Startup timeout and packet-delivery timeout

Hi,
At the moment it takes my first cluster node approximately 30 seconds to start and setting the packet-delivery timeout smaller than this means the system cannot start. I'm trying to reduce the packet-delivery setting to improve responsiveness during failover caused by hardware failures. I think 15-20 seconds would be ideal, allowing for GC pauses.
Subsequent nodes can start in a couple of seconds.
Is this a reasonable time to expect the first Coherence node in a cluster to start up? What kind of values is everyone else working with?
Thanks & Regards,
Martin
The error when packet-delivery is less than 30 seconds:
2009-11-24 13:04:05.568/24.141 Oracle Coherence GE 3.4/405 <Error> (thread=main, member=n/a): Error while starting cluster: com.tangosol.net.RequestTimeoutException: Timeout during service start: ServiceInfo(Id=0, Name=Cluster, Type=Cluster
MemberSet=ServiceMemberSet(
OldestMember=n/a
ActualMemberSet=MemberSet(Size=0, BitSetCount=0
MemberId/ServiceVersion/ServiceJoined/ServiceLeaving
     at com.tangosol.coherence.component.util.daemon.queueProcessor.service.Grid.onStartupTimeout(Grid.CDB:6)
     at com.tangosol.coherence.component.util.daemon.queueProcessor.Service.start(Service.CDB:27)
     at com.tangosol.coherence.component.util.daemon.queueProcessor.service.Grid.start(Grid.CDB:38)
     at com.tangosol.coherence.component.net.Cluster.onStart(Cluster.CDB:317)
     at com.tangosol.coherence.component.net.Cluster.start(Cluster.CDB:11)
     at com.tangosol.coherence.component.util.SafeCluster.startCluster(SafeCluster.CDB:3)
     at com.tangosol.coherence.component.util.SafeCluster.restartCluster(SafeCluster.CDB:7)
     at com.tangosol.coherence.component.util.SafeCluster.ensureRunningCluster(SafeCluster.CDB:27)
     at com.tangosol.coherence.component.util.SafeCluster.start(SafeCluster.CDB:2)
     at com.tangosol.net.CacheFactory.ensureCluster(CacheFactory.java:951)
     at com.tangosol.net.DefaultConfigurableCacheFactory.ensureService(DefaultConfigurableCacheFactory.java:748)
     at com.tangosol.net.DefaultConfigurableCacheFactory.ensureCache(DefaultConfigurableCacheFactory.java:710)
     at com.tangosol.net.DefaultConfigurableCacheFactory.configureCache(DefaultConfigurableCacheFactory.java:919)
     at com.tangosol.net.DefaultConfigurableCacheFactory.ensureCache(DefaultConfigurableCacheFactory.java:277)
     at com.tangosol.net.CacheFactory.getCache(CacheFactory.java:689)
     at com.tangosol.net.CacheFactory.getCache(CacheFactory.java:667)
     at com.changingworlds.datagrid.cache.Cache.init(Cache.java:111)
     at com.changingworlds.datagrid.cache.Cache.initializeNamedCache(Cache.java:95)
     at com.changingworlds.discovery.DiscoveryMain.initSpring(DiscoveryMain.java:201)
     at com.changingworlds.discovery.DiscoveryMain.createDiscovery(DiscoveryMain.java:154)
     at com.changingworlds.discovery.DiscoveryMain.main(DiscoveryMain.java:78)
Edited by: MartinMc on Nov 24, 2009 1:15 PM

Hi Martin,
The packet delivery timeout relates to death detection, not to cluster formation. You'll want to have a look at the join-timeout-milliseconds specified within the multicast-listener element (see http://coherence.oracle.com/display/COH35UG/multicast-listener) if you wish to change the amount of time it takes to form a new cluster. The reason for this timeout is to prevent a new node from accidentally forming a secondary cluster if the existing cluster members are temporarily unreachable while the new node starts. Assuming you are running more then just a few nodes in your cluster, you should be fine lowering this value to 5-10s.
I don't however see how this relates to failover unless by failover you mean starting an entirely new cluster after the complete loss of the formerly running cluster.
thanks,
Mark
Oracle Coherence

