SFTP related processes do not die if parent process dies.

Java Version: 1.4.2_05
JRE: build 1.4.2_05-b04.
Running on Solaris 9.
Problem:
If the process using the SFTP API dies, the children processes created by the SFTP API do not end.
A side effect of this is that all the sockets the orginal parent had remain open as the child processes inherited them.
The JVM does close STDIN/STDOUT/STDERR to the child if the JVM has less than 1024 filedesriptors open at the time of the Runtime.exec() command.
However, if the JVM has more than 1024 filedesriptors open at the time of the Runtime.exec command the child will
a) inherit all filedescriptors above 1024
b) the STDIN/STDOU/STDERR towards the child is not closed when the JVM exits.
Has anyone seen this before?
I heard this maybe fixed in JDK 1.5. Is this true? Is there a fix on this in 1.4.2?
Your comments are appreciated?

I found the following in http://docs.sun.com/app/docs/doc/816-4555/rfsrefer-134?l=ja&a=view :
In this situation, the SIGLOST signal is posted to the process. The default action for the SIGLOST signal is to terminate the process.
For you to recover from this state, you must restart any applications that had files open at the time of the failure. Note that the following can occur.
- Some processes that did not reopen the file could receive I/O errors.
- Other processes that did reopen the file, or performed the open operation after the recovery failure, are able to access the file without any problems.
Thus, some processes can access a particular file while other processes cannot.
Edited by: Arvind_Srinivasan on Dec 10, 2009 12:33 AM

Similar Messages

  • PAPI: Get processId of subflow process from Activity list of parent process

    Using PAPI or PAPI-WS how can I get the processId of the related process of a subflow activity?
    Specifically, I want to get the processId of all the subflow processes of a parent process provided I only have the parent process ID.
    For example, lets say my project has three processes with the following ID's:
    MainID1
    SubflowID1
    SubflowID2
    Assume my main process is MainID1 and that is only made up of two activities:
    1) a subflow activity that has SubflowID1 as the related process.
    2) a subflow activity that has SubflowID2 as the related process
    So in my case I need to get back "SubflowID1" and "SubflowID2" provided that initially I only know the ID of the main process MainID1.
    With PAPI/PAPI-WS I can get the list of activities of a process. So I can get back a list of two activities of type "SUBFLOW" for MainID1. However the activity object only gives me the ID of the activity itself and of the activity's parent process. It doesn't contain the ID of the subflow related process. Is there a way to do this?
    Please let me know if this needs further explanation. I know it's somewhat convoluted. Thanks!

    So you need a way to traverse through a process hierarchy even though instances may not be in the process? Sounds like you've created some type of recursive process topology and want to find related activities in the sub-process structure even though instances are not in the structure. If so, I've never seen a use case for this and so I wouldn't know how to do it. FindInstanceByFilter is your closest thing you have. But not having instances in the sub-processes to get information about the process hierarchy prohibits this. Perhaps you can plant tracer instances that you can normally filter out from normal work-flow. When you need to obtain a process hierarchy then you just look for these tracers.
    Interesting problem. Sorry I don't have a good answer.
    Edited by: Mark Peterson on Feb 18, 2010 6:08 AM

  • All Process Types not visible in RSPC ( process chain)

    Hi,
    When I go to create a new process chain in RSPC & look for process types to choose from I am not able to see all the process types.
    when I go to table RSPROCESSTYPES or setting-->maintain process types , it shows all the enteries.
    Am I suppose to do some steps before I start seeing all the process types in the left panel of my RSPC--process types.
    looking forward to your help.
    regards
    Vikash

    Hi Agarwal
    check the table RSPCCATEGORY ..In this table check whether there is an entry for the following..
    General services
    Load Process and Post-Processing
    Data Target Administration    
    Reporting Agent               
    Other BW Processes            
    if not there maintain them..
    bye
    Shu Moh..

  • Mac: Firefox software update process will not authorize in parental controlled account-Bug?

    OS: 10.7.5
    Firefox 20
    Parental Controlled user account: Firefox and Firefox Plugin processes are checked in parental controls. No other Firefox processes are visible.
    When logged into parental controlled account, user launches Firefox from the Dock. Message appears:
    You do not have permission to run Firefox Software Update. Admin required authorization.
    Admin clicked Always Allow, entered admin name & password. Firefox would not launch.
    Logged out of Parental Controlled account.
    Logged in to Admin account.
    Launched Firefox. "About Firefox" shows Firefox up to date.
    Logged out.
    Logged back into parental controlled account.
    Firefox opens.
    Is this a bug or is there some other reason the admin authorization failed?

