CSM : Sorry server and Stickyness when reals are overloaded

Hi,
I have a portal of eight real servers and one sorry server, which should redirect new user to another portal, in case of an overload condition of all eight real servers. Server load is measured on each real server using a custom developed agent, which basically measures the real CPU load. If a real server experiences an overload, the local agent uses the CSM XML interface to set the maxcons value in the CSM to stop accepting new connections. However, I want to continue accepting sticky connections (request with a valid cookie). The experience shows that the CSM does accept to create new connections to real server reaching maxcons, even if a cookie exist.
This causes a problem if we want to redirect NEW users to another portal in case of overload, but to keep EXISTING users in the server farm, even if the number of connections could increase slightly above maxconns...
How can I solve the problem ?
Thank you
Yves Haemmerli

Hi Thomas,
Thank you for your comment. I also understand this behaviour like you, however this can have a devastating effect in a global portal environment. Imagine, you have three portals distributed over the world, each having let say 8 real servers. In the real life, it is seldom to replicate data in real time between data centers, due to the distance. However, the user roles and customized bookmarks and other user-specific settings are replicated. This allows to provide a global portal to users. But if a user connects to one particular regional portal, he has to stay on this portal for the duration of the whole browser session, do you follow me ? OK, now imagine that all 8 real servers in a portal reach the maxconns, because 10'000 users are connected to the portal. For new users (users with no sticky cookie), we want to send them to another regional portal. This is achieved with the global site selection provided by the GSS for example. But for existing user already connected to the overloaded portal, we want to KEEP them on the portal ! else, as the user browser continuously opens and closes TCP sessions, all 10'000 users will be immediately transferred to the other regional portal! This means the the other regional portal will becom overloaded as well, while the first portal load will be droped to zero very quickly ! Then, we not only create a situation where users loose their data by being transferred to another portal, but we also create a oscillations in the portal load !
I really don't know if there is a mean to solve this problem...Do you have any idea ?
Regards,
Yves Haemmerli

Similar Messages

  • In Mountain Lion how does OS select between ethernet and WiFi when both are activated??

    In Mountain Lion how does OS select between ethernet and WiFi when both are activated??  How does OS use "both" networl accesses??

    It will pick the highest service in the list which has an internet connection.
    To order the list, open Network System Preferences and select Set order of network services from the button below the service list.

