Load balancing on CPU cores
Hi,
We have a supermicro server with 2 x E5-2660 processors and 160GB of memory.
The only role assigned to this server is Hyper-V and we have about 100 VMs each with two vCores provisioned on this server.
I checked the cpu load by monioring the %Processor Time counter and noticed that while 10 out of 32 cores are at 100%, 15 of them are below 10%.
I am wondering if it possible to balance the load on the CPU cores to improve the performance and utilize the full processing power of each CPU.
Hyper-V will schedule things appropriately - most likely better than any manual intervention could provide. It could be the nature of your applications.
Hyper-V does provide the ability to assign memory weightings to different VMs if you really want to get into that.
.:|:.:|:. tim
Similar Messages
-
How is load-balancing achieved in MPLS L3 vpns and equal cost multiple links exist to reach egress PE along with per-destination load-balancing enabled on interfaces.
I have tried to simulate the network below
Ingress PE--->P1--->>P2--->Egress PE
Multiple equal cost links exist between P1 and P2, cisco platform,LDP, IGP-ospf being used.Hi,
Destination based load balancing in MPLS L3VPNs can be categorized into two scenarios:
1) multiple pathes between two PE routers
2) multiple access links to a single CE or site
Your question as I understand it was about the first scenario. So let me first quickly review how customer traffic is forwarded between VRFs on two different PE routers.
The VRF routing table will have BGP entries for the routes learned from the remote PE usually with next hop addresses being the remote PE loopback IP used for PE-to-PE BGP peering.
The traffic will be forwarded across P routers using the label for the BGP next hop.
Thus the load balancing accross the MPLS core in a first step is decided by the IGP, which has to insert several equal cost pathes into the global routing table for the BGP next hop networks (PE loopbacks).
Side note: MPLS traffic engineering in the core would allow for unequal cost load balancing.
The decision, which labeled packet to send across which path in the core is done by CEF using a hash algorithm. To achieve the same load balancing as with unlabeled IP traffic, a Cisco MPLS enabled router will look for the bottom label - the one with bottom-of-stack bit set to 1 - and try to determine, if the transported packet behind the bottom label is IP. If so, the hash is calculated for the customer IP header like for normal IP traffic. This ensures all traffic for a certain customer destination will always go through the same path. No unwanted packet reordering will occur.
Be aware, that the customer IP packet header will only be used for CEF hash calculation, no IP lookup will be performed, as core routers in MPLS L3VPNs do not have any knowledge about customer addresses.
As a side note: if the traffic transported is not IP (e.g. Ethernet over MPLS), the bottom label will be used for the CEF load balancing (e.g. the VC label).
For the second scenario - CE load balancing with multihomed CE/sites - it is first required to have two equal cost entries in the VRF routing tables. The difference will be the two different PE BGP next hop addresses. The first load balancing decision is the performed by CEF based on the IP packet received by the CE and the VRF routing table entries. Once CEF decided, which VRF entry to use, the required BGP next hop label (and the VPN label) is applied and the packet is transported across the MPLS core. load balancing there is done as described above.
Hope this helps! Please rate all posts.
Regards, Martin -
MPLS/VPN network load balancing in the core
Hi,
I've an issue about cef based load-balancing in the MPLS core in MPLS/VPN environment. If you consider flow-based load balancing, the path (out interface) will be chosen based on source-destination IP address. What about in MPLS/VPN environment? The hash will be based on PE router src-dst loopback addresses, or vrf packet src-dst in P and PE router? The topology would be:
CE---PE===P===PE---CE
I'm interested in load balancing efficiency if I duplicate the link between P and PE routers.
Thank you for your help!
GaborHi,
On the PE router you could set different types and 2 levels of load-balancing.
For instance, in case of a DUAL-homed site, subnet A prefix for VPN A could be advertised in the VPN by PE1 or PE2.
PE1 receives this prefix via eBGP session from CE1 and keep this route as best due to external state.
PE2 receives this prefix via eBGP session from CE2 and keep this route as best due to external state.
eBGP
PE1 ---------CE1
PE3----------P1 Subnet A
PE2----------CE2 /
eBGP
Therefore from PE3 point of view, 2 routes are available assuming that IGP metric for PE3/PE1 is equal to PE3/PE2.
The a 1rst level of load-sharing can be achieve thanks to the maximum-paths ibgp number command.
2 MP-BGP routes are received on PE3:
PE3->PE1->CE1->subnet A
PE3->PE2->CE2->subnet A
To use both routes you must set the number at 2 at least : maximum-paths ibgp 2
But gess what, in the real world an MPLS backbone hardly garantee an equal IGP cost between 2 Egress PE for a given prefix.
So it is often necessary to ignore the IGP metric by adding the "unequal-cost" keyword: maximum-paths unequal-cost ibgp 2
By default the load-balancing is called "per-session": source and destination addresses are considered to choose the path and the outgoing interface avoiding reordering the packets on the target site. Overwise it is possible to use "per-packet" load-balancing.
Then a 2nd load-sharing level can occur.
