JOXSHM files in /dev/shm

I have a bunch of old files with names like JOXSHM_EXT_0_DEMDB_121208837 in the /dev/shm folder. I believe these are suppose to be removed when the database is shutdown. The DEMDB instance has been down for months. Do I have to reboot to remove these or is it safe to delete them?
I am running 11.2 on REHL 5

uptime
10:36:01 up 229 days, 1:16, 3 users, load average: 0.46, 0.49, 0.45

Similar Messages

/dev/shm not release space after delete file.

when I remove file from /dev/shm, the available space of shm still not increase.
Is it having a command for shrinking shm ?

http://www.linuxquestions.org/questions … on-806387/
Edit: Even the wiki has a thing or two on this subject https://wiki.archlinux.org/index.php//dev/shm
Last edited by karol (2011-05-29 10:22:27)

/dev/shm not mounted /dev busy

Recently I was trying out a boot disk I had made, and basically, I switched it off several times due to it booting the completely wrong kernel.
Now, I'm getting to the stage "Checking File systems" and then comes up [Failed]. I then get a message saying Reboot required, and that it will reboot in 15 seconds, just a few seconds before it reboots I get "/dev/shm not mounted, /dev busy" or something similar.
I've booted up my sysresccd, ran "fsck.ext4 -fcv /dev/sda2" to force a check and scan for any bad blocks, it came up clean, then I rebooted and got the same error, so I copied the kernel and system.map over to /boot to make sure there's no corruption and reinstalled initscripts and util-linux-ng, rebooted, same error.
Tried different kernels, I've checked fstab and menu.lst, no problems there, so I still don't get why I still get the same problem.
Anybody know of any fix other than reinstalling arch?
EDIT: Editing /etc/rc.sysinit and commenting out the fsck part made it work, finding out why fsck failed now.
Last edited by compgenius999 (2010-03-03 19:21:35)

anybody?

/dev/shm on Oracle Linux 6.x to run Oracle 11g R2 - manual configuration?

Hello
We are building a server to run Oracle 11g R2 database (11.2.0.3 x64) on Oracle Linux 6.2 with UEK R2.
Our preference is to use AMM to have Oracle 11g R2 manage memory. We may impose some minium SGA and PGA memory allocations but basically aim to use MEMORY_TARGET to manage overall memory.
By default Linux makes the size of /dev/shm ~50% of server physical RAM, as far as I can tell.
Here is the /etc/fstab entry created by the installation:
tmpfs /dev/shm tmpfs defaults 0 0
Given this Linux server will only run Oracle 11g R2 database and some monitoring software, almost application code will run on the server. The application code will run on the separate application server and is Java based.
Can I change the */etc/fstab* entry for /dev/shm to manually increase the size to take up ~80-90% of the server physical RAM ? Is it a good idea?
The server is 64-bit, the RAM = 64 GB, so I am thinking to manually make /dev/shm to be = ~55GB, leaving ~8GB for other purposes.
Right now it's about 32GB (50%?) if I leave the /dev/shm 'defaults' on.
many thanks

thanks,
I have read the doc (what little there is on this topic).
I have asked on the database forum......
just FYI - below is the proof:
SQL> show parameter mem
NAME                    TYPE     VALUE
hi_shared_memory_address     integer     0
memory_max_target          big integer 4G
memory_target          big integer 0
shared_memory_address     integer     0
SQL> show parameter ga
NAME                    TYPE     VALUE
lock_sga               boolean     FALSE
pga_aggregate_target          big integer 1600M
pre_page_sga          boolean     FALSE
sga_max_size          big integer 3G
sga_target               big integer 1600M
still does not work.
And I cant change memory_max_target = 0 because I get error on startup:
SQL> alter system set memory_max_target=0 scope=spfile;
System altered.
SQL> shutdown immediate;
Database closed.
Database dismounted.
ORACLE instance shut down.
SQL> startup;
ORA-01078: failure in processing system parameters
ORA-00843: Parameter not taking MEMORY_MAX_TARGET into account
ORA-00849: SGA_TARGET 3221225472 cannot be set to more than MEMORY_MAX_TARGET 0.
BUT if memory_max_target is > 0 then alert log says hugepages can not be used
it feels like catch-22.....
thanks
Edited by: yurib on Jun 1, 2012 4:53 PM