    I have a student that is having the same issue. Whenever she attempts to open Firefox, either from the dock or applications folder, she gets the permission error. I tried this under an admin account and get the same error, and it will not go past it, even with the admin user/pass. A strange icon appears in the dock for a second and the disappears, but Firefox never opens.
    We have around 1600 student laptops in the school and I am seeing this on one of them. I have confirmed that she is running the latest version of Firefox and have tried to clear out the preferences, but there don't appear to be any.
    Does anyone have anything useful to add to the conversation? Tyler just doesn't seem to be grasping the issue.

  • Process locking while running the current process running in Unix

    Hi Guys,
    I am sorry to ask you the question here . I have a situtation like I am running one process using crontab which kicks of every 30 mins but due to database slow my another batch is overlapping the process . Could you please give me some suggetions in unix
    Process A is running at 10:30 but this process did not complete
    The same process is started 11:00 -- Both process are overlapping each other .
    Please suggest me any thing .
    Clark

    No Oracle version number
    No description of the process or the SQL or PL/SQL being run
    No explain plans
    No traces
    No timings
    Here are your choices in no particular order.
    1. Buy faster hardware
    2. Load less data into your tables
    3. Modify all SQL statements to contain "AND ROWNUM = 1"
    4. Improve the performance of the first job
    5. Start the second job later than 11:00
    6. Provide those trying to help you with sufficient information for them to do so
    Well you did ask for "anything." <g>
    Seriously ... I'd suggest you choose #6.

  • Process chain Error: This AND process is not waiting for event RSPROCESS

    Hi All,
    I am facing an error in the process chain in PRD.
    Error message: This AND process is not waiting for event RSPROCESS.
    We had a process chain which had two sub chains which run parallel and below of this we had AND process type and below of the AND process we had 4 more jobs.
    Once the two subchains got successfully loaded, the and process should turn to green and further 4 jobs should start.
    It is a daily load , it worked fine from past years,but suddenly the AND processtype is getting failed.The thing is in the two subchains if one got completed, the AND process process is not waiting for 2nd subchain to get completed.The AND process is turning to RED (status: cancelled) . I tried to repeat the AND process once after above 2 subchains loaded, but it didnt worked.
    And i removed the existing AND process and created a new one and placed it in same place and activated and schedules again.but still it is getting failed with same error.
    Thanks in advance,
    Sai Chand.S

    Hi,
    If you did any transports related to that metachain we face similar kind of issues.
    not only the AND process , we need to remove all the process before executing the AND process and create it again.
    It helps you, you said your in production So you should take the proper approvals and do it.
    Regards,
    Yerrabelli.

  • My MacBookPro (mid 2012) will not complete the boot process. Yosemite's progress widget gets about 1/3 to 1/2 way across when the logo disappears and the screen goes all white for 10-15 secs or so, then the machine shuts down and restarts.

    My MacBookPro (mid 2012) will not complete the boot process. Yosemite's progress widget gets about 1/3 to 1/2 way across when the logo disappears and the screen goes all white for 10-15 secs, then the machine shuts down and restarts. This sequence will repeat seemingly forever unless I shut it down manually.
    I upgraded to Yosemite from Mountain Lion a couple months ago, skipping Maverick completely, and I've done any associated updates that Apple told me about. I don't recall which version of Yosemite I have, but I'm sure it is the "latest and greatest".
    I've tried re-starting in Safe Mode / Safe Boot and that does the same thing as described above except without the endless restarts. Instead of restarting at some point, SafeBoot proceeds to the white screen and then it hangs until I restart it manually. (I've waited as long os 20-30 minutes with no response).
    When I tried the option-command-p-r key combo I never get any chimes or other noises besides the original 1 time boot sound that you would expect to get when starting up a macbook pro.
    I was using the mac quite normally (meaning I was doing work I normally do on it and there was no indication of anything wrong) immediately prior to this happening.
    The machine is BootCamp(ed) with Windows 8.1 which I use from a VMWare virtual machine instead of natively so that I can have both OSes running at the same time. I also think this is unrelated to my problem. I have been using this setup for more than 2 years.
    My upgrade to Yosemite was uneventful and I thought it went well and was very smooth. I don't want to lead attention away from Yosemite, but at this point my problem does not seem related to the upgrade.
    Any help is appreciated.