For instance:
__P1__PE1__CE1
PE3 \/ Subnet A
\ __P2__PE2__CE2
There is still 2 MP-BGP paths :
PE3->P1->PE1->CE1->subnet A
PE3->P1->PE2->CE2->subnet A
But this time for 2 MP-BGP paths 4 IGP path are available:
PE3->P1->PE1->CE1->subnet A
PE3->P1->PE2->CE2->subnet A
PE3->P2->PE1->CE1->subnet A
PE3->P2->PE2->CE2->subnet A
For a load-balancing to be active between those 4 paths, they must exist in the routing table thanks to the "maximum-path 4 "command in the IGP (ex OSPF) process.
Therefore if those 4 paths are equal-cost IGP paths then a 2nd level load-balancing is achieved. the default behabior is the same source destination mechanism to selected the "per-session" path as mentionned before.
On an LSP each LSR could use this feature.
BR -
4 virtual CPUs are how much physical CPU core
Hi Everybody,
I've two VMs with 4 VCPU per VM.
First VM is created on VMware the second VM is on Oracle Virtual Server 3.1.1.
The VMware server is on the same HW architecture as Oracle VM Server (two blade as twins).
I've created Oracle database server 11gR2 on both VM.
The Enterprise Manager Cloud Control "Database - Performance Home" page shows 4 CPU Cores in case of VMware VM,
but only 1 CPU Core in case of Oracle VM.
I'd like to understand why.
Help me please.
Regards,
BattilaDid you set processors and the max processors the same on the VM guests?
One CPU is one CPU core. If you do not use CPU pinning..... CPU core allocation is load balanced across all cores in the system. When you use CPU pinning. (Which you must do in a oracle database to adhere to licensing requirements where they apply) CPU cores are dedicated to the VM guest and not load balanced. You can overcommit CPU resources in Oracle VM. In other words. You can set max CPU across all VM guests on a single VM server to be greater than the total amount of CPU cores available in the system. Check out article 464754.1 -
hy guys,
i read a previous post regarding how the
load-balancing is done for eompls.
someone says that the inner label (vc label) is responsable for this.
sincerely, if someone wants to explain me much deep how this works, it will be
very appreciate.
also, what is the meaning of the control word in the vc setup procedure ?Load-balancing in the core is indeed based on the inner label (vc label) for EoMPLS.
So if the P router has many outbound interfaces for a given LSP, it will use that innner label to compute the hash value to select the outbound interface for that specific packet.
For an explanation on the control word please refer to the Martini encapsulation draft:
http://www.ietf.org/internet-drafts/draft-martini-l2circuit-encap-mpls-10.txt
Hope this helps, -
Webform loading too long : only 1 cpu core used out of 16 core available?
Hi,
We have one webform which is quite heavy and takes 1min30 to load :
1 minute for essbase calculation (spreadsheet extractor)
30 seconds for network & browser rendering
We have seen that Essbase is only using 1 cpu core (100% usage) out 16 core available
Does it is possible to give some calculations to other cpu core? I know it's possible for HBR rules with CALCPARALLEL but for a webform?
Thanks!
Edited by: Tipiak on 3 juil. 2012 23:06The system has no ability to use multiple threads for one web form. It sounds as if your form is not designed for optimal performance which usually means you have a large number of total cells and/or are not using dense dimensions for your rows and columns (usually accounts/periods).
Suggest you refer to the KB article which discusses Hyperion Planning Form design at:
https://support.oracle.com/CSP/main/article?cmd=show&type=NOT&id=779502.1
Regards,
john A. Booth
http://www.metavero.com -
CSM - Load balance using Server CPU
Hi
I have a customer who requires the load-balancing prediction
algorithm to be based on the CPU level of the Server. So the server with the least CPU is chosen at the laod-balancing stage.
Is there a way to do this?
Thanks JamesHi James
With CSM the only option is DFP (Dynamic feed back protocol). If your application vendor provides DFP agents (which is very unlikely) for the application then these agents can be installed on App servers for the desired purpose.
Config details at
http://www.cisco.com/en/US/docs/interfaces_modules/services_modules/csm/4.2.x/configuration/guide/rsfarms.html#wp1039774')">http://www.cisco.com/en/US/docs/interfaces_modules/services_modules/csm/4.2.x/configuration/guide/rsfarms.html#wp1039774
With ACE you can use SNMP based probes to achieve what you are looking for.
Syed Iftekhar Ahmed -
Dears
I need your assistance please regarding A-VPLS core load-balance
As I understood, A-VPLS inserts a new label "Flow Label" so that this label can be used in core routers to acheive load-balancing as shown in figure 2
Now my questions
1- P router has to use port-channel so it can calculate a hash based on the flow label, correct ?
2- Also I should configure "port-channel load-balance mpls" at P router, correct ?
3- If I am correct, flow label is the 3rd label according to figure 1 and according to command reference for "port-channel load-balance label" only last 2 labels are taken into considersation !?
------------Quote----------
If you select label, these guidelines apply:
• With only one MPLS label, the last MPLS label is used.
• With two or more MPLS labels, the last two labels (up to the fifth label) are used.
---------Unquote------------
Regards,
Sherif IsmailHello Sherif,
I am not completely sure of the guidelines that cisco have quoted, but the document which you are referring doesnt seem to be updated to the RFC recomendations. According to the RFC 6790, the Flow label ( or the entropy label ) is no longer placed after the application label ( which is the pseudowire label in your diagram ). This is now placed after the tunnel label ( which is the ldp label or RSVP label ). There should also be an Entropy label indicator ( reserved label value of 7 ) after the tunnel label ( and before the entropy label ) to indicate that the following label is the one that will be used for load balancing.
