Ext3fs file system access on Sparc SCSI disk

We've got a requirement to read data from a SCSI disk with an ext3fs file system which will be connected to a SunBlade system. Write capability would be a bonus.
We've found some old (and not working) implementations to allow Solaris to mount and read ext2, but these don't work on Sparc. Things like LTools won't work either, since Solaris won't allow access to a SCSI disk that doesn't have a Solaris label on it.
Does anyone know of any possible solutions to this problem? Any one know of ext3fs for Solaris? Any one know how to access an internal SCSI disk on Sparc that doesn't have a Solaris label on it?
Thanks
Ed

Hmmmmmm........Thanks for the suggestions - have written a test and the results arent good.
Using file filter to "find" the file is just too slow (1.4 secs and i need to do this 4million times for my application).
So I understand why the filter is slow compared to directly naming the file as we have to accept/reject them all, whereas if we know the file we can go straight to it. But this is a nightmare for my app. Looks like Im going to have to think again....
  public static void main(String[] args) throws Exception {
    String dir = "/path/files";
    File dirf = new File(dir);
    long l = System.currentTimeMillis();
    File f1 = new File(dirf,"TheRealFile.3253.ser"); // The file in the 30,000 we want
    InputStream i = new FileInputStream(f1);
    System.err.println("--> "+i.read());
    i.close();
    System.err.println("Took: "+(System.currentTimeMillis()-l)+" ms");
    // Test 2
    FileFilter ffx = new FileFilter() {
      public boolean accept(File s) {
        return s.isFile() &&
               s.getName().startsWith("TheRealFile")
               && s.getName().endsWith(".ser");
    l = System.currentTimeMillis();
    File [] f2 = dirf.listFiles(ffx);
    //System.err.println("Found: "+f2[0]);
    InputStream i2 = new FileInputStream(f2[0]);
    System.err.println("--> "+i2.read());
    i2.close();
    System.err.println("Took: "+(System.currentTimeMillis()-l)+" ms");
--> 172
Took: 0 ms
--> 172
Took: 1482 ms

Similar Messages

  • Direct File System access problems from JAWS application

    Hello,
    I have built a Web Start application that consists of a Webserver (Jetty) ,
    a WAR file and a Java (main) class that deploys the web application on
    the server and starts the server. It all works fine, apart from the fact
    that I am getting java.security.AccessControlExceptions when I try to
    access the local filesystem or system variables like the java.io.tmpdir.
    I have signed all the jar files and I included the<security> <allpermissions />
    </security> tag in the jnlp file. Still, I can't seem to get out
    of the sandbox.
    I have read in this article (http://mindprod.com/jgloss/javawebstart.html)
    that direct file system access from a Web Start application is impossible
    (Quote: "There is still no way for even a signed JAWS app to
    find some persistent disk space in an easy way. It pretty well
    has to ask the user for the name of some directory to use.")
    Is this true?
    Thank you,
    Peter

    Hi Guys,
    I found a way to access the local filesystem...
    Besides signing all the jar files and including the<security><allpermissions /></security> tag in the jnlp file I have to include this line in the code I execute on the client machine:
    System.setSecurityManager(null);
    Regards,
    Peter

  • SAP File System Access - SLD Naming Convention/Suggestions

    I would like to access our ECC file system to pick up files we will use to create Idocs.   
    I'm wondering the best way to describe the file system access in the SLD.   
    I have a business system for the main client on the ECC  system (BS_ED1CLNT010) for example but the OS isn't client specific.    I could use this as the Business System in the scenario and define a file adapter that connects to the unix server.
    Any thoughts?

    Maybe I didn't frame my question properly.  
    In the ECC system we have multiple clients (20,30,40, etc).  If I am going to post an Idoc to a client in this system I need to define each as a business system in the SLD and import this to the Integration Directory.    So I would have BS_XXXCLNT010, BS_XXXCLNT020, etc, one for each client.   These all share the same Technical System.    If I want to post an Idoc to a client on the ECC system I have to define a Business System and interface to that and every client that will receive an Idoc. (as well as the ALE settings on the ECC)
    Each of these reside on the same SAP server (sap00001 for example) and there is a directory (/public for example)  on this server.     This isn't client specific.  
    I wish to pick up a file from the ECC file system and post a client on the ECC system (maybe different ones based on the data in the file).   
    I don't want to define the file adapter under BS_XXXCLNT020 since it isn't specifc to client 020 although that would work.  
    Do I create a new TS in the SLD as third party, stand alone java, and a BS for that?    TS_XXX_FILESYSTE (3rd party).  BS_XXX_FILESYSTEM for the TS?
    I'm really looking for clarity in the definition of the SLD.