    Install or Reinstall OS X from Scratch
    Be sure you backup your files to an external drive or second internal drive because the following procedure will remove everything from the hard drive.
    Boot to the Recovery HD:
    Restart the computer and after the chime press and hold down the COMMAND and R keys until the menu screen appears. Alternatively, restart the computer and after the chime press and hold down the OPTION key until the boot manager screen appears. Select the Recovery HD and click on the downward pointing arrow button.
    Erase the hard drive:
      1. Select Disk Utility from the main menu and click on the Continue button.
      2. After DU loads select your startup volume (usually Macintosh HD) from the
          left side list. Click on the Erase tab in the DU main window.
      3. Set the format type to Mac OS Extended (Journaled.) Optionally, click on
          the Security button and set the Zero Data option to one-pass. Click on
          the Erase button and wait until the process has completed.
      4. Quit DU and return to the main menu.
    Reinstall OS X: Select Reinstall OS X and click on the Install button.
    Note: You will need an active Internet connection. I suggest using Ethernet if possible
               because it is three times faster than wireless.

  • The process could not execute 'sp_repldone/sp_replcounters' error for Log Reader Agent and SQL Server Assertion 17066 & 3624 errors in SQL Logs

    One of our SQL Server started creating SQLDUMP file and and on investigation I found the error longs are filled with Errors 3624 & 17066. There is transnational replication configured on one of the databases is the LogReader Agent is failing error "The
    process could not execute 'sp_repldone/sp_replcounters' on XXXXX". 
    Not sure if both these Assertion & Logreader Agent errors are related. Before I remove and put the replication, I wanted to check if anyone has experienced the same issues or aware of what the cause. 
    ***********Error messages from SQL Logs******
    **Dump thread - spid = 0, EC = 0x0000000111534460
    Message
    A system assertion check has failed. Check the SQL Server error log for details. Typically, an assertion failure is caused by a software bug or data corruption. To check for database corruption, consider running DBCC CHECKDB. If you agreed to send dumps to
    Microsoft during setup, a mini dump will be sent to Microsoft. An update might be available from Microsoft in the latest Service Pack or in a QFE from Technical Support.
    Error: 3624, Severity: 20, State: 1.
    SQL Server Assertion: File: <logscan.cpp>, line=2123 Failed Assertion = 'UtilDbccIsInsideDbcc () || (m_ProxyLogMgr->GetPru ()->GetStartupState () < RecoveryUnit::Recovered)'. This error may be timing-related. If the error persists after rerunning
    the statement, use DBCC CHECKDB to check the database for structural integrity, or restart the server to ensure in-memory data structures are not corrupted.
    Error: 17066, Severity: 16, State: 1.
    External dump process return code 0x20000001.
    External dump process returned no errors.
    Thank you in advance.

    You need to determine if this error is a transient one or a show stopper one.
    It sounds like your log reader agent has crashed and can't continue.
    If so your best bet is to call Microsoft CSS and open a support incident.
    It also sounds like DBCC CHECKDB was running while the log reader agent crashed.
    If you need to get up and running again run sp_replrestart, but then you might find that replicated commands are not picked up. You will need to run a validation to determine if you need to reinitialize the entire publication or a single article.
    I have run into errors like this, but they tend to be transient, ie the log reader agent crashes, and on restart it works fine.
    looking for a book on SQL Server 2008 Administration?
    http://www.amazon.com/Microsoft-Server-2008-Management-Administration/dp/067233044X looking for a book on SQL Server 2008 Full-Text Search?
    http://www.amazon.com/Pro-Full-Text-Search-Server-2008/dp/1430215941

  • Process is not visible in BPM Workspace in Application panel 2

    Hi dear developers,
    I had a problem about BPM process. I was trying to implement this tutorial -> http://st-curriculum.oracle.com/obe/fmw/obpm/11g/r1/firstProcess/firstprocess_obpm11g.htm#t3
    And all the steps had been done. But the deployment step, i couldn't see my process on the BPM Workspace. I searched the other opinions in the forum. It's completely same with this thread -> Process is not visible in BPM Workspace in Application panel
    Nevertheless, couldn't find anything helpful. Could you help me please?
    Thanks in advance.
    Erdo

    Hi Erdo
    I hope you already ran the demo community seed app to create all the test users and groups etc in the default authenticator. If not, please see this below post:
    Re: New to BPM: How to get the tasks in the Workspace
    1. First login into bpm/workspace with domain adminstrator username/password. Then on right side, click on Adminstration link and see if you could see all the roles related to hello world app and also the users that you added to those roles.
    One other very basic thing is, you should and must deploy the TaskForms project that has all the taskdetails UI pages. I hope you did this while deploying the workflow and checking the check boxes to deploy other ui projects also.
    If seeding and getting demo uses is confusing, you can always use your own users. From weblogic console, create some extra users like weblogic1, weblogic2, weblogic3 etc with some password. Then use these users to assign to the roles from hello world app. Then login with these users. This will also work.
    Thanks
    Ravi Jegga