In the initial drafts of this RFC, the egress had to infer from the BOS bit on the application label whether or not there was an entropy label
All the hashing is calculated at the ingress. The output of the hash is the entropy label. The core routers would just use the label to loadbalance the traffic.
Regards,
Shreeram -
Load balance not work,still sending new connects to node at100% CPU usage
Trying to verify that Oracle RAC load balancing ( using Load Balance Advisor and services) is working before I recommend changes to a production RAC DB.
Even though I bury (100%CPU usage) one of 2 nodes ,new connections(using a new service I created and set GOAL=SERVICE_TIME and CLB_GOAL = SHORT ) still connecting there.( connection split stays at 50/50) . Any ideas why ?
nodes are HP Itanium ( hpux 11.23) , database version is 10.2.0.5
There are 2 RAC databases on this cluster (one TEST and one Model Office)
srvctl config service -d gdrm -s BATCHQ_GDRM_SERVICEBATCHQ_GDRM_SERVICE PREF: gdrm1 gdrm2 AVAIL:Hi Dan,
you will have to point all your systems to the Web Dispatcher in the SLD. If you don't do this, the browser will always retrieve the URL to the central instance, bypassing the Web Dispatcher. After you configured as the system to contact the Web Dispatcher, the browser will receive the URL of the Web Dispatcher and send the request to the Web Dispatcher.
On the Web Dispatcher, you'll have to configure an appropriate rule to handle that request. You can use 1 IP and different ports. Each port means another system. Or several virtual IPs and the same/different port.
IP1:8001 -> WD -> BI 1
IP1:8002 -> WD -> BI 2
or
IP1:8001 -> WD -> BI1
IP2:8002 -> WD -> BI2
In this case, you can keep the flexibility of using several backend systems with different names/IPs/Ports and one Web Dispatcher.
For refence, you can consult the following 2 documents:
General Information about how to use Web Dispatcher:
https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uuid/8105db90-0201-0010-cfbe-ae94d43e2e48
How to implement Web Dispatcher as a single point of entrance
p. 40-42: "
Internet Connectivity to SAP Applications.pdf"
(Unfortunatly, I'm right now not able to locate this document again in SDN, but it exists somewhere. Good luck!)
br,
Tobias -
2 locations, 2 core switch stacks, fibre in between, equal cost load balancing between?
Hi,
We've recently inherited a job that another company was doing, so we've had our hand slightly forced on the kit and overall topology involved, however that's all fine and we can make it work.
This is a collapsed core topology with core and access switches, split over 3 blocks (fibre connections between), one core switch/stack is in block B and the other in block C, with access switches throughout.
They require all access switches to be connected to the Core in B and the Core in C, and then obviously cross connects between the two cores.
They state:
"Core switches shall be linked with 2x 1Gbps links bonded into a standard compliant Etherchannel"
"Uplinks between access and core switches shall be non-blocking - for example equal cost load balancing at layer 3, or layer 2 bonded multi-chassis Etherchannel"
The specced kit for the core are 3850's, in an ideal world I'd use VSS (Virtual Switch System) to achieve the above statements beyond repute; but this is only supported on 4500/6500 and Nexus platforms.
Do we think a cross stack etherchannel (LACP between both core switch stacks) would satisfy the above statements? Or the statements may just be badly worded...
I look forward to your thoughts and views on this! Thanks!Disclaimer
The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.
Liability Disclaimer
In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.
Posting
As the others have noted, the 3850s, to stack, are restricted to the length of the longest stack cables.
As you have noted, VSS physical units would allow the "logical" unit to be far apart.
For a "small" VSS core, the 4500-X might be an idea unit. (Other than cost, the 4500 would be a better choice for a core device.)
Something to watch for, or understand, when running VSS, Etherchannel doesn't load balance as it does on a single chassis or stack. VSS will avoid using the VSL cross link unless it must.
As many access switches, today, support basic L3 routing, you might also determine whether a L3 edge would be a suitable alternative choice. It would allow retention of the 3850s and can offer some advantages even over VSS. (Where VSS is very nice [as too the Nexus] supporting servers with Etherchannels.) -
Failover clustering Hyper-V + load balancing
I have successfully setup a number of Hyper-V (5) on Server 2008 R2 fail-over cluster environment with two nodes connected to a SAN using MS iSCSI Target 3.3 with cluster shared volume and all is working well.
Now each Hyper-v is run by node-A or node-B. is there a way to have both nodes process all Hyper-Vs at once?
MikeYes you are right I was not using the proper term. I have Hyper-V installed on both nodes and added the Hyper-V service in the Fail-over Clustering service. I can run the VMs on either nodes and distribute the 5 VMs but instead of selecting which node to
run each VM I would like to run all 5 VMs in one environment where both servers (nodes) can manage the processes of all 5 VMs. In other words each server have 64GB RAM and 16 core CPUs, I would like to have my 5 VMs processed by 128GBRAM and 32 core CPU. My
understanding is I am looking at Load balancing, I have seen how to do this with a web application but couldn't find anything on Load Balancing VMs.
