Archival process eating up disk space

Hi,
I am running on Oracle Database version 9i and the db is running on HPUX 64 bit RISC platform.
I need to know whether killing the archival process will cause any isue to the Production environment? Or will it respawn another arc process and continue to work normally?
$ ps -ef | grep dynocop5 | grep arc
dynodba 4409 1 0 Jan 23 ? 159:54 ora_arc2_dynocop5
dynodba 4388 1 0 Jan 23 ? 197:50 ora_arc0_dynocop5
dynodba 4396 1 0 Jan 23 ? 75:18 ora_arc1_dynocop5
Actually, the mount point shows 85GB filled when only 35GB is being used! As per our Unix system, these type of issues happen only when someone remove/move a file when it is still accesses by some process. And currently these 3 processes are attached to this filesystem.
$ du -sk .
35234737 .
$ bdf .
Filesystem kbytes used avail %used Mounted on
/dev/vx/dsk/dynodg03/dyno5_arch_vol
206413824 86682808 112248054 44% /dyno5/arch
ebzdbp03: /usr/local/CPR/bin # ps -ef|grep 4388
dynodba 4388 1 0 Jan 23 ? 197:50 ora_arc0_dynocop5
root 15222 10499 1 06:29:22 pts/ta 0:00 grep 4388
ebzdbp03: /usr/local/CPR/bin # ps -ef|grep 4396
dynodba 4396 1 0 Jan 23 ? 75:18 ora_arc1_dynocop5
root 15445 10499 0 06:29:34 pts/ta 0:00 grep 4396
ebzdbp03: /usr/local/CPR/bin # ps -ef|grep 4409
dynodba 4409 1 0 Jan 23 ? 159:07 ora_arc2_dynocop5
root 15649 10499 0 06:29:48 pts/ta 0:00 grep 4409
ebzdbp03: /usr/local/CPR/bin #

Hi,
If you kill by brute force an arc process, your instance will probably die. I would open a SR case, of wait for a maintenance window and then restart your database.
Regards,
Mario Alcaide
http://marioalcaide.wordpress.com