  • ACE sorry server and sticky

    I have configured 3 different serverfarms with including realservers
    2 of them are with websites, the other 1 is with webservices
    I also have configured a sorry server farm and the including rserver.
    On the sorry rserver i have configured 2 maintenance websites, listening to an unique hostheader.
    So for serverfarm A & B i have configured a seperate maintenance website.
    Now when i take rservers from serverfarm A or B down, the sorry server will get active for the needed farm.
    However i can only reach 1 maintenance website. And even so, an url used to reach farm A gets on maintenance site from B
    This is strange behaviour, doesnt a sorryserver just accept requests with the requested hostheader by the client ?
    Also, when i put the rservers from A and B back into service i have to do a "clear stick database all" otherwise the sorryserver will remain active.
    What is wrong here ?
    probe http EHIC-http
    description Test op WWW functionaliteit
    interval 10
    passdetect interval 30
    request method get url http://acc.site-B.nl/web/
    expect status 200 200
    header Host header-value "acc.site-B.nl"
    expect regex 1.8.0.2
    probe http WWW-http
    description Test op WWW functionaliteit
    interval 10
    passdetect interval 30
    request method get url http://acc.site-A.nl/web/default.aspx
    expect status 200 200
    header Host header-value "acc.site-A.nl"
    expect regex v1.9.2.327
    serverfarm host EHIC-FARM
    failaction purge
    predictor leastconns slowstart 30
    probe EHIC-http
    rserver ehic_server01.site-B.nl
    inservice
    serverfarm host SORRY-FARM
    failaction purge
    predictor leastconns
    rserver sorrypage.site-C.nl
    inservice
    serverfarm host WBS-FARM
    failaction purge
    predictor leastconns slowstart 30
    probe ICMP-PROBE
    rserver acc-wbs01v.site-D
    inservice
    rserver wbs_01.site-D
    inservice
    rserver wbs_02.site-D
    inservice
    serverfarm host WWW-FARM
    failaction purge
    predictor leastconns slowstart 30
    probe WWW-http
    rserver acc-www01v.site-A
    inservice
    rserver acc_server01.site-A
    inservice
    rserver acc_server02.site-A
    inservice
    sticky ip-netmask 255.255.255.255 address source EHIC-FARM-STICKY
    serverfarm EHIC-FARM backup SORRY-FARM
    sticky ip-netmask 255.255.255.255 address source WWW-FARM-STICKY
    serverfarm WWW-FARM backup SORRY-FARM
    class-map match-any EHIC-VIP
    2 match virtual-address 172.30.9.4 tcp eq https
    3 match virtual-address 172.30.9.4 tcp eq www
    class-map match-any WBS-VIP
    6 match virtual-address 172.30.5.4 tcp eq www
    7 match virtual-address 172.30.5.4 tcp eq https
    class-map match-any WWW-VIP
    2 match virtual-address 172.30.6.4 tcp eq www
    3 match virtual-address 172.30.6.4 tcp eq https
    policy-map type loadbalance first-match EHIC-FARM-STICKY-BALANCE
    class class-default
    sticky-serverfarm EHIC-FARM-STICKY
    policy-map type loadbalance first-match WBS-FARM-BALANCE
    class class-default
    serverfarm WBS-FARM
    policy-map type loadbalance first-match WWW-FARM-STICKY-BALANCE
    class class-default
    sticky-serverfarm WWW-FARM-STICKY
    policy-map multi-match LOADBALANCING-EHIC
    class EHIC-VIP
    loadbalance vip inservice
    loadbalance policy EHIC-FARM-STICKY-BALANCE
    loadbalance vip icmp-reply active
    appl-parameter http advanced-options EHIC-PARAMETERS
    policy-map multi-match LOADBALANCING-WBS
    class WBS-VIP
    loadbalance vip inservice
    loadbalance policy WBS-FARM-BALANCE
    loadbalance vip icmp-reply active
    appl-parameter http advanced-options WBS-PARAMETERS
    policy-map multi-match LOADBALANCING-WWW
    class WWW-VIP
    loadbalance vip inservice
    loadbalance policy WWW-FARM-STICKY-BALANCE
    loadbalance vip icmp-reply active
    appl-parameter http advanced-options WWW-PARAMETERS
    Regards,
    Sebastian

    Hi Gilles,
    Here is our full config, i only changed some domain names.
    I'll try to describe the problem again ;
    We have published a website by vip 172.30.6.4
    We have another website published by vip 172.30.9.4
    These websites are hosted by realservers configured in 2 serverfarms and can be reached from the internet (secured by an ASA)
    For both of these farms i have configured a sorryserver. This sorry server should serve a webpage containing a maintenance message whenever a farm should get down.
    The sorry server is configured with 2 websites, each listening to the specific hostheader. This hostheader is the same as configured on the rservers for the specific farm 172.30.6.4 or 172.30.9.4.
    So what i am trying to accomplish is that i only need 1 sorryserver to server 2 sorry webpages, ofcourse listening to a hostheader to get 2 different sorrypages to be returned.
    Now when i take all realservers for both serverfarms down, except for the sorryserver, i can only reach 1 sorrypage.
    For example, site A and B are down, when i try to reach site A i get to the sorrypage of site A. But when i try to reach site B i too get served the sorrypage of site A.
    And also when i "inservice" all rservers again i have to do a "clear sticky database", otherwise the sorryserver will remain active.
    Now i have upgraded to the last version of the ACE ios, but i still have to test if the same problem persists so i will give feedback on this later.
    Regards,
    Sebastian

  • Reg: HTTP and RTMP when these are in Network Delays

    Video streaming in HTTP and RTMP, which one performs better when delays are introduced. Streaming from one PC to other PC through LAN. Here i introdced delays in both HTTP and RTMP streaming. So, far i have seen RTMP is quite good in low delay variations such as -+10ms. Here Qn is which is better when high delays are introduced.
    Thanks in Advance,
    sree7k7