You need to configure CSV on top of shared LUN, put your VM images there and run VMs on every host having access to LUN / CSV. See:
http://technet.microsoft.com/en-us/library/dd630633(v=ws.10).aspx
http://technet.microsoft.com/en-us/library/gg610692.aspx
Load balancing is another thing, you can use Live Migration to move VMs from one host to another (no downtime). See:
http://blogs.technet.com/b/iftekhar/archive/2010/01/18/microsoft-virtualization-hyper-v-live-migration-network-configuration.aspx
You can configure HA so one VM would re-boot on another node if primary host is down (minimal downtime). See:
http://blogs.technet.com/b/canitpro/archive/2011/04/18/creating-ha-vms-for-hyper-v-with-failover-clustering-using-free-microsoft-iscsi-target-3-3.aspx
You can configure guest VM cluster so VM would transparently failover to another host if primary one is down (no downtime). See:
http://blogs.technet.com/b/mghazai/archive/2009/12/12/hyper-v-guest-clustering-step-by-step-guide.aspx
http://blogs.technet.com/b/keithmayer/archive/2013/03/21/virtual-machine-guest-clustering-with-windows-server-2012-become-a-virtualization-expert-in-20-days-part-14-of-20.aspx#.Uss8YvRdWSo
Hope this helped :)
StarWind VSAN [Virtual SAN] clusters Hyper-V without SAS, Fibre Channel, SMB 3.0 or iSCSI, uses Ethernet to mirror internally mounted SATA disks between hosts. -
Apache 2 + Sun Java 8.1 load balancer connector dies under load
Hi,
We are testing the apache module "mod_loadbalancer.so" for connecting an apache 2 to a Sun Java 8.1 App Servers running on a different box.
When we perform a basic load testing (60 concurrent users) we are experiencing a very high CPU usage: apache's CPU consumption is around 30-40%per child.
This generates or very slow responses from the App Server or at some point the apache thread die, and therefore, lose of connection to the app server, generating an internal server error.
If we perform this same test directly to the App Server httpd listener, it bahaves fine. We have also performmed the same test againt static content of apache, and again the everything is ok, we even have tried using iWS instead of apache, and the CPU does not go over 4%, so it looks like the problem resides on the apache connector.
Have any of you experienced something similar?
This is the basic conf:
Web Server:
SunOS 5.9 Generic_118558-19 sun4u sparc SUNW,UltraSPARC-IIi-cEngineApache 2.0.54 with MPM worker (threads support) compiled with gcc
apache:~>/apps/WebServer/apache/2.0/bin/httpd -v
Server version: Apache/2.0.54
Server built: May 23 2006 07:10:19
apache:~>/apps/WebServer/apache/2.0/bin/httpd -l
Compiled in modules:
core.c
mod_access.c
mod_auth.c
mod_log_config.c
mod_env.c
worker.c
http_core.c
mod_mime.c
mod_negotiation.c
mod_dir.c
mod_actions.c
mod_alias.c
mod_so.c
httpd.conf:
# Sun AppServer Load Balancer Configuration #
LoadFile /usr/lib/libCstd.so.1
LoadModule apachelbplugin_module /apps/WebServer/apache/2.0/modules/mod_loadbalancer.so
config-file /apps/WebServer/apache/apache2-sunsecd1/conf/loadbalancer.xml
locale en
loadbalancer.xml:
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE loadbalancer PUBLIC "-//Sun Microsystems Inc.//DTD Sun ONE Application Server 7.1//EN" "sun-loadbalancer_1_1.dtd">
<loadbalancer>
<cluster name="sunsecd1">
<instance disable-timeout-in-minutes="30" enabled="true" listeners="http://appserver:38082" name="sunsecd1"/>
<web-module context-root="petstore" disable-timeout-in-minutes="30" enabled="true"/>
<web-module context-root="opc" disable-timeout-in-minutes="30" enabled="true"/>
<web-module context-root="admin" disable-timeout-in-minutes="30" enabled="true"/>
<web-module context-root="supplier" disable-timeout-in-minutes="30" enabled="true"/>
<health-checker interval-in-seconds="30" timeout-in-seconds="10" url="/"/>
</cluster>
<property name="response-timeout-in-seconds" value="60"/>
<property name="reload-poll-interval-in-seconds" value="60"/>
<property name="https-routing" value="false"/>
<property name="require-monitor-data" value="false"/>
<property name="route-cookie-enabled" value="true"/>
</loadbalancer>Hi Balaji,
The Apache webserver runs on Solaris 9 on a Netra T1:
$ /usr/platform/`uname -i`/sbin/prtdiag
System Configuration: Sun Microsystems sun4u Netra t1 (UltraSPARC-IIi 440MHz)
System clock frequency: 110 MHz
Memory size: 1024 Megabytes
========================= CPUs =========================
Run Ecache CPU CPU
Brd CPU Module MHz MB Impl. Mask
0 0 0 440 2.0 12 9.1
========================= IO Cards =========================
Bus# Freq
Brd Type MHz Slot Name Model
0 PCI-1 33 1 ebus
0 PCI-1 33 1 network-SUNW,hme
0 PCI-1 33 2 scsi-glm Symbios,53C875
0 PCI-1 33 3 network-SUNW,hme
0 PCI-2 33 1 pci-pci1011,22The Sun App Server is Sun Java System Application Server Enterprise Edition 8.1_02 (build b11-fcs)
Did I miss anything else? -
Does SL really utilize both CPU cores?