Similar Messages

  • RMAN9I HOW TO RESTORE ARCHIVE LOGS WITH LIMITED DISK SPACE

    제품 : RMAN
    작성날짜 : 2002-12-09
    RMAN9I HOW TO RESTORE ARCHIVE LOGS WITH LIMITED DISK SPACE
    ==========================================================
    PURPOSE
    이 자료는 Oracle 9.2 이상의 RMAN에서 사용 가능한 MAXSIZE 기능에 대하여
    설명할 것이다.
    How to restore archive logs with limited disk space
    Old Backup을 사용하여 RMAN으로 database를 recvoery할때, 모든 archived
    redo logfile들을 Restore할 destination의 disk space가 모자란 경우를 종종 만난다.
    이러한 경우 여러 개의 restore and recovery job으로 나누어 수행한다.
    즉, 첫 번째 restore and recovery job이 끝나면 archived redo logfile들을 지우고
    다음 적용될 것들을 restore한다. 그리고 다시 Recovery를 수행한다. 이러한 작업을
    원하는 시점까지 recovery를 하기위해 반복적으로 수행한다.
    Oracle 9iR2(9.2.0.x)부터 RMAN은 MAXSIZE option을 제공하는데 이것은 archive log
    file이 restore될 disk의 space를 control할 수 있게 해 준다.
    만약 disk space가 매우 제한적이라면, 즉 예를 들어 모든 archive log의 size 합보다
    disk free space가 적다면 MAXSIZE option을 사용하는 것은 매우 유용하다.
    이 OPTION이 기술되면 RMAN은 Media Manager에게 disk space 크기에 부합하는 만큼만
    archive log들을 restore하도록 한다. 부가적인 restore operation들은 restore된
    마지막 archive log가 적용될 때마다 발생한다.
    MAXSIZE option은 이러한 작업들을 하나의 rman job으로 처리하게 해 주고
    실수를 하지 않게 해 준다.
    아래는 MAXSIZE를 설명하기 위한 예제이며 총 6 단계로 구성되어 있다.
    STEP 1: Add data to the database to enforce log switches
    STEP 2: Backup the database and archive logs and delete the logs
    STEP 3: Add additional data to enforce new log switches
    STEP 4: Remove the data file and simulate a database crash
    STEP 5: Restore the data file from the backup
    STEP 6: Recover the database using MAXSIZE
    위 예제는 Unix와 Windows 모두에 적용될 수 있다.
    STEP 6 는 다음과 같은 조건으로 두 번 수행될 것이다.
    1) MAXSIZE가 archive log size보다 작은 경우:
    이 경우 RMAN-6558 Error message를 만나게 된다.
    그러므로 MAXSIZE를 archive log보다는 크게 설정해야한다.
    2) MAXSIZE가 archive log size보다 큰 경우:
    예를 들어 여러 개의 archive log를 포함할 수 있을 정도로 크게 설정하면
    restore/recovery는 user에게 transparent하게 수행되어진다.
    즉 archive log들은 restore된 후 applied되고 deleted되어 진다. 다시 새로운
    archive log들이 restore되어지고 applied된 후 deleted 되어 진다. 이러한 작업은
    recovery가 끝날 때까지 반복되어진다. 이러한 작업들이 진행되는 동안 RMAN은
    아무런 Message도 발생시키지 않는다.
    # Step 1: INSERT enough new data to generate log switches
    create table rman_tst (col1 varchar2 (10));
    begin
    for i in 1..30000 loop
    insert into rman_tst values(i,'test');
    commit;
    end loop;
    end;
    # Step 2: BACKUP the database and the archive logs automatically
    # and then delete the input
    run {
    backup database format='/web01/usupport/krosenme/admin/backups/db_%d%s%t'
    plus archivelog format='/web01/usupport/krosenme/admin/backups/arch_%d%s%t'
    delete input;
    # Step 3: INSERT enough new data to generate new log switches
    begin
    for i in 1..30000 loop
    insert into rman_tst values(i,'test');
    commit;
    end loop;
    end;
    # Step 4: REMOVE users01.dbf file and crash the database
    mv users01.dbf users01.org
    shutdown abort
    # Restore is now needed as the data file is deleted. The backup was
    # taken before the new data was added to it, thus archive logs are
    # needed to bring the database up to date
    # Step 5: RESTORE the data file from the full backup
    run {
    restore datafile '/web01/usupport/krosenme/oradata/kro_920/users01.dbf';
    # Step 6: RECOVER
    run {
    recover database delete archivelog maxsize 10 K;
    # This will fail with RMAN-6558 as the archived log has a size of 16 KB,
    # which is bigger than MAXSIZE limit of 10 KB. So the error is expected
    # and MAXSIZE works as designed.
    # Now rerunning STEP 6: but with MAXSIZE 50 K
    run {
    recover database delete archivelog maxsize 50 K;
    RELATED DOCUMENTS
    Recovery Manager Reference, Release 2 (9.2)

    cold backup means offline backup, you shut database normally and copy datafiles to another location.
    i think operating system does not matter that much but it is RedHat Linux Enterprise Server.
    backing up archivelogs whith rman script which is:
    run {
    allocate channel c1 type disk format '$BKUPLOC/arch_%d_%u_%s_%p.bkp';
    change archivelog all validate;
    sql 'ALTER SYSTEM ARCHIVE LOG CURRENT';
    backup archivelog time between 'SYSDATE - (30*60/(60*60*24))' and 'sysdate';
    release channel c1;
    so here is the details.
    1. because everything is lost, we restored all files(datafiles, controlfiles, redologs etc) from offline backup meaning copied all files from another location to corresponding location.
    2. want to apply archive logs which were created after cold backup till disk failure and we have these archive logs with backups which were taken by rman.
    So
    first of all, we want to register these archive log rman backups(because these backup information does not exist in restored controlfile)
    and then restore them and then apply them.
    we need rman commands to register these backups and restore them.
    Regards,
    Kamil

  • Upgrade, or Archive and Install? Disk space?

    I'm thinking of upgrading to Tiger soon because I want those shadows I make in Pages to show up on PDFs. But before I do give up my money for the disc, I've got a few questions, just to make sure every thing goes as planned:
    - Upgrade or Archive and Install? I seem to remember back in the old days when there was mayhem with Upgrade when that option was first introduced (in Jaguar?). Is Upgrade reliable now?
    - If I do Upgrade, how much additional disk space do I need (I have Panther on my PowerBook right now)? I know Apple says 3GB on its website, but that's only for a clean install, right?
    - If I do Archive and Install, how much disk space will I need (including the Previous System folder)?
    PowerBook G4 (Gigabit Ethernet, 667MHz)   Mac OS X (10.3.9)  