Similar Messages

New in 3.4 - ReplicatedCache startup timeout triggered while resyncing

Something changed from Coherence 3.3 -> 3.4 in handling or setting-up the start-up timeout of the ReplicatedCache service. There was no problem in 3.3 but in 3.4 the service start-up times out while resyncing caches. I found out that the time it takes to time-out is equal to the cluster's packet publisher's resend time-out (which according to recommendations should be set to max(60 sec, 2 * greatest GC pause). In my situation I set it to 5 minutes (with greatest GC pauses of about 1 minute or so). But with large caches this is not nearly enough (it takes up to 30 minutes to resync caches in my situation).
I worked arround the problem by creating a subclass of ReplicatedCache with an overriden getStartupTimeount() method and configuring it in the tangosol-coherence-override.xml in place of the default ReplicatedCache service.
Maybe a "startup-timeout" configuration parameter should be adequate here.
Peter

Hi Peter,
Yes, this is indeed a new "deterministic" behavior in 3.4. I will open a corresponding enhancement request.
The important question though is: why does it take so long to replicate the caches in a first place. I just tested a replicated cache with 200,000 objects (not optimized Java serialization) totaling 100MB and it took about 20 seconds to start a new node (relatively slow development machines). There must be something else going on in your environment if it takes so long to start a replicated cache service. Do you see any "remote GC" messages? What is the Publisher success rate during the start? Did you run the datagram test?
Regards,
Gene
Correction: I was wrong in the original suggestion that the service would not be stopped.
Edited by: ggleyzer on Oct 15, 2008 2:33 PM

Msg regarding "Dgraph did not start in startup timeout of 120 seconds"

Hi All,
Baseline update failed with following error message.
"WARNING: Component 'Dgraph1' did not start in startup timeout of 120 seconds.
SEVERE: Server component 'Dgraph1' did not start in the allotted startup time.Refer to component logs in /usr/local/endeca/EndAppR7/./logs/dgraphs/Dgraph1 on host MDEXHost1.
Occurred while executing line 5 of valid BeanShell script:
2|
3| DgraphCluster.cleanDirs();
4| DgraphCluster.copyIndexToDgraphServers();
5| DgraphCluster.applyIndex();
6|
7|
I've checked the dgraph1 logs and i don't see any difference than usual logs - like what i had yesterday.
"all dgraph transactions completed"
"Shutting down dgraph (pid=17398)"
It had been running well as daily scheduled job since last 1 week.
it has very small data to be indexed, just few MB.
I've checked main application log, dgraphs logs, forge logs just in case, and process.0.log in PlatformServices. I don't see any useful detailed information to find out what causes this issue.
And there was no locked flag on the application.
to resolve this issue, i just started the Dgraph1 from Workbench and run the baseline update to make sure if it does not fail again. it ran well.
But i'm wondering why... it happened.
Can anyone kindly tell me the possible cause or where i can find out the root cause?
Thanks,
Edited by: julia on Oct 31, 2012 11:06 AM

Julia,
After you issue the start command (either as part of a baseline update or with a runcommand.sh Dgraph1 start), you should see your Dgraph1.log file (in [your_app_folder]/data/dgraphs/Dgraph1/) get updated with information related to dgraph startup.
You should look there to see if your dgraph is actually starting in the 2 minute timeframe, the dgraph is up when you see the following line:
pid=[SOME_NUMBER] listening for HTTP connections on port [SOME_NUMBER] at [SOME_DATE_TIME]
If you don't see that, there is either an error starting the dgraph (you should see that in the same log file) or an error sending the command to start the dgraph (port 8888 not open, EAC Agent not running on the MDEX server, etc.).
If you do see a successful message starting the dgraph, it's likely that the EAC Central Server (i.e. likely where you ran the command) was not able to determine that the dgraph started. It usually does this by issuing an admin?op=ping to the dgraph port so it's likely that port 15000 (or whereever your dgraph is running) is blocked between those two servers.
Hopefully, that's enough for you to go on.
Patrick

Mac won't startup anymore and gives IP packet massage?