I've been wondering about the 32/64 bit and multi-core technologies. So we've had the availability of 64bit technology for over two years, yet the new O/S only now exploits 64bit technology. I ran multiple tasks; video, internet speed tests, etc. and monitored via Activity Monitor. It didn't seem to me SL uses both cores independently. Does anyone know if SL uses both CPU cores for multiple independent tasks or processes?
To All,
I would like to add an addendum to Edmund's latest post, as quoted below. This will, perhaps, illuminate what will occur now, and the possibilities for the near future...
2) as pointed out by someone else if app used Grand Central services. Can't imagine many do yet.
Again keep in mind MULTIPLE apps, even if single threaded, will be assigned to balance out core/CPU loads. And the kernel/OS itself will use the multiple cores for various services used by the applications.
This is all true, and I would like to highlight that no third-party applications yet take advantage of GCD (none of which I am aware).
However, we can almost certainly assume that many of the underlying components of SL, and probably all of those newly "Cocoa-ized" SL applications (including the Finder) will be taking advantage of Grand Central. The benefits will apply directly to all the "native" processes, and will carry over to third-party processes even if only by virtue of the native ones "getting out of the way."
If you have 3 single-threaded processes actively consuming CPU cycles, and you have two cores, the ability to "packetize" and dynamically dispatch even one of them allows the other two to be handled more efficiently and smoothly.
We haven't even started mentioning OpenCL. This is still waiting in the wings, for the most part, but what I have seen so far is... almost shocking. There are a few rudimentary "benchmark" apps floating about that do use OpenCL (all command line), and what they show is that my 8600M GT will complete in less than 2 seconds a "general computing" calculation that it takes my 2.5 GHz "Penryn" CPU 14 seconds to complete (using both cores!).
So, the question becomes: "How much of Snow Leopard is taking advantage of OpenCL?" I think none of it. Not yet. Whether or not we begin to see the leveraging of OpenCL in updates to SL remains to be seen. Let's not step over the line with speculation, here
Scott -
Z77A-GD65 Gaming CPU Core Speed Lock 1.6Ghz
Hello everyone, this is my first time using a MSI board at first I'm using the default setting in BIOS and the core speed at idle is at 1.6GHz which is fine. However, when the CPU cores are at full load it still remains at 1.6GHz.
I've done a bit of research tried different settings with EIST, Intel Turbo Boost, Enhanced Turbo, Adjust CPU Ratio, C1E, but it is not working.
Just to clarify the power settings in Windows is set to balanced.
Here's my rig's spec:
Motherboard: Z77A-GD65 Gaming
CPU: Intel 2600k
RAM: Corsair vengeance 16GB kit
VGA: AMD 290x
Storage: Corsair Force GT 120GB
OS: tried both Win7 and Win8 64bit
Thanks in advance
P.S. I'm away from home at the moment if needed I can take some screen caps from the BIOS.For initial setup, establish everything works correctly at stock/auto/default on all the OC settings. Make sure EuP 2013 and Spread Spectrum are disabled. EIST and C1E enabled is fine, and just try with standard Turbo enabled and Enhanced Turbo disabled. Might try with C-State disabled. Download the utility Intel Burn Test (IBT) and see if the CPU speed ramps up to the Turbo speed.
-
Load Balancing Directory Servers with Access Manager - Simple questions
Hi.
We are in the process of configuring 2 Access Manager instances (servers) accessing the same logical LDAP repository (comprising physically of two Directory Servers working together with Multi-Master Replication configured and tested) For doing this, we are following guide number 819-6258.
The guide uses BigIP load balancer for load balancing the directory servers. However, we intend to use Directory Proxy Server. Since we faced some (unresolved) issues last time that we used DPS, there are some simple questions that I would be very grateful to have answers to:
1. The guide, in section 3.2.10 (To configure Access Manager 1 with the Directory Server load balancer), talks about making changes at 4 places, and replacing the existing entry (hostname and port) with the load balancer's hostname and port (assuming that the load balancer has already been configured). It says that changes need not be made on Access Manager 2 since the LDAPs are in replication, and hence changes will be replicated at all places. However, the guide also states that changes have to be made in two files, namely AMConfig.properties, and the serverconfig.xml file. But these changes will not be reflected on Access Manager 2, since these files are local on each machine.
Question 1. Do changes have to be made in AMConfig.properties and serverconfig.xml files on the other machine hosting Access Manager 2?
Question 2: What is the purpose of putting these values here? Specifically, what is achieved by specifying the Directory server host and port in AMConfig.properties, as well as in serverconfig.xml?
Question 3. In the HTTP console, there is the option of specifying multiple primary LDAP servers, as well as multiple secondary LDAP servers. What is the purpose of these? Are secondary servers attempted when none of the list in the primary list are accessible? Also, if there are multiple entries in the primary server list, are they accessed in a round robin fashion (hereby providing rudimentary load balancing), or are other servers accessed only when the one mentioned first is not reachable etc.?