    Hi Kim!
    Here is some additional information.
    Tiger is available for purchase at The Apple Store (U.S.).
    If you know what to look for, a Full Retail Version, of the Tiger Install DVD, can also be purchased rather inexpensively, at some online Apple retailers, Amazon, eBay, AppleRescue, FastMac, etc.
    As Gulliver posted, be sure not to purchase grey, upgrade or machine specific CDs or DVDs.
    The discs should look like the images in the above links.
    Additional info in these links.
    Using OS X Install CDs/DVDs On Multiple Macs
    What's A Computer Specific Mac OS X Release
    Software Update, Upgrade: What's The Difference?
    Tiger System Requirements
    Additionally, "Tiger ships on a DVD, but if your Mac doesn’t have a built-in DVD-ROM player, you’ll need CD media. When you buy Mac OS X Tiger, you qualify to purchase Tiger CDs for only $9.95."
    Download the Media Exchange Program Order From Here.
    Orders must be mailed by March 19, 2007.
    AppleRescue, also sells a Tiger installation set on CD
    Shop Carefully, Examine All Documentation, And Good Luck!
    ali b

  • Three invisible volumes eating up disk space

    I used Carbon Copy Cloner to load up a new external drive with the contents of my MBP. After it finished my Users folder it began copying "Volumes / LaCie 500GB."
    It wouldn't show up in Finder, so I presumed it was an invisible file, as best I understand it. I ran GrandPerspective to get a multicolor image depicting the relative sizes of all files hierarchically. It displayed clearly not just that LaCie, but two others as well. The three volumes use up some 85GB — a third of my drive.
    I've confirmed that the contents are all duplicates (and triplicates) of various parts of my main drive. It seems evident that I can just delete them, but I'd like to know of any cautions anyone may know.
    And what might I have done to copy the same external onto my MBP — three times? OR what other process might make that happen? I have CCC scheduled to back up my MBP onto the LaCie, but not back the other way.
    Thanks.

    WZZZ wrote:
    Why this would have happened three times is certainly something to investigate. Is the LaCie mounting properly otherwise?
    Now that I've read the page through more completely, I find that the last paragraph on that page evidently wraps it all up:
    "...you must ensure that the target will remain available throughout the cloning so that you don't have to do this clearing up too often!"
    This brings to mind that occasionally one of my FW cables to this drive gets bumped and disconnects. Not good, I know. But I'd bet this is what's caused it. I have the page bookmarked and now I'll know where to return.

  • Lost disk space after each reinstall.

    I've had to reinstall leopard on my macbook three times in the last week for different reasons. After each reinstall(erase and install) it appeared that I lost a little more GB to the HDD. I have a 250GB HDD, and I know it actualls reads as 232.87, but without uploading any media and only using the reinstall disks that came with the box, I have only 218 or so GB. Is this normal? I can't find the numbers anywhere; they just don't add up.

    Anthony Turtzo wrote:
    I can't find the numbers anywhere; they just don't add up.
    My guess is that you have added up the sizes of the folders visible in Finder at the root level of the drive (Applications, Library, System, & Users) & compared that total to the total space used on the drive. The two numbers will not match because there are also hidden files & folders at the root level of the drive that Finder does not normally show. The hidden items are necessary for the proper operation of the Mac & in normal use are managed by the system. Since users should not tamper with them directly without good reason & it requires expert knowledge to do so without breaking anything, they have been made invisible in the Finder view.
    It is normal for the drive space used by these items to fluctuate, & for the total to grow somewhat after first installing the OS. Some of the items contain temporary files that are created & destroyed as needed. Some contain usage, crash, & other informational logs that are managed by routines that the system automatically runs periodically to keep their total from becoming too large. Others contain initialization data gathered over time that the system uses to start up or to run processes efficiently.
    Thus, there is no fixed correct number for their total size since that depends on the history & current state of the machine. On my iMac G5, one large hidden folder ("var") currently uses about 1.05 GB of drive space, but consider that only a ballpark number since it may be considerably different on different Macs at different times.
    If you are still concerned about the health of the drive, especially that something is eating up disk space unnecessarily, then run Disk Utility to check it. In the "First Aid" tab select your drive & click the "Verify Disk" button. If this reports no problems then you can be pretty sure there is nothing to worry about.