Upgrade breakage (file called /dev/tty[0-9]*)

Today I updated a laptop with the usual pacman -Syu. After the upgrade (with no errors visible in pacman.log) i rebooted into a fscked system. Instead of /dev/tty0, /dev/tty1, ... I had a file called /dev/tty[0-9]* , and when the boot process completed I was left without a gui or a console login.
I have no idea what could have caused this as I updated successfully on several other machines.
Is there a quick fix for this? I tried reinstalling the base group, reinstalling ALL packages, still no go. If possible I'd like to avoid wiping the root partition and installing/configuring all over again - any ideas?
Thanks,
schuay
Last edited by schuay (2009-09-25 19:41:12)

I encountered this problem and was able to solve it using this thread, so thanks, everyone!
Also, I did quick check to see what the differences in the /etc/udev/ directory were:
liquidsunshine@ruaidhri:/etc$ diff -r udev udev-bak
Only in udev-bak: lomoco_mouse.conf
Only in udev-bak/rules.d: 10-cups-usb.rules
Only in udev-bak/rules.d: 45-libnjb.rules
Only in udev-bak/rules.d: 60-pcmcia.rules
Only in udev-bak/rules.d: 60-virtualbox.rules
Only in udev-bak/rules.d: 65-libmtp.rules
Only in udev-bak/rules.d: 80-lomoco.rules
Only in udev-bak/rules.d: 90-hal.rules
Only in udev-bak/rules.d: 97-bluetooth-serial.rules
Only in udev-bak/rules.d: 99-fuse.rules
Only in udev-bak/rules.d: device-mapper.rules
Only in udev-bak/rules.d: udev.rules
and the same for /lib/udev/:
liquidsunshine@ruaidhri:/lib$ diff -r udev udev-bak/
Only in udev-bak/: bluetooth_serial
Only in udev-bak/: check-mtp-device
Only in udev-bak/: check-ptp-camera
File udev/devices/console is a character special file while file udev-bak/devices/console is a character special file
File udev/devices/fuse is a character special file while file udev-bak/devices/fuse is a character special file
File udev/devices/kmsg is a character special file while file udev-bak/devices/kmsg is a character special file
File udev/devices/loop0 is a block special file while file udev-bak/devices/loop0 is a block special file
File udev/devices/net/tun is a character special file while file udev-bak/devices/net/tun is a character special file
File udev/devices/null is a character special file while file udev-bak/devices/null is a character special file
File udev/devices/ppp is a character special file while file udev-bak/devices/ppp is a character special file
File udev/devices/zero is a character special file while file udev-bak/devices/zero is a character special file
Only in udev-bak/: devkit-disks-dm-export
Only in udev-bak/: devkit-disks-part-id
Only in udev-bak/: devkit-disks-probe-ata-smart
Only in udev-bak/: lomoco.sh
Only in udev-bak/rules.d: 52-libmtp.rules
Only in udev-bak/rules.d: 53-sane.rules
Only in udev-bak/rules.d: 54-gphoto.rules
Only in udev-bak/rules.d: 55-hpmud.rules
Only in udev-bak/rules.d: 56-hpmud_support.rules
Only in udev-bak/rules.d: 95-devkit-disks.rules
Only in udev-bak/rules.d: 95-devkit-power-battery-recall-dell.rules
Only in udev-bak/rules.d: 95-devkit-power-battery-recall-fujitsu.rules
Only in udev-bak/rules.d: 95-devkit-power-battery-recall-gateway.rules
Only in udev-bak/rules.d: 95-devkit-power-battery-recall-ibm.rules
Only in udev-bak/rules.d: 95-devkit-power-battery-recall-lenovo.rules
Only in udev-bak/rules.d: 95-devkit-power-battery-recall-toshiba.rules
Only in udev-bak/rules.d: 95-devkit-power-csr.rules
Only in udev-bak/rules.d: 95-devkit-power-hid.rules
Only in udev-bak/rules.d: 95-devkit-power-wup.rules
Only in udev-bak/rules.d: 97-bluetooth-hid2hci.rules
Only in udev-bak/rules.d: 97-bluetooth.rules
Unfortunately, I'm not curious enough at this point to do more in-depth investigation, but it looks to me like there were quite a few left-over files lying around from previous udev upgrades...