  • Whether will falsh player support  Unrestricted File System Access in Full Trust ?if it will , when?

        I am doing a software based on Flash , to provide my customer to edit pictures online.
        In some scenario, to get Full Acess Local File is necessary. I find SliverLight 5.0 had supported such features:
    Unrestricted File System Access in Full Trust
    Full Trust in-browser for enterprise scenarios
    Default Filename in SaveFileDialog and OpenFileDialog
       I really desire those features , so , I am wondering whether adobe will support it .If will, when?

    Please read the below flash player administrator guide
    http://wwwimages.adobe.com/www.adobe.com/content/dam/Adobe/en/devnet/flashplayer/pdfs/flas h_player_11_7_admin_guide.pdf
    check the section "The User FlashPlayerTrust directory". It will help.

  • What is the minimum file system access needed to run ODI 10.1.3.4.0 client?

    Hi ODI discussion folks,
    I have a couple of questions from an Oracle partner that I'm trying to find a definitive answer for if possible. The partner is setting up ODI 10.1.3.4.0 for a customer who insists that the absolute minimum amount of access to the file system is granted due to corporate security policies.
    I have checked the bundled ODI documentation but couldn't really find anything about file system permissions needed to run the ODI client. I was pointed towards the "Setting Up Security for an Integration Project — What to Consider" document but this does not mention a great deal about how much access to the file system is needed for the ODI client to function.
    What the partner is asking is the following:
    "1. What are the minimum file/folder permissions needed for the ODI client installation? I'm installing at xxxxx
    and their machines have to be locked down as much as possible.
    2. Say you have 3 users all wanting to run integrations etc and the Master and Work
    repositories have been set up. An admin installs the ODI client but doesn't
    create the connection to the Master repository. What are the minimum
    file/folder permissions required on the client machine to:
    a) create the connection to the repository
    b) run any subsequent integrations?"
    If anyone can advise on this then that would be much appreciated.
    Regards
    Craig Huggans
    Oracle Hyperion Support
    Message was edited by:
    user648991

    Hi Craig,
    How are you?
    Let me try to contribute a little....
    1) The minimum requirement is for its own installation directory, there is no reason to have access to other directories unless if it is necessary to read files from some other directory at the client
    2) Again only to its own install directory. The connection setting is recorded at \bin install directory. After that, all information are recorded at repository, there is no client work.
    Be free to contact me by email or phone if you have any new doubt. You can get my email from my profile.
    Does it respond your doubts?
    Cezar Santos

  • Confusion with OCFS2 File system for OCR and Voting disk RHEL 5, Oracle11g,

    Dear all,
    I am in the process of installing Oracle 11g 3 Node RAC database
    The environment on which i have to do this implementation is as follows:
    Oracle 11g.
    Red Hat Linux 5 x86
    Oracle Clusterware
    ASM
    EMC Storage
    250 Gb of Storage drive.
    SAN
    As of now i am in the process of installing Oracle Clusterware on the 3 nodes.
    I have performed these tasks for the cluster installs.
    1. Configure Kernel Parameters
    2. Configure User Limits
    3. Modify the /etc/pam.d/login file
    4. Configure Operating System Users and Groups for Oracle Clusterware
    5. Configure Oracle Clusterware Owner Environment
    6. Install CVUQDISK rpm package
    7. Configure the Hosts file
    8. Verify the Network Setup
    9. Configure the SSH on all Cluster Nodes (User Equivalence)
    9. Enable the SSH on all Cluster Nodes (User Equivalence)
    10. Install Oracle Cluster File System (OCFS2)
    11.Verify the Installation of Oracle Cluster File System (OCFS2)
    12. Configure the OCFS2 (/etc/ocfs2/cluster.conf)
    13. Configure the O2CB Cluster Stack for OCFS2
    BUT, here after i am a little bit confused on how to proceed further. The next step is to Format the disk and mount the OCFS2, Create Software Directories... so and so forth.
    I asked my system admin to provide me two partitions so that i could format them with OCFS2 file system.
    He wrote back to me saying.
    *"Is what you want before I do it??*
    */dev/emcpowera1 is 3GB and formatted OCFS2.*
    */dev/emcpowera2 is 3GB and formatted OCFS2.*
    *Are those big enough for you? If not, I can re-size and re-format them*
    *before I mount them on the servers.*
    *the SAN is shared storage. /dev/emcpowera is one of three LUNs on*
    *the shared storage, and it's 214GB. Right now there are only two*
    *partitions on it- the ones I listed below. I can repartition the LUN any*
    *way you want it.*
    *Where do you want these mounted at:*
    */dev/emcpowera1*
    */dev/emcpowera2*
    *I was thinking if this mounting techique would work like so:*
    *emcpowera1: /u01/shared_config/OCR_config*
    *emcpowera2: /u01/shared_config/voting_disk*
    *Let me know how you'd like them mounted."*
    Please recommend me what i should convey to him so that i can ask him to the exact same thing.
    My second question is, as we are using ASM, for which i am gonna configure ASM after clusterware installation, should i install Openfiler??
    Pls refer the enviroment information i provided above and make recommendations.
    As of now i am using Jeffery Hunters guide to install the entire setup. You think the entire install guide goes well with my enviroment??
    http://www.oracle.com/technology/pub/articles/hunter_rac11gr1_iscsi.html?rssid=rss_otn_articles
    Kind regards
    MK