  • Archivelog mode ans disk space

    Hi,
    when we are in archive log mode, how disk space would be occupied ? Where the archive files would be put ? How the disk space can be controlled and managed ?
    many thanks before.

    1. how disk space would be occupied ?
    It depends, mostly of your redo log size and amount of transactions you are going to have.
    Where the archive files would be put
    I assume you are not going to transfer these archive logs to a standby database.
    If you don't use flash recovery area for this purpose then you have to specify location in log_archive_dest_n parameters.
    If you use flash recovery area, then by default archive logs will be saved at destination specified in db_recovery_file_dest parameter.
    How the disk space can be controlled and managed ?
    Can you be more specific on this one?

  • Disk Space Disappearing -- 10.2.8

    I have a G3 PowerBook running OS X 10.2.8. I've noticed that my disk space disappears for some unexplained reason, although it does seem to happen most often after my daughter has been playing games on disney.com. Does anyone know how to pinpoint where the disk space is going, and more importantly, how to get it back? I've got an external hard disk and have moved lost of programs off of my Mac, but the gains are short-lived.
    Thanks for any help/suggestions.

    If you go to "more options" in the Search pane (upper right of this page) and enter "disney" and then search "all categories," you'll find other reports of "concerns" with visiting the Disney web site. Some are obviously just plain paranoia but such a search might be useful.
    I've been rehabbing three old G3 PowerBooks for a friend who knew nothing about the Mac OSX maintenance scripts. On one machine, simply running one of the scripts freed up 2GB of space on his 8GB boot partition (the Wallstreet G3 PB has the "Old World ROM" OSX boot partition limitation).
    To learn about this useful function, please refer to this article:
    Running the Mac OS X maintenance scripts
    The scripts run automatically but only if your computer is turned on and not sleeping in the wee hours of the morning. Obviously, 95 percent of Mac users won't benefit as their computer are either off or asleep at this time of day. The article shows how to run them manually.
    There is a freeware program called "WhatSize" that reports what is eating up disk space but I've not been able to get it to run on a G3 PowerBook under 10.2.8, Seems to need Panther.
    You might post which G3 PowerBook you have; this Apple article will help you identify yours:
    http://docs.info.apple.com/article.html?artnum=24604

  • Zones - reducing disk space used

    We're currently looking at zones to consolidate our license servers, since fault management tends to be easier/faster when the license server is running in its own zone.
    However, as of right now the "sparse" root model configuration requires 578M of space - 495M taken by packages. Having several of zones using the same size like this would eat up disk space quickly.
    Ideally, I'd like the zones to be exactly like the global zone with a few changes under /etc
    Does anyone know a way to reduce or eliminate the number of packages needed to be installed, or other ways I might cut down disk usage further?

    We're trying to figure out how to speed up patching as well -- it seems silly how long it takes to effectively just update the /var/sadm structures for a sparse zone...
    One note though: to reduce space consumed in /var/sadm in a very safe way, you can delete the "obsolete*" files -- these are the backout information from patches that have been superseded at least once, so the only risk is that you will only be able to back out a single "layer" of patching. You can remove the "undo" files as well, but then you cannot backout the related patches.
    Infodoc 14295 has a nice overview of what the files in /var/sadm are for those interested.

  • Arch process using more swap space

    Hi,
    DB : 11.2
    OS : Aix 6
    While taking archive backup by oem,only 9 archive process consuming more swap space.
    The parameter value is LOG_ARCHIVE_MAX_PROCESSES=15 at db level.
    Why only 9 archive process use swap space at os level?
    how to determine all the archive process needed?
    Thanks & Regards,
    VN

    user3266490 wrote:
    Hi,
    Thanks for your reply.
    Physical memory 120 GB,Swap space 45GB.
    We getting TOP consuming SWAP space PID's from OS team.Then we checked v$session with that PID's.We found that PID's belongs to archive processes.
    Nealry 25 log switch per hour happened.the log file size 8gb,8 groups. Then REDO log files is is way too small by around a factor of SIX
    Increase REDO log file size to 50GB to reduce the number of switches per hour to around 3 0r 4.