PRKC-1031: error checking free space for /dev/shm

Dear,
hope someone can help me with this error.
install oracle 11g on linux OUL5 2 nodes cluster.
the cluster is fine. when install database, i always have this error. did increase /dev/shm to 2G but still get this error.and it would not let me continue.
the specific Prerequisite checks were fine.only failed on swap
Linux rac11g1 2.6.18-92.1.13.0.1.el5PAE #1 SMP
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/mapper/VolGroup00-LogVol00
22282108 8542032 12589920 41% /
/dev/hda1 101086 40264 55603 42% /boot
tmpfs 2097152 0 2097152 0% /dev/shm
/dev/mapper/VolGroup01-LogVol01
25189484 769252 23140644 4% /u01
Thanks in advance.

Hi
you are quite welcome, so you don't any issues with the user equiv part during the cluvfy controls?
how many instances are you running on your server that hosts ASM?
More than 1 then read that: Did you change your shm values in the kernel ? Please check those values in 2.6 Configuring Kernel Parameters
64 bits: http://download.oracle.com/docs/cd/B19306_01/install.102/b15667/pre_install.htm#i1011296
32 bits: http://download.oracle.com/docs/cd/B19306_01/install.102/b15660/pre_install.htm#sthref264
Note that you have to increase : semmni, semmns,shmmni should be multiplied by # of instances, SHMMAX should be set to avail memory
Can you please give us your values from your server please?
Your physical memory :
grep MemTotal /proc/meminfo
And also the result of
ipcs -l
Edited by: Hub on Oct 24, 2008 11:07 PM

Compile in /dev/shm?

Hello,
I have an SSD installed and followed the instructions on wiki page: https://wiki.archlinux.org/index.php/SSD
Now I'm in heavy developement for a project, using gwt. How can I ensure that if I compile my Application, that it takes place in /dev/shm or some other place in vm?

https://wiki.archlinux.org/index.php/Makepkg.conf
https://wiki.archlinux.org/index.php//dev/shm

[SOLVED] Trying to allocate more ram to /dev/shm

I want to allocate more than 50% ram to '/dev/shm'.
In the wiki it says to edit '/etc/fstab' and add the size parameter but I know that '/etc/fstab' no longer mounts '/dev/shm', '/etc/rc.sysinit' does.
Is it safe to add the '/dev/shm' entry in to '/etc/fstab or should I edit '/etc/rc.sysinit' now?
Thanks for any responses
Last edited by pluckypigeon (2012-10-01 23:28:04)

You should add an entry to /etc/fstab.
rc.sysinit (or systemd) will indeed mount /dev/shm with standard options in early boot, but then rc.sysinit (or systemd) will call /usr/lib/systemd/systemd-remount-fs, which will remount all the api filesystems (such as /dev/shm) that have entries in fstab, with the correct options specified there.

Question about 11gR2 Grid, RAC, /dev/shm and Automatic Memory Management

Hello,
i've recently installed grid and rdbms software 11.2.0.2 on a two node Oracle Linux cluster with 128gb ram each node.
I'm using ASM to store data and ocr and I'm testing Automatic Memory Management.
When I finished Grid+RDBMS installation I've seen that /dev/shm size is 64gb (half of my total RAM).
I've created a database with dbca and when I was asked to choose if I wanted to use AMM I've noticed that I could
allocate only about 60gb for Oracle. If I chose more than 90gb I got an error saying:
Using Automatic Memory Management requires 60gb available in my two nodes.
The current available space in the two nodes is only 30gb and 30gb.
If you want to use AMM you should either free up some space in /dev/shm
or reduce the memory allocated to Oracle
I was wondering when (during the installation or the settings of kernel parameters) did I define the space of /dev/shm ?
Since I have 128gb of RAM wouldn't it be better to use more than 64gb of ram for my /dev/shm tmpfs partition ?
Is there a limit or a ratio for best practice for my RAM and the /dev/shm ?
thanks in advance.

user9051299 wrote:
Is the "half of the RAM size" a kernel's default value or Oracle's ? Neither. There are a number of unique factors that determine the best memory size and fit for Oracle - including just how much memory is effectively available (i.e. how much is needed for other services and processes).
And from what I understand i don't "break" any Oracle's best practice by increasing the /dev/shm right ?Correct. (at least none that I'm aware of, and none that I have read in Oracle's RAC Starter Kit documentation).

