Cisco Application :: Ace 4710 Response Sticky Only
Dec 15, 2011
I've been using my pair of ACE-4710s for quite some time and have usually stuck to the Class C Subnet sticky settings, as that's what we migrated from in Windows NLB. In one instance of load balancing I'm trying to create an L4 inspection policy that looks for a certain payload (much like a http header) and would like to persist on this. The problem is that the client portion of the conversation starts with a 'SessionID' of 0, and the server responds with a unique 'SessionID'. If I setup the sticky policy with 'Enable Sticky For Response', I get entries populated in the sticky database, but they all go to the same server as there is a sticky session setup for the SessionID = 0. Is there a way to setup sticky entries on server response only.Currently using ACE DM v4(1.0).
I'm looking for some documentation I can share with a customer to explain why I can't configure them a back-up sticky server farm when we're not terminating SSL on the the ACE (we pass SSL from the client through to the rservers, sticking the client to the rserver by source IP address). I've not been able to find anything that addresses this particular scenario in my googling so far. I remember discussing this in my training class with the instructor, but I can't find any reference to it. Have any of you run into this and have a link they can share?
We are using a sticky serverfarm with 2 real servers, one server was down for maintenance for an extended period of time. When it came inservice again it was not getting any connections. is it because all the connections had stuck to the other server ? we want sessions to be sticky but we also want to LB?I got it working by bouncing the server that had been online all the time. things started to LB then.BTW the ACE 4710 is running 4.2.1
I have a requirement to load balance OWA 2010 inbound connectivity to 2 CAS servers using a ACE 4710 with sticky sessions enabled.
The CAS servers are currently responding on 80 or 443 at this moment in time. Eventually I want to off load the SSL to the ACE 4710, its currently running on the CAS servers. I need to enable sticky sessions to keep the session to the same CAS server for each internet based connection. I also have a proxy enabled for inbound connectivity so I cannot use source IP.
Here is my configuration but it doesn’t seem to be working, i am currently testing with port 80 connections not SSL.
We are using a ACE 4710 with A3(2.6) software release.I had to change our sticky load balancing method for HTTPS to cookie based.However while connections appear to work if I look at the show sticky database table I can not see or confirm sticky entries for the cookie based connections.Here or config snippets to show the config
I have a strange effect at my ACE 4710. I loadbalances normally reliable only 14 WEB-Services.
It's running on SW A3.25. Since several weeks I regognized a dramatical increase of Sticky entries. So when running in limitations (the stolen for reuse counter increased then) (show np 1 me-stats "-slb -v") gave more and more resources for sticky ... last it was at 65% and ran again into limits at round 650500 Sticky entries.
So I began to find out what of the services was affected with most sticky database entries and could Identify it. There were really to see round about640000 entries for that specific service.
The sticky for that service was defined to look at a specific cookie in the http header and the timeout defined is 120 minutes.
So round about 45000 Entries was to see with a "show sticky databse group Cookie_Sticky" with a time-to-expire value of zero in the database like the follwing examüple shows:
We want to mask part of the path prefix to hide development content: For example: the site(s) are: [URL]However we don't want anything with acme showing...so we would want the loadbalanced url to be: [URL] ...for requests and responses. I think this would be an http re-write request/response scenario?Is this possible to configure this on the ACE Device? We've got the load balance configuration down...not sure how to do this re-write type scenario?
I've been using my pair of ACE-4710s for quite some time and have usually stuck to the Class C Subnet sticky settings, as that's what we migrated from in Windows NLB. In one instance of load balancing I'm trying to create an L4 inspection policy that looks for a certain payload (much like a http header) and would like to persist on this. The problem is that the client portion of the conversation starts with a 'SessionID' of 0, and the server responds with a unique 'SessionID'. If I setup the sticky policy with 'Enable Sticky For Response', I get entries populated in the sticky database, but they all go to the same server as there is a sticky session setup for the SessionID = 0. Is there a way to setup sticky entries on server response only? Currently using ACE DM v4(1.0).
how a static entry under a "sticky" performs Configuring Static IP Address Sticky Table Entries Cisco Documentation Says When you configure a static entry, the ACE enters it into the sticky table immediately. Configuring the ACE Action on Server Failure failaction purge # The purge keyword specifies that the ACE remove the connections to a real server if that real server in the server farm fails after you enter the command. The ACE sends a reset (RST) to both the client and the server that failed. Cisco Documentation Says If you do not configure this command, the ACE takes no action when a server fails
Report run via Individual Web server URL’sThe report takes less than 20 minutes (average 15 minutes) to fetch and return the data. This is observed 9 out of 10 times.Report run via ACE Load Balanced URLThe report keeps on running for more than 20 minutes and never completes. The front end keeps showing report is running.The data in general when tested directly by running queries against the database (bypassing the platform) completes in 15-18 minutesThe network connectivity for each and every ports involved (Loadbalancer/Servers) have been throulgly checked.