  • Workflow could not start in HCM process and Forms

    Hi All
    It would be really helpful if you can suggest possible solution regarding case management case in HCM forms and process.
    I have set up a simple custom HCM process and Form regarding Infotype 2 updation. I have checked the process and form consistency and it seems fine. Now when I run the process from HRASR_DT it generates a process number but it also gives an error workflow could not start.I get following error (SWIA log - Step history)
    Executing flow work item - Transaction brackets of the workflow has been damaged
    Exception occurred - Error when starting work item 000000007031
    PROCESS_NODE - Error when processing node '0000000014' (ParForEach index 000000)
    CREATE - Error when creating a component of type 'Step'
    CREATE_WIM_HANDLE - Error when creating a work item
    CREATE_VIA_WFM - Exit CL_HRASR00_POBJ_WF_EXIT triggered exeception for event CREATED and (target) status
    EVENT_RAISED - Error updating the process object
    Executing flow work item - Exit CL_HRASR00_POBJ_WF_EXIT triggered exeception for event STATE_CHG and (target) status READY->ERROR
    EVENT_RAISED - Error updating the process object
    Executing flow work item - Transaction brackets of the workflow has been damaged
    Executing flow work item - Work item 000000007031: Object FLOWITEM method EXECUTE cannot be executed
    Executing flow work item - Error when processing node '0000000014' (ParForEach index 000000)
    I have searched few SAP notes such as 1384961 which talks about same error but it is implemented in our system (EA-HR 604 , Patch 27, )
    Appreciate any help.
    Thanks
    Umang

    Hi Swati
    One of the similar (not exact) issues were resolved by correcting the event linkage....please refer thread:
    http://scn.sap.com/message/14848795#14848795
    Not sure of the details of the the resolution, maybe Ragavendran Kanagaraj can throw more light.
    Just some high level checks (if not done already):
    Check event linkage errors in SWEQADM if anything is related to your WF
    Check in SWEL if it's showing any more details (not sure if it will)
    Any warnings in syntax check of WF template (PFTC) or WF definition (SWDD)
    Any error sin binding
    regards,
    Modak

  • How to determine the Parent process information in Oracle BPMS 10g

    Hi ,
    I am trying to retrieve the parent process information given the instanceId and Process Id of an instance of a process from a different process.
    I am getting the instanceId of the instance in workspace using the Fuego.Papi.ProcessService and by creating the Filter.
    I tried using the ProcessInstance.parent but it needs to be instantiated using the instance details of my referred instance, which i am not able to do.
    Can some one shed some light on how to get the parent process details from the referred instance.
    Thanks,
    shravan.

    Hi,
    If you mean to increase the request timeout i.e the time when a request is submitted to the server and the response is sent by the server it can be done in httpd.conf file located in $OracleHome/Apache/Apache/conf directory.
    The attribute to be used for setting timeout is "Timeout". By default teh value is 600 seconds i.e 5 minutes.
    After making these changes , run command dcmctl updateConfig -ct OHS and restart the HTTP Server. Please note these commands need to be run only if the httpd.conf changes are done through text editor.
    Hope this helps.
    Smita

  • Process System Not Responding

    I have Yoga 8. I was simply playing this game from Google Play, Fun Run when the battery died due to low power and then when i charged the tab and started it, it has been giving me the error "Sorry: Process System Not Responding". The tab keeps restarting and it locked so i can't even log in to update the app anyone and the wifi isn't working either.
    Can anyone help? Thanks

    Hi mmoorthy, 
    Welcome to the Sony Community! 
    Please try to clear the cache of the Chrome. Here are the steps. 
    IMPORTANT: A Google Chrome browser window should already be open to complete this procedure.
    On the keypad remote control, press the button.
    In the Options menu, select Settings.
    Press the ENTER button.
    In the Browser Settings menu, under Privacy & Safety, select Clear saved browsing data.
    NOTE: The Clear saved browsing data option will clear the history, cache and cookies. If you do not want to do all 3, you can instead select each one individually. They are located just below the Clear saved browsing data option on the menu.
    After you have selected the desired option, press the ENTER button.
    Then repeat the steps provided by SonyFanGirl.
    If my post answers your question, please mark it as "Accept as Solution"

  • BPEL Processes are not showing in BPEL Console after OPMN restart

    Hi All,
    BPEL Processes are not showing in BPEL Console after OPMN restart. It was showing earlier before restart.
    Can some one help on this?
    Thanks

    Vikky,
    Are you using 10g?
    If BPELConsole is up but you don't see any process, you can go to the log files and see what is the problem.
    It might relates to permissions, or DB issues etc....
    Arik

  • Killing a sub process from it's parent process

    does any one know if ther's a way of killing a sub process from it's parent process?