Problem running swf file outside dev environment

If I copy my bin-debug folder outside the dev environment
i.e. any where in hard disk and run the html file the swf is unable
to connect to the HttpService.
Is there any way to solve this problem.. Its working fine
when i am running it from inside of flex builder bin-debug dir.
"Crossdomain.xml" is also provided.
Thanks

This is probably due to security restrictions of running a
file on your local machine that access network resources. Try
running the SWF file via a network request and not opening it as a
file (use the
http:// protocol rather than file:// in your
browser).
matt horn
flex docs

Accessing java files for Dev't.

I would like to know how my developers can get access to some java files to facilitate the customization of the web interface, for my company.
e.g. oracle.ifs.clients.webui or
viewer.java
Thanks..

Please see the following post: http://technet.oracle.com:89/ubb/Forum36/HTML/000749.html

Temps on /dev/shm

I'd like to know if there are any temp directories besides /tmp and /var/tmp I can/could/should mount to a tmpfs and how I would go on about doing this. For starters I don't even know how big it's got to be and if I have to do this via fstab or some kind of bind.
Thanks for your input.

Thank you for your input! Is there a reason why you propose 250M? On gentoo I had the problem with big compiles that /var/tmp quickly ran out of space (at least with OOo), I don't want that again. So if I realize that I might need more space on a given compile, taken that arch uses /var/tmp for the build process, could I just assign more RAM to it (on reboot and, of course, after the process ran out of space lol)?
Besides, why the profiles folder, does that promote fragmentation, too? What happens to my FF settings and plugins and such?
I'm asking all this for two reasons, one being speed, the other less fragmentation. At first I thought I'll just give each a partition, as I always did on my installs. But when I read that I could just point it all to shm it was more appealing. I'm not saying that I'll immidiately fiddle with my own compiles, since it's been a while that my hdd has met the penguin, I need to reeducate myself. Arch will be my new home, since the-compile-everything-no-matter-if-it-even-needs-more-rice motto was just too much for my taste. Extra speed where it's needed makes more sense to me. End of rant, hehe.
Last edited by p2501 (2009-07-25 15:55:05)

Clean up file system size

Hi,
I found   /dev/shm file system utilization is more in DB server as below. Please help me to reduce the space.
$ df -kh
Filesystem Size Used Avail Use% Mounted on
/dev/mapper/root_vg-rootvol
38G 3.0G   33G   9% /
tmpfs 7.8G 7.1G 792M 91% /dev/shm
Files under /dev/shm
I don’t know what these files are.
-rw-r-----. 1 oracle oinstall 16777216 Apr 18 23:04 ora_SAPLAT02_1245213_2
-rw-r-----. 1 oracle oinstall 16777216 Apr 18 23:18 ora_SAPLAT02_1245213_3
-rw-r-----. 1 oracle oinstall 16777216 Apr 18 23:38 ora_SAPLAT02_1245213_4
-rw-r-----. 1 oracle oinstall 16777216 Apr 18 23:34 ora_SAPLAT02_1245213_5
-rw-r-----. 1 oracle oinstall 16777216 Apr 18 23:15 ora_SAPLAT02_1245213_6
-rw-r-----. 1 oracle oinstall 16777216 Apr 18 23:34 ora_SAPLAT02_1245213_7
Files under /dev/shm with sizes
oracle@server01 /dev/shm
$ du -sh * | grep G
4.0M    ora_COGNOS_917525_0
4.0M    ora_COGNOS_917525_1
1.6M    ora_COGNOS_917525_2
0       ora_COGNOS_950294_0
0       ora_COGNOS_950294_1
0       ora_COGNOS_950294_10
4.0M    ora_COGNOS_950294_100
4.0M    ora_COGNOS_950294_101
4.0M    ora_COGNOS_950294_102
4.0M    ora_COGNOS_950294_103
4.0M    ora_COGNOS_950294_104
4.0M    ora_COGNOS_950294_105
4.0M    ora_COGNOS_950294_106
4.0M    ora_COGNOS_950294_107
Regards,
Bala