  • Spotlight eating disk space os x snow leopard

    Hello,
    I have 2 major problems. The first is that Spotlight, for at least 1 or 2 years, has continually eat disk space on a half-hour basis. It would go from gigabytes and drain down to zero and then back up again and then down. This slows down my Mac Pro because the disk is continually writing. Sometimes it would stop if I restarted the computer but it would only be for a couple hours.
    I know that it is Spotlight because when I took my whole hard drive and dragged it to the "excluded" side in System Settings>Spotlight and it stopped doing what I mentioned above. I'm trying to find solutions that do not have to do with reinstalling Snow Leopard as I have so many things that I need to download of the internet and I have to keep doing so over and over everytime I re-install.
    Could it be something with the Spotlight plist? I've checked all my caches and they aren't large at all.
    2nd Major Problem:
    Spotlight is taking up over 20 gigabytes to index. Why?
    Thanks,
    spiral

    Erase the drive and reinstall Snow Leopard.
    Clean Install of Snow Leopard
    Be sure to make a backup first because the following procedure will erase
    the drive and everything on it.
         1. Boot the computer using the Snow Leopard Installer Disc or the Disc 1 that came
             with your computer.  Insert the disc into the optical drive and restart the computer.
             After the chime press and hold down the  "C" key.  Release the key when you see
             a small spinning gear appear below the dark gray Apple logo.
         2. After the installer loads select your language and click on the Continue
             button. When the menu bar appears select Disk Utility from the Utilities menu.
             After DU loads select the hard drive entry from the left side list (mfgr.'s ID and drive
             size.)  Click on the Partition tab in the DU main window.  Set the number of
             partitions to one (1) from the Partitions drop down menu, set the format type to Mac
             OS Extended (Journaled, if supported), then click on the Partition button.
         3. When the formatting has completed quit DU and return to the installer.  Proceed
             with the OS X installation and follow the directions included with the installer.
         4. When the installation has completed your computer will Restart into the Setup
             Assistant. Be sure you configure your initial admin account with the exact same
             username and password that you used on your old drive. After you finish Setup
             Assistant will complete the installation after which you will be running a fresh
             install of OS X.  You can now begin the update process by opening Software
             Update and installing all recommended updates to bring your installation current.
    Download and install Mac OS X 10.6.8 Update Combo v1.1.

  • Problem: There is not enough disk space available to process clips. Clear some space and try again.

    Hi there.
    I am running the newest FCPX 10.0.9 and Mavericks, and I am editing a large multicam project. I have approx. 190 GB left on my external event/project disk via USB3 (it has 2TB in total), and I get the error message:
    "There is not enough disk space available to process clips. Clear some space and try again."
    when I am trying to add clips to an angle in one of my multicam clips. I have made more than 8 multicam clips in the same event, without that problem, but now it won't let me add more.
    What's the problem?

    This was driving me nuts as well until I found the sneaky checkbox. In the editing tab in Preferences there is a checkbox for "Optimize media for multicam." If that's checked, Final Cut is rendering ProRes files for every clip you bring into multicam, so it just eats memory. Unchecking that solved my issues; hope it helps you or anyone who comes across this!

  • Song tracks are not downloaded properly from iTunes Store: they are not added to the playlists, and eat disk space

    Hi,
    Yesterday I wanted to upgrade my wife's iPhone (5, 16 GB) from iOS 7.0.6 to iOS 7.1.
    She had only 500 MB of free space, and 1.9 GB was needed, so I decided to temporarily remove a number of songs from her music library.
    After this operation there was 2 GB free, and the upgrade went successfully.  After the upgrade there was still 2 GB free.
    Then I wanted to download the songs again, and here something went wrong:
    I went to iTunes Store > More > Purchased > Music, and clicked on "Download all".
    When the download process was completed, there was about 500 MB free space, but I noticed that some of the tracks, and even a complete album, were actually not in the playlists; in another album only one of the 30 tracks or so was downloaded.  So I went again into iTunes Store > More > Purchased > Music, and indeed some of the tracks were still marked as "to download", with a little cloud in front of them.
    When I tried to click on one of these little clouds to download a given track, the track seemed to be downloaded, but as soon as the download was finished the little cloud reappeared again, and the track wasn't added to the playlists.
    I tried to repeat that process a few times, until I noticed that there was now only 90 MB of free space.
    I then tried to download one more track that was 10 MB large, and after this there was only 80 MB of free space.
    So it seems that the tracks are indeed downloaded (and downloaded again if I click again on the little cloud) in a place where they use disk space without being otherwise accessible.
    What can I do to:
    1. download the tracks correctly so that they appear in the playlists?
    2. clean up the space that was wasted?
    Many thanks in advance!