    Thanks for ur reply Mufalani,
    You have managed to solve half part of my query. But still i am stuck with what kind of mount point i should ask the system admin to create for OCR and Voting disk. Should i go with the mount point he is mentioning??
    Let me put forth few more questions here.
    1. Is 280 MB ok for OCR and voting disks respectively??
    2. Should i ask the system admin to create 4 voting disk mount points and two for ocr??
    3. As mentioned by the system admin.
    */u01/shared_config/OCR_config*
    */u01/shared_config/voting_disk*
    Is this ok for creating the ocr and voting disks?
    4. Can i use OCFS2 file system for formating the disk instead of using them as RAW device!!?
    5. As u mentioned that Openfiler is not needed for Configuring ASM... Could you provide me the links which will guide me to create partition disks, voting disks and ocr disks!! I could not locate them on the doc or else were. I did find a couple of them, but was unable to identify a suitable one for my envoirement.
    Regards
    MK

  • Disallow for file system access

    Hi All,
    I would like to not permit to access to file system, how can i do that with permission object. I saw only examples how to set contraints on location where it can access.
    Regards

    I'm sorry but i dont understand the problem. (Maybe someone else will)
    So you have told me that when you pass null to an file permission object, that you find out the mask is "NONE"?
    You want to stop users accessing some files?
    You need to build code that will stop the user accessing certain files
    You want to know if you can put restrictions on the whole VM(What is VM?) or just the context?
    I believe this Link could be helpful for you.

  • Motorola file system access

    hi,
    I wrote a Midlet to access image files on my motorola v3x.I also checked it by Device Emulator .It is working fine.But when i installed in my phone it is not worked.So,if any body worked on this please help me.
    sorry for my bad english.thanks in advance

    We have to start it from begining ..
    Does the device supports full implementation of JSR75. MOTO devices dont support complete implementations. like RAZRv3 i claims to support JSR75 but doesnt supports complete implementation.
    A. Check Call to System.getProperty with key microedition.io.file.FileConnection.version
    should return the implementation version number starting with 1.0
    B.Files read/write permissions should be supported by the device�s native system:
    � javax.microedition.io.Connector.file.read - should enable reading from the file
    system (hereinafter just �read permission�).
    � javax.microedition.io.Connector.file.write - should enable writing to the file
    system (hereinafter just �write permission�).
    If A and B are true ..u can start with below points.
    If it supports the complete implementation, then
    1. is it able to install properly.
    2. Have u specified permissions in the JAD file for the same?
    3. at what stage is it not working, when the file write code is being executed ??
    Message was edited by:
    aeran1

  • Files system - access problem (JSR 75)

    Hi All
    My cell phone is Nokia 6288, I'm from Israel.
    The phone should support JSR 75 (FileConnection and PIM API), and I get it without it.
    Probably, my Cell operator decided to remove the built-in support for JSR-75 (which suppose to be there).
    I wonder how can I add or restore the java thing for JSR-75 (if is it possible).
    or if there is someway to get it back because
    I know that Nokia 6288 should include the JSR 75 support.
    How can I be sure that it have be removed?
    Because many application which need this JSR 75 support
    couldn't find files and not work because of that.
    Thanks,
    Adi

    25-Jul-2008 10:33 AM
    burekas wrote:
    I know there is maybe a special burning which should get it back.
    It is not permitted to discuss such operations on this forum.
    Was this post helpful? If so, please click on the white "Kudos!" star below. Thank you!