    Start Firefox in <u>[[Safe Mode|Safe Mode]]</u> to check if one of the extensions (Firefox/Tools > Add-ons > Extensions) or if hardware acceleration is causing the problem (switch to the DEFAULT theme: Firefox/Tools > Add-ons > Appearance).
    *Do not click the Reset button on the Safe mode start window or otherwise make changes.
    *https://support.mozilla.org/kb/Safe+Mode
    *https://support.mozilla.org/kb/Troubleshooting+extensions+and+themes
    Do you have any ad-blocking extensions or (security) software?

  • Speed between server and client when using FML

    I using FML between server and client, the server access oracle using only 5 ms, but when tranfer back to client, it about 100ms long. I tuned my hp-ux 11 kernel and tuxedo config file, but useless, Why?

    Hi Tumecan,
    Your expected information available here, check.
    Frontend Network Load - Network Integration Guide (BC-NET) - SAP Library
    few related SAP Notes:
    164102 - Network load between application server and front end
    500235 - Network Diagnosis with NIPING
    62418 - Network Load of SAPGUI Frontend Communication
    679918 - The front-end network time
    578118 - Long response times on the SAP GUI
    161053 - Using SAP GUI in WAN
    Regards,
    V Srinivasan

  • Why purchase server AND client when I already bought the Lion client?

    When I try to purchase the Lion Server from the App Store it displays a message that OS X Lion is also required uand I have to purchase both even when I am logged in and next to the Server item the client version confirms that I have already purchased it with a "download" button - the app has been downloaded but cannot install it as it is waiting for the app store to download the server components... and what if I just want to use it as a client only? Any suggestions besides the obvious $80 purchase. Thanks.

    Solved. Purchased only the server portion from the machine used to purchase the Lion upgrade. Did not install it there... went to the MacMini Server and checked the purchases... for some reason it was hidden... unhid the purchase and downloaded the server app. Solved.

  • How does iPad choose between wifi and 3G when both are 'on'?

    Easy peasy question.
    I have 02 and Wifi going on my iPad. I know I can use 02 out and about and I know I can use wifi at home etc.
    But when I'm at home how do I make sure I'm using my wifi for data and not my cellular connection???
    I've been turning my cellular to 'off' when at home, just in case, but I don't want to have the faff of turning it off and on, so it would be nice to know how the iPad chooses between wifi and 3G when both data access points are active.
    Also I've had no wifi troubles at home at all. so far... The acid test will be setting it up at work as they have proxyies for everything, but obviously I don't have proxies at home. I'm not sure if the settings screen can cope with different proxies dependent upon different wifi networks... * grin *

    In the status bar I have
    a signal strength indicator then 'O2-UK' then the 'fan' No mention of 3G.
    I'm guessing you're going to say that when 3G is showing then 3G is being used for data?

  • Radio buttons (from Indesign) get larger and overlap when fields are highlighted in Acrobat X

    I have created my first set of interactive forms using Indesign CS6 Buttons and Forms dialog box. As the formatting of the fields does not seem to carry over, I have left the text boxes as Indesign objects and added text fields over the top of them in Acrobat X. I also intended to add drop down menus in Acrobat instead because I didn't like how the Indesign list box behaved in Acrobat (having to click through each item rather than dropping down the entire list). There does not seem to be a way to add a drop down menu item from within Indesign. (Or if there is, can someone please point it out? :-) )
    However I used the sample buttons supplied within Indesign to create radio buttons and check boxes (which automatically included the graphics for the different states). On my first test export to Acrobat it appeared that the buttons retained their formatting, but something rather strange is happening when the fields are highlighted, whereby the buttons enlarge and overlap each other. Very frustrating I must say! I have just about gone nuts trying to find a solution. I've tried several different button options from the sample set; I've created them from scratch within Indesign; I've attempted to resize the fields in Acrobat form editing mode, but this screws up the original buttons. I also have found that once selected in Acrobat, I cannot deselect them again. (Don't know if this is significant info but it may point to some kind of stacking problem??)
    Anyway, to make it clearer here are some screen captures:
    1. Form as it looks when opened in Acrobat X. The two different check box styles are from me experimenting in trying to fix this issue. But they all seem to behave the same way once highlighted.                                                               
    2. Form when interactive fields are highlighted. The text fields are perfect but note the big ugly overlapping circles... :-(
    And some further explanation of how I created the buttons and their issues:
    I would very much appreciate some guidance on what I'm doing wrong here, as my poor little head is going to explode very shortly (and my client is getting pretty annoyed at the delay!!)
    Thank you in advance :-)