2790985 wrote:
Hi,
I found   /dev/shm file system utilization is more in DB server as below. Please help me to reduce the space.
$ df -kh
Filesystem Size Used Avail Use% Mounted on
/dev/mapper/root_vg-rootvol
38G 3.0G   33G   9% /
tmpfs 7.8G 7.1G 792M 91% /dev/shm
Files under /dev/shm
I don’t know what these files are.
-rw-r-----. 1 oracle oinstall 16777216 Apr 18 23:04 ora_SAPLAT02_1245213_2
-rw-r-----. 1 oracle oinstall 16777216 Apr 18 23:18 ora_SAPLAT02_1245213_3
-rw-r-----. 1 oracle oinstall 16777216 Apr 18 23:38 ora_SAPLAT02_1245213_4
-rw-r-----. 1 oracle oinstall 16777216 Apr 18 23:34 ora_SAPLAT02_1245213_5
-rw-r-----. 1 oracle oinstall 16777216 Apr 18 23:15 ora_SAPLAT02_1245213_6
-rw-r-----. 1 oracle oinstall 16777216 Apr 18 23:34 ora_SAPLAT02_1245213_7
Files under /dev/shm with sizes
oracle@server01 /dev/shm
$ du -sh * | grep G
4.0M    ora_COGNOS_917525_0
4.0M    ora_COGNOS_917525_1
1.6M    ora_COGNOS_917525_2
0       ora_COGNOS_950294_0
0       ora_COGNOS_950294_1
0       ora_COGNOS_950294_10
4.0M    ora_COGNOS_950294_100
4.0M    ora_COGNOS_950294_101
4.0M    ora_COGNOS_950294_102
4.0M    ora_COGNOS_950294_103
4.0M    ora_COGNOS_950294_104
4.0M    ora_COGNOS_950294_105
4.0M    ora_COGNOS_950294_106
4.0M    ora_COGNOS_950294_107
Regards,
Bala
why do you think the current value is a problem?
what will be gained by making percentage smaller?
reduce value of PROCESSES parameter
post results of command below
ls -l /dev/shm
explain what you learned from what you saw
What do you see when Oracle DB has been shutdown?

Dynamic Accordion, fetching node name instead of value. Is it possible?

Here is the xml.
<?xml-stylesheet type="text/xsl" href="myweb.xsl"?>
<client product="ppd" date="4/26/11 3:17 AM">
<system>
       <osname>Linux
          </osname>
       <hostname>abhishek
          </hostname>
       <release>2.6.18-128.el5
          </release>
       <version>Red Hat Enterprise Linux Server release 5.3 (Tikanga)
          </version>
       <machine>ia64
          </machine>
       <bitmode>64
          </bitmode>
       <filesystem>
          <file mount='Mounted' home='Filesystem' total='1K-blocks' free='Available' used='Used' percentage='Use%' />
          <file mount='/' home='/dev/sda3' total='60300484' free='38305668' used='18882312' percentage='34%' />
          <file mount='/boot/efi' home='/dev/sda1' total='1046516' free='1037912' used='8604' percentage='1%' />
          <file mount='/dev/shm' home='tmpfs' total='4120800' free='4120800' used='0' percentage='0%' />
       </filesystem>
</system>
</client>
my spry region. Using {systemData::osname} I can point to "Linux" but instead I would like to get osname itself. So basically I want to make a tree like structure using accordion where data within tree comes from this xml's node names.
<div spry:region="systemData">
<div id="Accordion1" class="Accordion" tabindex="0">
<div spry:repeat="systemData" class="AccordionPanel">
<div class="AccordionPanelTab">{systemData::osname}</div>
<div class="AccordionPanelContent">
{systemData::hostname}
</div>
</div>
</div>
<script type="text/javascript">
var a1 = new Spry.Widget.Accordion("Accordion1");
</script>
</div>
last but not least here is my spry dataset
var systemData=new Spry.Data.XMLDataSet("include/xml/clientabhishek.xml","client/system");

hehe. Sorry for not being so clear. I wanted to know whether there is a way I can get osname(node name) instead of Linux (node value) from my dataset. Like using {osname} I can get value of osname but I was wondering if osname itself can be fetched someway. Is there a way? Thanks for passing by