  • Search event logs for file system access

    I'm looking to create a script that will allow me to search Windows 2012 security event logs for access to specific folders.  Ideally it would allow the granularity to search for read access events (4663) and specify specific users to view.  One
    example would be to show events for drive F:\ where the folder name is JSmith (including subfolders) and the username is not JSmith.
    I've tried something like this, but can't see how to filter.
    Get-EventLog security | ? {$_.Message.contains("F:\JSmith")}

    Is the match explicit?  How can I use wildcard?  How can I exclude events?
    I recommend asking a search engine and doing some initial research. Here's a starter:
    https://technet.microsoft.com/en-us/library/hh849682.aspx
    http://blogs.msdn.com/b/powershell/archive/2009/06/11/windows-event-log-in-powershell-part-ii.aspx
    http://blogs.technet.com/b/ashleymcglone/archive/2013/08/28/powershell-get-winevent-xml-madness-getting-details-from-event-logs.aspx
    http://blogs.technet.com/b/heyscriptingguy/archive/2011/01/24/use-powershell-cmdlet-to-filter-event-log-for-easy-parsing.aspx
    https://richardspowershellblog.wordpress.com/2009/03/08/get-winevent/
    Don't retire TechNet! -
    (Don't give up yet - 13,225+ strong and growing)

  • Extending RAID file system with larger disk

    Hi,
    I have a Mac pro running 10.7.5.   
    I used to have two 1TB disks in the internal bays configured as a raid 1 set with a single 1TB file system.  One of the disks failed and I replaced it with a 2TB disk which seems to be working ok.   However, I'm running out of space in the file system and thought to switch to 2TB.
    I just got another 2TB disk and thought to replace the 1TB disk with this one. Then I would have two 2TB disks but I guess after the rebuild the file system is still 1TB.  Is there any way to extend the file system and underlying raid set on the go or do I need to disband the raid set to two independent disks first, then repartition one of them, copy data from the remaining half to this now 2TB file system and add the source disk to the raid set afterwards?
    Hannu

    I too have a 1,1 and continue to "do things" and upgrade it.
    Sonnet Tempo Pro SSD PCIe is perfect for Aperture having it on another bus controller with SATA III 6G and 550MB/sec.
    $400 was never too much to spend for SCSI or for my first couple 160MB disk drive for perspective.
    Mavericks (cannot run normally on ours) was suppose to offer more advantages in RAID features but even that seems to be MIA.
    PS: A Sonnet Allegro USB3 PCIe $54 is a nice way to interface instead of using USB2.
    I have used mirror but AFTER having a couple good backups and strategy.
    So just buy a new set of 2TB or larger drives.
    We have had people with iTunes library alone that needed their own 2TB volume when the largest drives were 1TB.
    Done right a stripe RAID is fine, and one way I always used to have larger volumes.
    "Extended volume support" is more a Windows thing, not a Mac feature.
    As you found, Apple mirror RAIDs cannot 'grow' and you can't add a 2TB and suddenly have larger mirror w/o delete and create a new set.
    You may be lucky that you never had a disaster. A 3-way mirror is really preferable, take one of the drives out while rebuild and you still have two in a mirror set. AND clone your volume or use something besides or in addition to TimeMachine.
    A backup is only as good as your ability to restore and have a recovery plan - hopefully one that you have practiced doing but never need to.

  • How to access file system in  kernel?

    hi
    i need to access disk file system in kernel.
    is it possible using UFS data structure and function in
    other modules in kernel?
    or i need to create file system access function based on DDI/DKI?

    Short answer is "you cannot".
    The longer one is "see topics in this forum" - there was an exhaustive discussion about this issue...
    The thead "File I/O from withing a driver" in Driver Development forum may be of particular interest.
    Hope this helps,
         Ilya.

  • Yosemite will not install and won't let me change startup disk, it's saying there is a file system problem but disk utilities can't fix it.

    I Downloaded the Yosemite upgrade and went to install it on my iMac. The install failed saying file System needed repair. Ran disk utility could do the verify but the repair button was grayed out. When I tried to change startup disk back it wouldn't let me , saying it couldn't gather enough file info. So I'm stuck and can't boot my Mac. How do I get out of this?

    Do a backup. Boot to the Recovery Volume (command - R on a restart or hold down the option/alt key during a restart and select Recovery Volume). Run Disk Utility Verify/Repair and Repair Permissions until you get no errors. Try to reboot normally.
    OS X Recovery
    OS X Recovery (2)