2. Since I do not have a load balancer setup yet, I tried the following deviation to the above, which, according to me, should have worked. If viewed in the HTTP console, LDAP / Membership / MSISDN and Policy configuration all pointed to the DS on host 1. When I changed all these to point to the directory server on host 2 (and made AMConfig.properties and serverconfig.xml on host 1 point to DS of host 2 as well), things should have worked fine, but apparently Access manager 1 could not be started. Error from Webserver:
[14/Aug/2006:04:30:36] info (13937): WEB0100: Loading web module in virtual server [https-machine_1_FQDN] at [search]
[14/Aug/2006:04:31:48] warning (13937): CORE3283: stderr: Exception in thread "EventService" java.lang.ExceptionInInitializerError
[14/Aug/2006:04:31:48] warning (13937): CORE3283: stderr: at com.iplanet.services.ldap.event.EventServicePolling.run(EventServicePolling.java:132)
[14/Aug/2006:04:31:48] warning (13937): CORE3283: stderr: at java.lang.Thread.run(Thread.java:595)
[14/Aug/2006:04:31:48] warning (13937): CORE3283: stderr: Caused by: java.lang.InterruptedException
[14/Aug/2006:04:31:48] warning (13937): CORE3283: stderr: at com.sun.identity.sm.ServiceManager.<clinit>(ServiceManager.java:74)
[14/Aug/2006:04:31:48] warning (13937): CORE3283: stderr: ... 2 more
In effect, AM on 1 did not start. On rolling back the changes, things again worked like previously.
Will be really grateful for any help / insight / experience on dealing with the above.
Thanks!Update to the above, incase anyone is reading:
We setup a similar setup in Windows, and it worked. Here is a detailed account of what was done:
1. Host 1: Start installer, install automatically, chose Directory server, Directory Administration server, Directory Proxy server, Web server, Access Manager.
All installed, and worked fine. (AMConfig.properties, serverconfig.xml, and the info in LDAP service, all pointed to HOST1:389)
2. Host 2: Start installer, install automatically, chose Directory server, Directory Administration server, Directory Proxy server, Web server, Access Manager.
All installed, and worked fine. (AMConfig.properties, serverconfig.xml, and the info in LDAP service, all pointed to HOST2:389)
3. Host 1: Started replication. Set to Master
4. Host 2: Started replication. Set to Master
5. Host 1: Setup replication agreement to Host 2
6. Host 2: Setup replication agreement to Host 1
7. Initiated the remote replica from Host 1 ----> Host 2
Note that since default installation uses abc.....xyz as the encryption key, setting this to same was not an issue.
9. Started webserver for Host 1 and logged into AM as amadmin.
10. Added Host 2 FQDN in DNS Aliases / Realms
11. Added http://HOST2_FQDN:80 in the Platform server (instance) list.
12. Started Host 2 webserver. Logged in AM on Host 2, things worked fine.
At this stage, note the following:
a) Host 1:
AMConfig.properties file has
com.iplanet.am.directory.host=host1_FQDN
and
com.iplanet.am.directory.port=389
serverconfig.xml has:
<Server name="Server1" host="host1_FQDN" port="389" type="SIMPLE" />
b) Host 2:
AMConfig.properties file has
com.iplanet.am.directory.host=host2_FQDN
and
com.iplanet.am.directory.port=389
serverconfig.xml has:
<Server name="Server1" host="host2_FQDN" port="389" type="SIMPLE" />
c) If one logs into AM, and checks LDAP servers for LDAP / Policy Configuration / Membership etc services, they all contain Host2_FQDN:389 (which makes sense, since replica 2 was initialized from 1)
Returning back to the configuations:
13. On Host 1, login into the Admin server console of the Directory server. Navigate to the DPS, and confgure the following:
a) Network Group
b) LDAP servers
c) Load Balancing
d) Change Group
e) Action on-bind
f) Allow all actions (permit modification / deletion etc.).
g) any other configuations required - Am willing to give detailed steps if someone needs them to help me / themselves! :)
So now, we have DPS configured and running on Host1:489, and distributing load to DS1 and DS2 on a 50:50 basis.
14. Now, log into AM on Host 1, and instead of Host1_fqdn:389 (for DS) in the following places, specify Host1_fqdn:489 (for the DPS)--
LDAP Authentication
MSISDN server
Membership Service
Policy configuation.
Verified that this propagated to the Policy Configuration service and the LDAP authentication service that are already registered with the default organization.
15. Log out of AM. Following the documentation, modify directory.host and directory.port in AMConfig.properties to point to Host 1_FQDN and 489 respectively. Make this change in AMConfig.properties of both Host 1 as well as 2.
16. Edit serverconfig.xml on both hosts, and instead of they pointing to their local directory servers, point both to host1_FQDN:489
17. When you start the webserver, it will refuse to start. Will spew errors such as:
[https-host1_FQDN]: Sun ONE Web Server 6.1SP5 B06/23/2005 17:36
[https-host1_FQDN]: info: CORE3016: daemon is running as super-user
[https-host1_FQDN]: info: CORE5076: Using [Java HotSpot(TM) Server VM, Version 1.5.0_04] from [Sun Microsystems Inc.]