ORA-29516: Bulk load of method failed; insufficient shm-object space

Hello,
Just installed 11.2.0.1.0 on CentOS 5.5 64-bit. All dependencies satisfied, installation/linking went without a problem.
Server has 32GB RAM, using AMM with target set at 29GB, no swapping is occuring.
No matter what i do when loading Java code (loadjava with JARs or "create and compile java source") I keep getting the error:
ORA-29516: Error in module Aurora: Assertion failure at joez.c:3311
Bulk load of method java/lang/Object.<init> failed; insufficient shm-object space
Checked shm-related kernel params, all seems to be normal:
# Controls the maximum size of a message, in bytes
kernel.msgmnb = 65536
# Controls the default maxmimum size of a mesage queue
kernel.msgmax = 65536
# Controls the maximum shared segment size, in bytes
kernel.shmmax = 68719476736
# Controls the maximum number of shared memory segments, in pages
kernel.shmall = 4294967296
kernel.shmmni = 4096
kernel.sem = 250 32000 100 128
net.core.rmem_default = 262144
net.core.rmem_max = 4194304
net.core.wmem_default = 262144
net.core.wmem_max = 1048576
Please help.

Hi there,
I've stumbled into exactly the same issue for 11g. After I start the database up and I ran loadjava on an externally
compiled class (Hello.class in my instance) I got the following error:
Error while testing for existence of dbms_java.handleMd5
ORA-29516: Aurora assertion failure: Assertion failure at joez.c:3311
Bulk load of method java/lang/Object.<init> failed; insufficient shm-object space
ORA-06512: at "SYS.DBMS_JAVA", line 679
Error while creating class Hello
ORA-29516: Aurora assertion failure: Assertion failure at joez.c:3311
Bulk load of method java/lang/Object.<init> failed; insufficient shm-object space
ORA-06512: at line 1
The following operations failed
class Hello: creation (createFailed)
exiting : Failures occurred during processing
After this, I checked the trace file and saw the following error message:
peshmmap_Create_Memory_Map:
Map_Length = 4096
Map_Protection = 7
Flags = 1
File_Offset = 0
mmap failed with error 1
error message:Operation not permitted
ORA-04035: unable to allocate 4096 bytes of shared memory in shared object cache "JOXSHM" of size "134217728"
peshmmap_Create_Memory_Map:
Map_Length = 4096
Map_Protection = 7
Flags = 1
File_Offset = 0
mmap failed with error 1
error message:Operation not permitted
ORA-04035: unable to allocate 4096 bytes of shared memory in shared object cache "JOXSHM" of size "134217728"
Assertion failure at joez.c:3311
Bulk load of method java/lang/Object.<init> failed; insufficient shm-object space
It seems as though that "JOXSHM" of size "134217728" (which is 128MB) corresponds to the java_pool_size setting in my init.ora file:
memory_target=1000M
memory_max_target=2000M
java_pool_size=128M
shared_pool_size=256M
Whenever I change that size it propagates to the trace file. I also picked up that only 592MB of shm memory gets used. My df -h dump:
Filesystem Size Used Avail Use% Mounted on
/dev/sda7 39G 34G 4.6G 89% /
udev 10M 288K 9.8M 3% /dev
/dev/sda5 63M 43M 21M 69% /boot
/dev/sda4 59G 45G 11G 81% /mnt/data
shm 2.0G 592M 1.5G 29% /dev/shm
The only way in which I could get loadjava to work was to remove java from the database by calling the rmjvm.sql script.
After this I installed java again by calling the initjvm.sql script. I noticed that after these scripts my shm-memory usage
increased to about 624MB which is 32MB larger than before:
Filesystem Size Used Avail Use% Mounted on
/dev/sda7 39G 34G 4.6G 89% /
udev 10M 288K 9.8M 3% /dev
/dev/sda5 63M 43M 21M 69% /boot
/dev/sda4 59G 45G 11G 81% /mnt/data
shm 2.0G 624M 1.4G 31% /dev/shm
However, after I stopped the database and started it again my Java was broken again and calling loadjava produced
the same error message as before. The shm memory usage would also return to 592MB again. Is there something I
need to do in terms of persisting the changes that initjvm and rmjvm does to the database? Or is there something else
wrong that I'm overlooking like the memory management settings or something?
Regards,
Wiehann

JOXSHM files in /dev/shm

Similar Messages

Maybe you are looking for