My mac wont startup anymore and gives this massage:
darwin/BSD (login) (console)
IP packet filtering initialized divert enabled loging disabled.
IPv6 packet
IP firewall loaded.
how can i fix this or do i have to install Osx again? HELP please thnx.
Powerbook G4 1GHZ Mac OS X (10.3.9) -
Powerbook G4 1GHZ Mac OS X (10.3.9) -

In this case I would try reformatting the drive:
Extended Hard Drive Preparation
1. Boot from your OS X Installer Disk. After the installer loads select your language and click on the Continue button. When the menu bar appears select Disk Utility from the Installer menu (Utilities menu for Tiger.)
2. After DU loads select your hard drive (this is the entry with the mfgr.'s ID and size) from the left side list. Note the SMART status of the drive in DU's status area. If it does not say "Verified" then the drive is failing or has failed and will need replacing. Otherwise, click on the Partition tab in the DU main window.
3. Set the number of partitions from the dropdown menu (use 1 partition unless you wish to make more.) Set the format type to Mac OS Extended (Journaled.) Click on the Partition button and wait until the volume(s) mount on the Desktop.
4. Select the volume you just created (this is the sub-entry under the drive entry) from the left side list. Click on the Erase tab in the DU main window.
5. Set the format type to Mac OS Extended (Journaled.) Click on the Options button, check the button for Zero Data and click on OK to return to the Erase window.
6. Click on the Erase button. The format process will take 30 minutes to an hour or more depending upon the drive size.

High ping and packet loss.

How do i fix high ping and packet loss? Also when i ping my own ip, it results in 100.0% packet loss.

Please read this whole message before doing anything.
This procedure is a diagnostic test. It’s unlikely to solve your problem. Don’t be disappointed when you find that nothing has changed after you complete it.
The purpose of the test is to determine whether the problem is caused by third-party software that loads automatically at startup or login, by a peripheral device, by a font conflict, or by corruption of the file system or of certain system caches.
Disconnect all wired peripherals except those needed for the test, and remove all aftermarket expansion cards, if applicable. Start up in safe mode and log in to the account with the problem. You must hold down the shift key twice: once when you turn on the computer, and again when you log in.
Note: If FileVault is enabled, or if a firmware password is set, or if the startup volume is a software RAID, you can’t do this. Ask for further instructions.
Safe mode is much slower to start up and run than normal, with limited graphics performance, and some things won’t work at all, including sound output and Wi-Fi on certain models. The next normal startup may also be somewhat slow.
The login screen appears even if you usually login automatically. You must know your login password in order to log in. If you’ve forgotten the password, you will need to reset it before you begin.
Test while in safe mode. Same problem?
After testing, restart as usual (not in safe mode) and verify that you still have the problem. Post the results of the test.

Using SQVI to generate report of open and released delivery schedule lines

All,
I'm using SQVI to generate an excel spreadsheet for some buyers to show open released schedule lines because they are a 1 line item per scheduling agreement company.
I used the logical database MEPOLDB instead of a table joint and pulled fields from EKKO(vendor, SA #,&purchasing group), EKPO(Material Number), EKEH(schedule line type), and EKET(delivery date, scheduled qty,previous qty).
Does this sound like I'll get the results I want on paper as long as I use the right selection criteria, because the report I'm getting isn't quite what I expect? I am unable to identify which lines are authorized to ship vs. trade-off zone, planning, etc. in the report thus far.