    There are a few ways to do this depending on how the subprocess was invoked and if you are inside the parent work item instance or if you are using PAPI.
    If you are not using PAPI, here's how you could abort the child work item instances spawned by a parent process's work item instance.
    Asynchronously Spawned Children
    If a child process was invoked using an asynchronous Process Creation activity, the work item instance in the parent has a predefined hash table variable called "children". Downstream of the parent's Process Creation activity (called "SpawnChildren" in the PBL logic below), you can get the child's instance id using the children hash table and use this to send a notification to a Message Wait activty in the spawned child process instance using logic like this:
    send Notification
            using instanceId = children["SpawnChildren"],
                  activityName = "TerminateChild",
                  parameters = nullThe Message Wait activity needs to have its "Allows Interrupt" property checked in the child process. As you add this Message Wait activity (called "terminateChild" in the example above) make it an orphan with no transitions in or out initially. Once notified by the parent process, it still needs to be aborted so add a new Automatic activity in the child with this logic:
    action = ABORTFinally, add a transition from the Message Wait activity in the child subprocess to the Automatic activity.
    Synchronously Spawned Children
    If all you know is the parent work item instance's id and the child subprocess was invoked using a synchronous SubFlow activity, you could send a notification to a Message Wait activity that has it's "Allows Interrupt" property checked in the parent. If the Message Wait activity flows to an Automatic activity, you could use logic like this to kill the child subprocess work item instances synchrously spawned by the parent:
    for each ch in ProcessInstance.children do
        send Notification
            using instanceId = ch,
                  activityName = "TerminateChild",
                  parameters = null
    end
    action = SKIPSince the parent's instance was stuck in the Subflow activity, you'd need to send a notification to the parent's Message Wait activity in the parent process.

  • Firefox process won't die

    When I quit/close Firefox, wait 1 minute, then reopen Firefox, Firefox does not open...it just hangs.
    When looking in my Task Manager, there are two firefox.exe processes running. Killing the process via the Task Manager is required to get Firefox to open/run again.

    I'm quite familiar with the issue you've proposed via this KB article. My issue, though, is quite different. I don't get the annoying "Firefox is already running, but is not responding. To open a new window, you must first close the existing Firefox process, or restart your system.." alert box.
    Further, with the known issue, the firefox process does eventually die. With this new issue, the firefox process never dies and appears orphaned.
    Keep in mind this only started occurring after I upgraded to version 3.6.9 two days ago. There is no doubt in my mind something is broken in the newest build.
    As a developer, I like adhering an important and well-known principle in the software development world when I build and maintain stuff: if it ain't broke, don't fix it. Sometimes I wonder/question if Mozilla developers hold the same conviction, given numerous new issues that make it into every new build of Firefox.
    This also brings up another subject: does anyone do QA anymore? I like to test my stuff before letting others use it. It's extremely important to make sure your stuff is solid, before letting others use it. Cross your T's and dot your I's, for heaven's sake!

Maybe you are looking for

  • Problem with Photoshop Elements 11

    Hi, I recently deleted an old version of Photoshop Elements 7 as I'm now using V11. Since doing this when I try to open a photo in Photoshop the photo does not appear. Photoshop 11 opens but no photo. Any suggestions? I have made Photoshop Elements 1

  • How to connect WAP200 and WRV210

    Hello! I need to connect two dislocated wired ethernet cables (about 50 meters), so i bought one WAP200 (suport AP, bridge mode and repeater mode) and one WRV210. How do i need to configure the routers? I configer WAP200 as AP  and in WRV210 configua

  • Re-installation photoshop elements 11 on a new PC

    My old PC have Photoshop Element 11 installed in 2013 (I still have the Serial #) As this OLD PC cannot connect to the Internet, I am unable to disconnect Photoshop via Creative Cloud What should I do ?????

  • Blob at the start of a calligraphic line taper

    Hi all, I have a problem with the start of the calligraphic line taper in Illustrator CS2, using a Wacom Bamboo Fun. When I start to add pressure lightly, a blob appears at the start of my line much bigger than the minimum diameter. The rest of the l

  • Weather WebPart through customaization

    hi all, I want to create a custom weather web part through C# code if it'll be possible. Suggest me any code snippet to implement this through API. I want to give below look and feel in my Webpart. Please help me if any one has any better way to impl