[https-host1_FQDN]: info: WEB0100: Loading web module in virtual server [https-host1_FQDN] at [amserver]
[https-host1_FQDN]: warning: WEB6100: locale-charset-info is deprecated, please use parameter-encoding
[https-host1_FQDN]: info: WEB0100: Loading web module in virtual server [https-host1_FQDN] at [ampassword]
[https-host1_FQDN]: warning: WEB6100: locale-charset-info is deprecated, please use parameter-encoding
[https-host1_FQDN]: info: WEB0100: Loading web module in virtual server [https-host1_FQDN] at [amcommon]
[https-host1_FQDN]: info: WEB0100: Loading web module in virtual server [https-host1_FQDN] at [amconsole]
[https-host1_FQDN]: warning: WEB6100: locale-charset-info is deprecated, please use parameter-encoding
[https-host1_FQDN]: info: WEB0100: Loading web module in virtual server [https-host1_FQDN] at [search]
[https-host1_FQDN]: warning: CORE3283: stderr: netscape.ldap.LDAPException: error result (32); matchedDN = dc=sun,dc=com; No such object (DN changed)
[https-host1_FQDN]: warning: CORE3283: stderr: Got LDAPServiceException code=-1
[https-host1_FQDN]: warning: CORE3283: stderr: at com.iplanet.services.ldap.DSConfigMgr.getConnection(DSConfigMgr.java:357)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.iplanet.services.ldap.DSConfigMgr.getNewFailoverConnection(DSConfigMgr.java:314)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.iplanet.services.ldap.DSConfigMgr.getNewConnection(DSConfigMgr.java:253)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.iplanet.services.ldap.DSConfigMgr.getNewProxyConnection(DSConfigMgr.java:184)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.iplanet.services.ldap.DSConfigMgr.getNewProxyConnection(DSConfigMgr.java:194)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.iplanet.ums.DataLayer.initLdapPool(DataLayer.java:1248)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.iplanet.ums.DataLayer.(DataLayer.java:190)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.iplanet.ums.DataLayer.getInstance(DataLayer.java:215)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.iplanet.ums.DataLayer.getInstance(DataLayer.java:246)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.sun.identity.sm.ldap.SMSLdapObject.initialize(SMSLdapObject.java:156)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.sun.identity.sm.ldap.SMSLdapObject.(SMSLdapObject.java:124)
[https-host1_FQDN]: warning: CORE3283: stderr: at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
[https-host1_FQDN]: warning: CORE3283: stderr: at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:39)
[https-host1_FQDN]: warning: CORE3283: stderr: at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:27)
[https-host1_FQDN]: warning: CORE3283: stderr: at java.lang.reflect.Constructor.newInstance(Constructor.java:494)
[https-host1_FQDN]: warning: CORE3283: stderr: at java.lang.Class.newInstance0(Class.java:350)
[https-host1_FQDN]: warning: CORE3283: stderr: at java.lang.Class.newInstance(Class.java:303)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.sun.identity.sm.SMSEntry.(SMSEntry.java:216)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.sun.identity.sm.ServiceSchemaManager.(ServiceSchemaManager.java:67)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.iplanet.am.util.AMClientDetector.getServiceSchemaManager(AMClientDetector.java:219)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.iplanet.am.util.AMClientDetector.(AMClientDetector.java:94)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.sun.mobile.filter.AMLController.init(AMLController.java:85)
[https-host1_FQDN]: warning: CORE3283: stderr: at org.apache.catalina.core.ApplicationFilterConfig.getFilter(ApplicationFilterConfig.java:262)
[https-host1_FQDN]: warning: CORE3283: stderr: at org.apache.catalina.core.ApplicationFilterConfig.setFilterDef(ApplicationFilterConfig.java:322)
[https-host1_FQDN]: warning: CORE3283: stderr: at org.apache.catalina.core.ApplicationFilterConfig.(ApplicationFilterConfig.java:120)
[https-host1_FQDN]: warning: CORE3283: stderr: at org.apache.catalina.core.StandardContext.filterStart(StandardContext.java:3271)
[https-host1_FQDN]: warning: CORE3283: stderr: at org.apache.catalina.core.StandardContext.start(StandardContext.java:3747)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.iplanet.ias.web.WebModule.start(WebModule.java:251)
[https-host1_FQDN]: warning: CORE3283: stderr: at org.apache.catalina.core.ContainerBase.start(ContainerBase.java:1133)
[https-host1_FQDN]: warning: CORE3283: stderr: at org.apache.catalina.core.StandardHost.start(StandardHost.java:652)
[https-host1_FQDN]: warning: CORE3283: stderr: at org.apache.catalina.core.ContainerBase.start(ContainerBase.java:1133)
[https-host1_FQDN]: warning: CORE3283: stderr: at org.apache.catalina.core.StandardEngine.start(StandardEngine.java:355)
[https-host1_FQDN]: warning: CORE3283: stderr: at org.apache.catalina.startup.Embedded.start(Embedded.java:995)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.iplanet.ias.web.WebContainer.start(WebContainer.java:431)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.iplanet.ias.web.WebContainer.startInstance(WebContainer.java:500)
[https-host1_FQDN]: warning: CORE3283: stderr: at com.iplanet.ias.server.J2EERunner.confPostInit(J2EERunner.java:161)
[https-host1_FQDN]: failure: WebModule[amserver]: WEB2783: Servlet /amserver threw load() exception