  • Difference between ASM Disk Group, ADVM Volume and ACFS File system

    Q1. What is the difference between an ASM Disk Group and an ADVM Volume ?
    To my mind, an ASM Disk Group is effectively a logical volume for Database files ( including FRA files ).
    11gR2 seems to have introduced the concepts of ADVM volumes and ACFS File Systems.
    An 11gR2 ASM Disk Group can contain :
    ASM Disks
    ADVM volumes
    ACFS file systems
    Q2. ADVM volumes appear to be dynamic volumes.
    However is this therefore not effectively layering a logical volume ( the ADVM volume ) beneath an ASM Disk Group ( conceptually a logical volume as well ) ?
    Worse still if you have left ASM Disk Group Redundancy to the hardware RAID / SAN level ( as Oracle recommend ), you could effectively have 3 layers of logical disk ? ( ASM on top of ADVM on top of RAID/SAN ) ?
    Q3. if it is 2 layers of logical disk ( i.e. ASM on top of ADVM ), what makes this better than 2 layers using a 3rd party volume manager ( eg ASM on top of 3rd party LVM ) - something Oracle encourages against ?
    Q4. ACFS File systems, seem to be clustered file systems for non database files including ORACLE_HOMEs, application exe's etc ( but NOT GRID_HOME, OS root, OCR's or Voting disks )
    Can you create / modify ACFS file systems using ASM.
    The oracle toplogy diagram for ASM in the 11gR2 ASM Admin guide, shows ACFS as part of ASM. I am not sure from this if ACFS is part of ASM or ASM sits on top of ACFS ?
    Q5. Connected to Q4. there seems to be a number of different ways, ACFS file systems can be created ? Which of the below are valid methods ?
    through ASM ?
    through native OS file system creation ?
    through OEM ?
    through acfsutil ?
    my head is exploding
    Any help and clarification greatly appreciated
    Jim

    Q1 - ADVM volume is a type of special file created in the ASM DG.  Once created, it creates a block device on the OS itself that can be used just like any other block device.  http://docs.oracle.com/cd/E16655_01/server.121/e17612/asmfilesystem.htm#OSTMG30000
    Q2 - the asm disk group is a disk group, not really a logical volume.  It combines attributes of both when used for database purposes, as the database and certain other applications know how to talk "ASM" protocol.  However, you won't find any general purpose applications that can do so.  In addition, some customers prefer to deal directly with file systems and volume devices, which ADVM is made to do.  In your way of thinking, you could have 3 layers of logical disk, but each of them provides different attributes and characteristics.  This is not a bad thing though, as each has a slightly different focus - os file system\device, database specific, and storage centric.
    Q3 - ADVM is specifically developed to extend the characteristics of ASM for use by general OS applications.  It understands the database performance characteristics and is tuned to work well in that situation.  Because it is developed in house, it takes advantage of the ASM design model.  Additionally, rather than having to contact multiple vendors for support, your support is limited to calling Oracle, a one-stop shop.
    Q4 - You can create and modify ACFS file systems using command line tools and ASMCA.  Creating and modifying logical volumes happens through SQL(ASM), asmcmd, and ASMCA.  EM can also be used for both items.  ACFS sits on top of ADVM, which is a file in an ASM disk group.  ACFS is aware of the characteristics of ASM\ADVM volumes, and tunes it's IO to make best use of those characteristics. 
    Q5 - several ways:
    1) Connect to ASM with SQL, use 'alter diskgroup add volume' as Mihael points out.  This creates an ADVM volume.  Then, format the volume using 'mkfs' (*nix) or acfsformat (windows).
    2) Use ASMCA - A gui to create a volume and format a file system.  Probably the easiest if your head is exploding.
    3) Use 'asmcmd' to create a volume, and 'mkfs' to format the ACFS file system.
    Here is information on ASMCA, with examples:
    http://docs.oracle.com/cd/E16655_01/server.121/e17612/asmca_acfs.htm#OSTMG94348
    Information on command line tools, with examples:
    Basic Steps to Manage Oracle ACFS Systems

  • Problem with the EFI system partition's file system.

    I installed latest yosemite 10.10.2 update - did a verify disk on “WDC WD1001FALS-403AA0 Media” (internal HD on my iMac 21.5-inch, Mid 2011) and got
    Problems were found with the partition map which might prevent booting
    Error: The partition map needs to be repaired because there’s a problem with the EFI system partition’s file system.
    However a verify disk on the volume partition gets a clean bill of health!
    I haven't done a 'repair disk' yet. Thought I'd check if anybody else has same message first.
    Cheers

    I had issues where my MacPro 8-core would not boot after the 10.7.4 upgrade, but my MacBook Air 13" 2011 upgrade worked just fine.
    I had to restore my MacPro HD from my TimeMachine back up. Apparently, the Softraid raid drivers that I use on my MacPro are not compatible yet with 10.7.4. See here for more info:
    http://macperformanceguide.com/blog/2012/20120511_1-MacOSX-SoftRAID.html
    I contacted Softraid and they are already aware of the issues and are working with Apple for a fix.
    Cheers.
    Bud

Maybe you are looking for