    Depending upon what country that you are in (music can't be re-downloaded in all countries) then try deleting the incomplete tracks from your iTunes library and redownload them via the Purchased link under Quick Links on the right-hand side of the iTunes store home page on your computer's iTunes : re-downloading.
    If you aren't in a country where you can re-download music or if they re-download in the same state then try the 'report a problem' link from your purchase history : log into your account on your computer's iTunes via Store > View My Account and you should then see a Purchase History section with a 'see all' link to the right of it ; click on that and you should see a list of your purchases ; find those tracks and use the 'Report a Problem' link and fill in details about the problem (iTunes support should reply within, I think, 24 to 48 hours).
    Some people have had a problem with the 'report a problem' link (it's been taking people to this site on a browser instead of showing a form in iTunes) - if it does that to you then try contacting iTunes support via this page : http://www.apple.com/support/itunes/contact/- click on Contact iTunes Store Support on the right-hand side of the page.

  • How to monitor available disk space in archive directory

    I need to monitor the available disk space of the archive directory during a long running PL/SQL program which does a lot of inserts and updates. This program is running in a background session and should regularly check for the available space. If there is less then a customizable number of MBs free the program must terminate itself. How can I access this information from the file system? (I'm using 8.1.6 and the application will run on both AIX and NT)

    The directory itself is not taking space (It actually does but in most cases you can neglect its size). It is files that it has that take space. Also, the more files you create the more space gets allocates from the media that holds them. And the last, the media (Hard Drive, Floppy, or other) has free space, not directory. That is how all OS's that I know about work.

  • When i import movies from my iPhoto app to my iMovie app does it make a copy (thereby using double the hard disk space) or does it link to my iPhoto app? The reason i ask is i want to archive all my videos from iPhoto/iMovie. HELP!

    When i import movies from my iPhoto app to my iMovie app does it make a copy (thereby using double the hard disk space) or does it link to my iPhoto app? The reason i ask is i want to archive all my videos from iPhoto/iMovie. Do i need to archive the movies in iPhoto AND iMovie? HELP!

    Hi - I'm having the same problem with freeing up space on the HD. If I can't free space if I'm using any part of the clip, how can I split up the clips so that I can discard the part I won't use? This is crazy! I have only 4 GB of free space left!
    Also, I've tried to transfer the project file to my external HD but I keep getting an error saying I can't transfer it. I've been told this is because I don't have enough space on the HD for the transfer. I tried a similar transfer with another computer with plenty of HD space but the same error occurred. Is there a way to break up the project file into small pieces for the transfer (I'm thinking the whole movie file is just too big)?
    Thanks for any info!

  • How do I tell how much hard disk space an archive will take up?

    Hi we are trying to determine whether or not we have the capacity to run an archive export on one of our WebCenter spaces content folder (the root folder and ALL contents). Before we begin we'd like to be able to tell how much disk space it will take up.
    Thanks,
    S

    The table Documents contains a field called dFileSize, which contains the size of file in bytes.
    Now, you can use this field to calculate the size of all files contained in a folder (or its subfolders), or contained in an archiving batch.
    I think easier is the former. Actually, I have done something similar - created a component that checks so that files in a folder do not exceed the given quota. It already contains hierarchical queries that could be re-designed (simplified) to your needs. For instance, an occupied quota of a folder is
    WITH tree AS (select dCollectionId, case when level>1 then dCollectionQuota end as effective_quota from collections start with dCollectionId = ? connect by dParentCollectionId = prior dCollectionId and (prior dCollectionQuota IS NULL or level = 2)) SELECT SUM(NVL(t.effective_quota, NVL(docs.dFileSize, 0))) as TOTALDOCUMENTSIZE from tree t LEFT OUTER JOIN DOCMETA meta ON t.dCollectionId = meta.xCollectionId and t.effective_quota IS NULL LEFT OUTER JOIN DOCUMENTS docs ON meta.dId = docs.dId and docs.dIsPrimary = 1 LEFT OUTER JOIN REVISIONS rev ON rev.dId = docs.dId(in your case, you don't need t.effective_quota, nor logic bound to dCollectionQuota as these values do not exist in your system)
    P.S. more info about hierarchical queries, see here
    http://docs.oracle.com/cd/E11882_01/server.112/e10592/queries003.htm#i2053935
    http://docs.oracle.com/cd/E11882_01/server.112/e10592/statements_10002.htm#i2077142

Maybe you are looking for