[https-host1_FQDN]: javax.servlet.ServletException: WEB2778: Servlet.init() for servlet LoginLogoutMapping threw exception
[https-host1_FQDN]: at org.apache.catalina.core.StandardWrapper.loadServlet(StandardWrapper.java:949)
[https-host1_FQDN]: at org.apache.catalina.core.StandardWrapper.load(StandardWrapper.java:813)
[https-host1_FQDN]: at org.apache.catalina.core.StandardContext.loadOnStartup(StandardContext.java:3478)
[https-host1_FQDN]: at org.apache.catalina.core.StandardContext.start(StandardContext.java:3760)
[https-host1_FQDN]: at com.iplanet.ias.web.WebModule.start(WebModule.java:251)
[https-host1_FQDN]: at org.apache.catalina.core.ContainerBase.start(ContainerBase.java:1133)
[https-host1_FQDN]: at org.apache.catalina.core.StandardHost.start(StandardHost.java:652)
[https-host1_FQDN]: at org.apache.catalina.core.ContainerBase.start(ContainerBase.java:1133)
[https-host1_FQDN]: at org.apache.catalina.core.StandardEngine.start(StandardEngine.java:355)
[https-host1_FQDN]: at org.apache.catalina.startup.Embedded.start(Embedded.java:995)
[https-host1_FQDN]: at com.iplanet.ias.web.WebContainer.start(WebContainer.java:431)
[https-host1_FQDN]: at com.iplanet.ias.web.WebContainer.startInstance(WebContainer.java:500)
[https-host1_FQDN]: at com.iplanet.ias.server.J2EERunner.confPostInit(J2EERunner.java:161)
[https-host1_FQDN]: ----- Root Cause -----
[https-host1_FQDN]: java.lang.NullPointerException
[https-host1_FQDN]: at com.sun.identity.authentication.UI.LoginLogoutMapping.init(LoginLogoutMapping.java:71)
[https-host1_FQDN]: at org.apache.catalina.core.StandardWrapper.loadServlet(StandardWrapper.java:921)
[https-host1_FQDN]: at org.apache.catalina.core.StandardWrapper.load(StandardWrapper.java:813)
[https-host1_FQDN]: at org.apache.catalina.core.StandardContext.loadOnStartup(StandardContext.java:3478)
[https-host1_FQDN]: at org.apache.catalina.core.StandardContext.start(StandardContext.java:3760)
[https-host1_FQDN]: at com.iplanet.ias.web.WebModule.start(WebModule.java:251)
[https-host1_FQDN]: at org.apache.catalina.core.ContainerBase.start(ContainerBase.java:1133)
[https-host1_FQDN]: at org.apache.catalina.core.StandardHost.start(StandardHost.java:652)
[https-host1_FQDN]: at org.apache.catalina.core.ContainerBase.start(ContainerBase.java:1133)
[https-host1_FQDN]: at org.apache.catalina.core.StandardEngine.start(StandardEngine.java:355)
[https-host1_FQDN]: at org.apache.catalina.startup.Embedded.start(Embedded.java:995)
[https-host1_FQDN]: at com.iplanet.ias.web.WebContainer.start(WebContainer.java:431)
[https-host1_FQDN]: at com.iplanet.ias.web.WebContainer.startInstance(WebContainer.java:500)
[https-host1_FQDN]: at com.iplanet.ias.server.J2EERunner.confPostInit(J2EERunner.java:161)
[https-host1_FQDN]:
[https-host1_FQDN]: info: HTTP3072: [LS ls1] http://host1_FQDN:58080 [i]ready to accept requests
[https-host1_FQDN]: startup: server started successfully
Success!
The server https-host1_FQDN has started up.
The server infact, didn't start up (nothing even listening on 58080).
However, if AMConfig.properties is left as it originally was, and only serverconfig.xml files were changed as mentioned above, web servers started fine, and things worked all okay. (Alright, except for some glitches when viewed in /amconsole. If /amserver/console is accessed, all is good. Can this mean that all is still not well? I am not sure).
So far so good. Now comes the sad part. When the same is done on Solaris 9, things dont work. You continue to get the above error, OR the following error, and the web server will refuse to start:
Differences in Solaris and Windows are as follows:
1. Windows hosts have 1 IP and hostname. Solaris hosts have 3 IPs and hostnames (for DS, DPS, and webserver).
No other difference from an architectural perspective.
Any help / insight on why the above is not working (and why the hell does the documentation seem so sketchy / insecure / incorrect).
Thanks a bunch!
Maybe you are looking for
-
Having to re-add Music folder to iTunes everytime
Everytime I open iTunes on my somewhat new Sony laptop, I have to use "Add Folder to Library" to re-add my Music folder. For whatever reason, my iTunes library is empty otherwise upon turning on my laptop. I lose all my playlists and play counts from
-
How to get the functionality of FM 'RV_PRICE_PRINT_ITEM' for archived data
Hi All, I am trying to print archived billing data using archiving object SD_VBRK. I am able to print all the related information except the pricing. Apart from reading all the tables archived under the archiving object SD_VBRK, I also need to use t
-
Macbook Pro date setting problem
I have a macbook pro running 10.6. Today, twice, the computer told me that my date and time were set to 2008. I was able to reset it, but I'm afraid it ill do it again. What would cause it to suddenly lose the date and time settings? And what should
-
Dear Support Team, i am having the error ''The specified forest functional level is invalid. "Lync Server" requires forests running in Windows 2003 mode or higher'' from lync 2013 during the schema master prepare on windows server 2008r2 and my fores
-
Is Leopard a free upgrade to Tiger?
Hi. I'm coming from the Windows world and would be moving semi-full time to Mac OS X. I'm just wondering if the Leopard release is a free upgrade to Tiger or is it a completely new upgrade similar to going from Windows XP to Vista.