    I am having the exact same problem. I can't find a way around it. It seems that they included some states like normal, rollover, and click—but they didn't include a highlighted state. So when highlight fields is on it just shows the generic standard check box, which looks ridiculous if you have customized it at all in inDesign. I read a lot of people saying they can't find a way to change the highlight state—anyone out there have any answers?

  • How to change VM Server IP address when VMs are provisioned

    Dear all,
    I have deployed some Oracle VM servers and provisioned some VMs running upon on them.
    However, the user requests to change the management IP address of these VM servers. How can I do it?
    In my practice, I need backup all VMs and remove them. Then remove these VM servers from pool. Then change management IP...add them into server pool...restore VMs from backup....
    Pls correct me if I am wrong.
    Thanks,
    Mike

    You don't have to remove or backup/restore the VMs. Just migrate them to another server (if shared storage) or to Unassigned VMs (if local storage). The machines should be stopped (not suspended). There is a bug where a suspended machine can migrate to Unassigned VMs but can't migrate back (without manual intervention on the console). If you are changing the pool virtual IP, you have to delete/recreate the server pool. This will lose the "simple names" of the disks and ISOs, so record them before deleting the pool. It will not remove the Repository contents. After recreating the pool, you can rediscover the server, present the repository and refresh its contents.

  • Message server hogs CPU when clients are disconnected

    Hi, I have a problem where java process for the Sun One Message Queue (3.0.1) hogs the cpu when a client is disconnected. I am trying to use the MQ in an environment where clients might crash and try to reconnect in an unpredictable fashion. Most of the time, when I kill the client and restart it, I do not see the problem. Sometimes, however, doing so makes the java process of mq take almsot 100% of Cpu and it does not go down unless i kill the mq process.
    I am using asynchrnous message listeners with transactions, on a persisted queue, although i saw this problem when transaction was disabled.
    Does anyone have any idea what might be the source of the problem? I hope it's just a matter of configuring MQ correctly, but i could figure it out. Help would be greatly appreciated.

    Are you running the enterprise edition and
    using round-robin queues with multiple
    receivers ?
    If so, you may be running into a timing related bug
    in multiple receivers that has been fixed in service
    pack releases of the software
    You should contact sun support to discuss testing with
    the lastest service pack.
    If not, can you provide more information (platform,
    # and type of receivers, etc) on how you are using
    the product
    Thanks

  • I unpluged my headphones, and now when they are out my computer is on a permeant mute.

    I tried to change my output setting to internal speakers, but the setting doesn't even seem to be there.
    I did some reasearch and it says that it must be something in my earphone jack.Because even when i unplug them a red light is still shining in there.
    What can i do to fix it?

    Try plugging and unplugging the headphones several times, and see if  that turns it from digital out (red light) to analog out (no red light)  Sometimes rebooting the computer with the headphones plugged in, then pulling them out will reset it. There are a couple of more things to check, but start wit these

  • FF10 crashes every time I view the download manager, and sometimes when downloads are taking place in the background. WHTF FF10?

    crash reports have not been submitted because FF10 crashes completely and I have to close using WTM
    never happened under previous FF releases
    OS: Win 7 HP 64 bit
    PC:
    2500k
    MSI P67-GD65
    Radeon 6950
    Crucial M4 120GB boot drive (with FF on)
    WDCBlue 500GB data drive

    Hey duddbudda,
    Have you tried starting Firefox in Safe Mode? Hold down "shift" while Firefox is starting up. If the crash doesn't happen in Safe Mode, you should read [[Troubleshooting extensions and themes]].
    There is also a lot of useful stuff in the [[Firefox crashes]] Knowledge Base article. If you've recently upgraded, you might want to check out [https://support.mozilla.com/en-US/kb/latest-firefox-issues#w_firefox-crashes-when-i-start-it this article] as well.
    Hopefully this helps!

