/var filesystem 100% full

I need some help here please, I cleared /var filesystem from all the big files but it's still showing me 100% full, I only manage to get rid of this problem when I reboot the server then /var becomes 3% full, any idea what's causing this to happen.
I want to avoid rebooting the server in future, should /var/ become full again
Many thanks

Some process still has the deleted file open, so the kernel cannot
free the space occupied by the file(s) you've delete.
The space is deleted as soon as the process closes the file handle
(e.g. by stopping / restarting the process).
The following command lists the process ids / file handles that
do not have filesystem links but are still referenced by a process
(you must run it as user root):
# find /proc/*/fd -type f -links 0 -ls

Similar Messages

Unable to install gcc, says /var filesystem full

Hi,
when i am trying to intall gcc on solaris9, it says
NOTICE: alloc: /var: file system full
cpio: Cannot write "reloc/lib/libstdc++.a", errno 28, No space left on device
cpio: Cannot write "reloc/lib/libstdc++.so.5.0.5", errno 28, No space left on device
cpio: Cannot write "reloc/lib/libsupc++.a", errno 28, No space left on device
cpio: Cannot write "reloc/man/man1/cpp.1", errno 28, No space left on device
cpio: Cannot write "reloc/man/man1/gcc.1", errno 28, No space left on device
cpio: Cannot write "reloc/man/man1/gcov.1", errno 28, No space left on device
cpio: Cannot write "reloc/man/man7/fsf-funding.7", errno 28, No space left on device
cpio: Cannot write "reloc/man/man7/gfdl.7", errno 28, No space left on device
cpio: Cannot write "reloc/man/man7/gpl.7", errno 28, No space left on device
124787 blocks
when i try to check the disk free using df -h, i get output as
Filesystem size used avail capacity Mounted on
/dev/dsk/c0d0s0 149M 32M 103M 24% /
/dev/dsk/c0d0s6 1.3G 451M 784M 37% /usr
/proc 0K 0K 0K 0% /proc
mnttab 0K 0K 0K 0% /etc/mnttab
fd 0K 0K 0K 0% /dev/fd
/dev/dsk/c0d0s1 52M 3.6M 43M 8% /var
swap 585M 12K 585M 1% /var/run
swap 585M 0K 585M 0% /tmp
i can see clearly tht my first slice which i have used for /var filesystem is not full. what might be the problem?
thanx in advance
regards
bushan.

hi,
I am having a similar problem while trying to install gcc...This is the error messg I get-
cpio: Cannot write "reloc/man/man7/gfdl.7", errno 28, No space left on device
cpio: Cannot write "reloc/man/man7/gpl.7", errno 28, No space left on device
210439 blocks
2 error(s)
pkgadd: ERROR: attempt to process datastream failed
    - process </usr/bin/cpio -icdumD -C 512> failed, exit code 2
pkgadd: ERROR: unable to unpack datastream
Installation of <SMCgcc342s> failed (internal error).
No changes were made to the system.But when I do a df - h it looks like I have enough space...then why am I getting this error?
bash-2.05$ df -h
Filesystem             size   used avail capacity Mounted on
/dev/dsk/c0t0d0s0      2.5G   2.3G   138M    95%    /
/proc                    0K     0K     0K     0%    /proc
mnttab                   0K     0K     0K     0%    /etc/mnttab
fd                       0K     0K     0K     0%    /dev/fd
swap                   3.6G   176K   3.6G     1%    /var/run
swap                   3.6G   456K   3.6G     1%    /tmp
/dev/dsk/c0t0d0s7       14G   2.5G    11G    19%    /export/home
/vol/dev/dsk/c1t6d0/networker_vol_1
                       464M   464M     0K   100%    /cdrom/networker_vol_1
/export/home/dupont     14G   2.5G    11G    19%    /home/dupontthanks,
R

Oracle filesystem is 100% full

when using sql to check the tablespaces. One of the tablespaces shows me that it is 85% used, but when I bdf
(unix level tool) the filesystem where that tablespace resides, it shows me that is 100% full.
Can anyone explain why?

When you add a datafile to a tablespace, Oracle requests a file of that size from the O/S and formats it for it's own use. As a result, if you request a 50M file, Oracle will grab 50M of disk space.
Within the file, if Oracle has not allocated blocks to a segment, then it considers those blocks to be "free". So, at some point after the datafile(s) for that tablespace were allocated, the filesystem filled up, but Oracle has not yet allocated all of the available space in that tablespace.
If I were you, I would check right now to make sure that no datafiles that reside on that filesystem are set for autoextend.
John

[SOLVED] sda3 = 100% full, and I can't figure how to clean it out.

