp keep alive