  • I recently accidently upgraded to firefox 3.5.1 and when I did, my bookmarks now only list in recentlyy bookmarked and the when full are removed. How do I get my bookmarks to go in list. I use xp pro

    sorry for the repeat but I forgot to add my os.

    See http://kb.mozillazine.org/Bookmark_dialog
    Just to be sure that you understand how it works:
    * Clicking the star in the location bar will bookmark the page in the "Unsorted Bookmarks" folder and the star turns yellow (blue on Mac) to show that.
    * The "Unsorted Bookmarks" folder shows in the sidebar and the Library, but not in the Bookmarks menu drop down list.
    * Bookmarks > "Bookmark This Page" (Ctrl+D) will bookmark the page in the Bookmarks Menu folder (you need to confirm that).
    * If the URL in the location bar is bookmarked (star at the right end is yellow/blue) then click the star or use the Bookmark > Bookmark This Page (Ctrl+D) menu entry to edit the properties of the bookmark (name, location) and move it to another folder or remove the bookmark.
    * Click the star at the right end of the location bar once to bookmark the current site and click the highlighted star a second time to open the "Edit This Bookmark" window and move the bookmark to another folder or edit the name.
    * Click the Remove button in the "Edit This Bookmark" window to remove a bookmark.
    * Click the arrow at the far right of the Folder field in the "Edit This Bookmark" window to access the complete folder list. That full folder list also opens if you click the drop marker in the Folder field and select the "Choose" item. That window also has a button to create a "New Folder" (look for a button with "New Folder" or a "+" under the folder list).
    * Use the "Edit This Bookmark" window to locate a bookmarked link (bookmarks have a yellow/blue star) from the location bar drop down list. Do that in Offline mode ("File > Work Offline") to avoid the link from opening.
    See:
    * [[Bookmarks]]
    * [[Organizing Bookmarks]]
    Other things that need attention:
    Your above posted system details show outdated plugin(s) with known security and stability risks that you should update.
    # Shockwave Flash 10.0 r45
    # Java Plug-in 1.5.0_06 for Netscape Navigator (DLL Helper)
    Update the [[Managing the Flash plugin|Flash]] plugin to the latest version.
    *http://www.adobe.com/software/flash/about/
    Update the [[Java]] plugin to the latest version.
    *http://www.oracle.com/technetwork/java/javase/downloads/index.html (Java Platform: Download JRE)

  • Thunderbird stopped downloading emails from the web server and now when I try to add a new account I get a message to say it cannot find the email server.

    Thunderbird was working and then for no reason it stopped downloading emails from the web server. I can access my emails in the web server okay. I have tried to remove my account and then add it again as a new account but I get a message saying that Thunderbird cannot find the email server. I have also tried to uninstall and reinstall Thunderbird but with the same result. I have also tried to remove the default mail server but have been unable to do so. How can I fix this?

    I figured out a workaround.... Since I can get e-mails to populate via iSync and my other computer is running Leopard (it's a powerpc chip and can't run Snow Leopard), I added the account on the other computer and got the new account to populate into my SL machine with iSync. I don't like that I can't actually add the account directly, but at least this got me working.

  • Syncing BSO and ASO when there are extra members in ASO

    Hi All,
    I have a BSO cube for planning application and an ASO cube which is used for reporting purposes. Now the problem is heavy changes were made in planning like deleting members, moving members to a different parent etc and the ASO cube was not updated for weeks. Now I have to update the ASO Cube. What are my options?
    Thanks.
    Edited by: user9097967 on Dec 23, 2011 9:10 AM
    Version 11.1.2

    Depends on how you are managing the metadata to the planning application.
    Cheers
    John
    http://john-goodwin.blogspot.com/

Maybe you are looking for