Hi Mark,
I have faced same requirement. I am not sure about transporting to TST and PROD. I done by this way.
After generating SQVI program in DEV , I assigned that program to a transaction and tested in DEV. Later i have regenarated SQVI in Production. then I assigned the generated Program to same transaction in DEV. And transported the Tcode assignment of program to Production..
About authorization , if its not sensitive report, BASIS can restrict at transaction level.
Regards,
Ravi.

What is difference between Startup upgrade and startup migration.

what is difference between Startup upgrade and startup migration.

I am having the metalink note.
STARTUP MIGRATE was introduced in 9.2 as a mechanism to be sure that most everything that needs to be done to run an upgrade script or a patch script is done automatically. In the past, customers were expected to adjust certain initialization parameters prior to beginning an upgrade or applying a a patch,
but most of this is now done automatically by STARTUP MIGRATE.
When a customer starts a database in MIGRATE mode, the following ALTER SYSTEM
commands will be set automatically:
But i am not sure what exactly the meaning by startup upgrade.
and exactly the startup upgrade and startup migrate

Open production orders and outbound delivery after go-live

Hello All,
I tried to search answer for this question but could not find answer.
Could anybody suggest, what to do with open production orders and outbound delivery after go-live when WM is activated. As, after go-live WM-PP interface will be activated and system should consume material from storage bin specified in control cycle and that will happen for newly created production order after go-live,but what about existing production orders how to update production supply area against work center in production order.
Similarly with outbound delivery how to update WM status after go-live.
Thanks in advance

Thanks to both Jurgen and Sander
Regarding production order yes, one way is to re-read PP master data in production orders, but is that the only way to handle this or do we have any other option?
I was aware of re-read PP master data but it will real time consuming and also if it will problematic for already released or partially confirmed production orders.

Open order and Open delivery in Credit Exposure FD32

Hello Guys,
I am now trying to include Open order and Open delivery values in Credit Exposure FD32. I have configured the following settings, but the Open order and Open delivery values still cannot show in Sales value under Credit Exposure FD32. Can anyone here advise me what i missed?
1) In OVA8 --> 01 Credit Group for Sales Order, I have selected Static Checks and includes Open Orders and Open deliveries.
2) In V/08 --> Pricing Procedure --> Control , I also put "A" in PR00 Sub Total field.
3) In OB45 --> Credit Control Area, I put Update = 000012.
4) In OMO1 --> S066 , Period Split = Day and Updating = "Synchronous update (1)".
Thank you,
Benny

Hello Mrunal and GSL,
Thanks for your paticipation. I found my configurations are correct. I just haven't trigger the update program to update the S066 (Sales value in FD32). To trigger the update program I need to try to create delivery. If I just create a Sales Order and press Save, the PR00 value will not go to S066.
Hope that can help someone else .
Benny

Difference between Scheduling agreement delivery schedule and PO delivery s

Dear All,
Please tell the difference between the Scheduling agreement delivey schedule and PO delivery schdule.
In PO itself, we can give the delivery schedule.
Then what is the purpose of making scheduling agreement and maintain schedule in ME38?
Regards

Hi,
Difference between Scheduling agreement and PO is while PO is a legal document whatever the delivery schedules you mention in PO, you are liable to pay for the entire qty in case of cancellation
In case of scheduling agreement, we have Firm Zone and Tradeoff Zone to provide some cushion against paying for the entire qty you have ordered
Firm zone: (go-ahead for production)
If the date of a schedule line lies within the firm zone, the vendor has the go-ahead to produce the relevant quantity with the assurance that the orderer will bear any and all material and production costs associated with the scheduled quantity in the event of subsequent cancellation by the latter.
Trade-off zone: (go-ahead for procurement of input materials)
If the date of a schedule line lies within the trade-off zone, the vendor has the go-ahead to purchase any input materials necessary to produce the scheduled quantity. The orderer will recompense him in full for such outlay (but not for any other production costs) in the event that he should subsequently cancel the scheduled quantity.
Schedule lines that lie beyond the firm and trade-off zones fall within the planning zone (forecast zone) and are for rough guidance only, being without any commitment on the part of the orderer.
In case of PO if you want to cancel the order, you are liable to pay for the entire quantity, while in case of Scheduling agreement if you want to cancel the schedule and if the days fall in Firm zone or trade off zone or planning zone you have to pay only for raw materials or assembled products according to the above description
Regards
Venkat