I am trying to track down an application response problem on my network (the traffic goes through a 6509 and FWSM).I noticed in one of my WireShark captures, that the client at times seems to be sending ackowledgements (ACKs) over and over again, and I'm not sure if this indicates a problem/ retransmission.
Basically, a web server is delivering images to the client, but end users are complaining of slowness and freezes.Wire-Shark has not flagged this as a problem (comes up "green")
Having issues with HTTPS sites being very slow after applying KB2585542? Once you remove this Microsoft patch everything returns to normal. It appears that the CSS does not handle the split-ssl requests properly. I have opened a TAC case but am not really getting anywhere as we seem to be the only company that is having this issue.
i'm looking for a recommendation for a setup guide including ft i've had a quick look a wiki and i can get basics but i'm not sure about if i need to setup additional contexts etc when i'm the only one using the appliance?
I have an issue with a customer that wants to update a server behind the ACE. The problem is that when the application wants to update the server it does it with the name.Doing some research I found that you can rewrite the record DNS based on the static NAT you set up on the ACE. The feature is called DNS inspection. Is the same feature as the ASA (DNS doctoring).I apply it to the outside interface and it did not work.
With the current (A5) ACE 4710 lic setup, does the "X gigabit per second appliance throughput" that is licensed affect: -
A) Only "appliance" i.e. load balancing traffic, any other normal routed traffic is not included in the limit
B) Is it an overall throughput limit on the interfaces i.e. includes all traffic not only load balancing traffic but also normal routed traffic crossing the appliance
Looking at a scenario where the lic size I need for HTTP load balanacing would be one size if A) but would need to be much larger is B) to accomodate out of hours routed backup traffic crossing the ACE 4710
I've got a web app that the owners want to run over port 80, but also using SSL to secure private data in transit. The architecture is an ACE 4710 in SSL termination mode->Apache (port 2000)->Back-End app server.
I've got two VIPs set up already - one on port 443 and one on 2000 - both of which do the SSL termination quite nicely, but using the 3rd VIP set up on port 80, the connection steadfastly refuses to be HTTPS (i.e. doesn't show the padlock).
I've done all the set-up through the web interface so far, can this be done? If so, how?
I am currently running A3(2.6) and evaluate the possibility of upgrading to A4(2.1). The Instal & Upgrade Guide A4(2.0) mentions that A4(2.0) does not include all features of A4(1.1). Does this apply to A4(2.1)? The Release Notes mentions a list of features merged from A4(1.1) to A4(2.1) but does not clarify if there any features not merged.
we configued An ACE 4710 with SSL termination on Oracle Aplication Server 10g (10.1.2.0.2) ,so that SSL termination is done on the ACE and HTTP reaches the Oracle Aplication Server 10g (10.1.2.0.2) then we configure the ACE to enabled client authentication with Pkcs#11 smart card token certificate and this don succfully my problem need do this client certificate authentication for only the [URL] not for all SSL proxy service how can do that.
We have recently transitioned one of our Ecommerce products to a new data center, at which we now use a one-armed load balancing approach rather then the routed load balancing approach we used previously. This is casuing us some issues as we generally log the source IP address a user comes in on when he fills out an application. Now the logs only show the natted ip address recieved by the load balancer, which does us no good. Any way to log the source IP address when a new connection is created to a particular vip?
If we use an ACE4710 to load balance two real servers, obviously it will use health checks to determine if a server is down.When it detects a server is down, it will not send it any more traffic.But can we also have it take any other action? For example maybe email an admin, or send an SNMP trap? Or better yet, can we use a custom TCL script to do other things, like launch some custom activities?
I am new to the 4710 appliance.Apart from the 4 GE 'data' ports, there are 2 Ethernet 'management' ("console") ports. I find the description in the "quick start guide"somewhat confusing. URL, Is a first-time serial connection (at least to run the initial config. script) mandatory? Or can you obtain the same result via one of the 2 Ethernet management ports and using a default ip address (192.168.1.10 ? When running the initial config. script (only possible from the serially connected console i suppose), you have to select your management port. Why does the system in step 5 proposes you 4 ports, and not just 2? I suppose the intended port for management is one of the 2 management ports, not one of the 4 data ports?
I have a pair of ACE 4710s with 12 contexts sharing the load, running A4(2.1). esterday I upgraded one of them to A4(2.3) now I cannot telnet to the Admin context.Pings ok. I can telnet to other contexts on the box and everything seems to be working ok when i do a " sh telnet" comes back with
No Session Information is available sh telnet maxsessions telnet maxsessions 16
ACE# sh script code NORDICID_PROBE.Error: Called API is invalid or non-existant.Hardware is ACE-4710-K9 and software A3(2.7)The probe itself is functioning ok according to show probe detail.However show script script_name probe_name -counters all remain at zero for some reason. This wasn't the case on the previously use ACE software.To my recollection the command show script code has worked successfully before on the same ACE software. Not 100% sure though, but it definitely worked on the previous software we ran on the ACE.