Forecast delivery schdule and JIT delivery schedule difference

Hi All,
What isthe difference between Forecast delivery schdule and JIT delivery schedule in JIT scheduling agreement (LZJ)??
I am doing JIT Inbound process now. Is it necesary to update both the schedule lines(Forecast delivery schedule&JIT delivery schedule)??
Thanks,
Anbazhagan E

Hello,
Forecast Schedule updates your planning , i.e. in MD04 you could see system is planning for quantities updated in forecast schedule.
regards,
amol

How to add Total Quantity in Inbound and Outbound Delivery screen

Hi,
I want to add Total quantity field in Inbound and Outbound Delivery screens.
In document flow i can see the line item quantities in ALV Format, but if i select Display the totals above the entry check box
in Change Layout--> Display, i didn't see any totals displayed.
Please help me on this?
Regards
Bhuvana

Hi
If the field is a customer field, see BADI 'LE_SHP_TAB_CUST_HEAD'
Regards
Eduardo

Link between PO and Inbound Delivery - Delivery confirmation

Hi Gurus,
I have a question regarding the link of Purchase Order and Inbound delivery completed indicator.
This is my scenario. I had created a purhase order with qty 10, then i created a Inbound delivery with 8 qty then i did GR with reference to inbound delivery. Then looking at the purchase order , the PO is set to Delv. Completed. Looking also in the doc flow of the inbound delivery, the status is "complete".
Now , my question, if i go to VL31N and input the purchase order , i am still able to see the open qty which is "2". Is there is no way that the SAP will prompt me after i input the PO number in the VL31n as " Delivery completed" message or someting related to that. Since , in the PO its already Delv. completed?
Thanks.

Check the toleance limits set in the delivery tab of purchase order. If the quanytity delivered is with in tolerance limits then system will set delivery completed indicator automatically.
Even though delivery completed indicator set you can further post the GR for remaining qty.
Regards,
B K

Data Source combine - Customer Invoice and Outbound Delivery Item Details

Hello,
I need create report where getting the outbound delivery item details ( like ship date, quantities, product, identified_stock)
and also need invoice number ( customer invoice number) which is available in the Customer Invoice Header or Detail source.
But try to create combined data source not able to join Customer Invoice and Outbound Delivery Item Details.
Since after adding any one of them as first datasource and try to add another data source from above then it doesn't shows up.
Anybody have idea how to get the invoice information (invoice number) and shipping data.
Thanks,
Viral.

Hi Viral,
you only can comine data sources of the same access context. The customer invoice data sources have access context "company" and the outbound delivery sources have "site" or "sales". So unfortunately, you can`t combine these data sources.
Best regards,
Andreas

Regarding additon of Freight charges and Revised Delivery date in Z Trans

Hi All,
My client is using Z Transaction ZSales_Book(Billing report) and he wants to add two extra fields Freight Charges and Revised Delivery date to this report as it is related to VBAP saled document line item and how can we connect it to VBRP Billing document line item can i know the connecting fileds.
As i passed it to ABAPER as a requirement he asked me the same question..........

Hi
Is it is a revised or request delivery date? Assuming it is a request delivery date i am using request terminology
As you want to add two extra feilds freight charges and Request delivery date.
For Freight Charges - If they are item level freight charges then you give the following tables data - VBAP,KOMV
For Revised Delivery Date - use the following data - VBAP, RV45A-ETDAT also
Regards
Srinath

Startup timeout and packet-delivery timeout

Similar Messages

Maybe you are looking for