It turns out my sda3 partition is 100% full. My immediate reaction to this apart from being irritated was to clear my pacman cache, thinking this would solve the problem. It didn't. I've also tried pruning out some egregiously large log files but that doesn't do anything either.
Apart from clearing the cache and getting rid of log files though I'm not sure what else I can do to get this thing trimmed down to size. Presumably someone out there has run into a similar situation and solved it? Any advice on how to deal with it? Thanks for your time, everyone.
Last edited by dmadenski (2012-02-06 23:41:43)

Oddly enough, I just came home with intent to keep hacking away at this issue and I found that I was able to run xinit. I hadn't been able to do that before due to this disk space issue but I certainly won't say no to a little deus ex machina action.
Here's some more info now that I'm back on the machine itself:
[dmadenski@yggdrasil ~]$ df -h
Filesystem Size Used Avail Use% Mounted on
rootfs 7.3G 6.9G 14M 100% /
/dev 983M 0 983M 0% /dev
/run 986M 272K 985M 1% /run
/dev/sda3 7.3G 6.9G 14M 100% /
shm 986M 1.3M 984M 1% /dev/shm
/dev/sda1 99M 34M 60M 37% /boot
/dev/sda4 221G 36G 174G 18% /home
[dmadenski@yggdrasil ~]$ df -hi
Filesystem Inodes IUsed IFree IUse% Mounted on
rootfs 470K 325K 145K 70% /
/dev 246K 607 246K 1% /dev
/run 247K 413 246K 1% /run
/dev/sda3 470K 325K 145K 70% /
shm 247K 8 247K 1% /dev/shm
/dev/sda1 26K 46 26K 1% /boot
/dev/sda4 15M 68K 14M 1% /home
[dmadenski@yggdrasil ~]$ sudo du -a /var | sort -n -r | head -n 20
67840 /var
52272 /var/lib
37876 /var/lib/pacman
35396 /var/lib/pacman/local
10444 /var/log
9276 /var/lib/mlocate
9272 /var/lib/mlocate/mlocate.db
5040 /var/cache
3504 /var/cache/man
3132 /var/lib/upower
2676 /var/log/pm-powersave.log
2488 /var/cache/man/index.db
2476 /var/lib/pacman/sync
1668 /var/lib/pacman/local/chromo-erectus-icon-themes-1.2-1
1660 /var/lib/pacman/local/chromo-erectus-icon-themes-1.2-1/files
1300 /var/log/wicd
1232 /var/lib/pacman/sync/extra.db
1156 /var/lib/pacman/local/libreoffice-sdk-3.4.5-2
1148 /var/lib/pacman/local/libreoffice-sdk-3.4.5-2/files
1096 /var/lib/pacman/sync/community.db
Per your suggestion, karol, I downloaded ncdu and ran it. Here are the results.
ncdu 1.8 ~ Use the arrow keys to navigate, press ? for help
. 35.7GiB [##########] /home
6.3GiB [# ] /usr
140.2MiB [ ] /opt
139.4MiB [ ] /lib
66.2MiB [ ] /var
32.2MiB [ ] /boot
13.4MiB [ ] /etc
11.3MiB [ ] /sbin
5.7MiB [ ] /bin
272.0KiB [ ] /run
228.0KiB [ ] /dev
184.0KiB [ ] /root
56.0KiB [ ] /tmp
16.0KiB [ ] /srv
e 16.0KiB [ ] /lost+found
4.0KiB [ ] /lib64
e 4.0KiB [ ] /mnt
e 4.0KiB [ ] /media
0.0 B [ ] /proc
0.0 B [ ] /sys
Total disk usage: 42.4GiB Apparent size: 128.1TiB Items: 470053

Anyone having same problem? I phone 4 after updating to Vers.5.1.1. battery cannot be 100% full charge, always stop at 99% and battery drains off fast too :(

Anyone having same problem? I phone 4 after updating to Vers.5.1.1. battery cannot be 100% full charge, always stop at 99% and battery drains off fast too

I had major issues with the iPhone 4s battery, however it’s resolved.
The tech who set the phone up at the Apple store did so with little training.
if you have a mobile me account. First go and move all your data to the cloud by going on your computer and logging in at me.com/move. The cloud has replaced mobile me, so there is no need for those two accounts
Also make sure that for any of your email accounts you set them up to fetch, not push. My tech person set them all to have the email servers push data to the phone. The new iphone4s antenna is extremely strong so it will continually try to access stuff that is pushed–***** a lot of battery life doing this. It makes it worse if you have exchange 2010 accounts. Something about changes made to exchange really suck battery life from devices that access such accounts.
turning of locator and the push notifications from facebook--they have a lot!

How do I set ethernet speed to 100 Full?

Right now we connect at 100 half duplex. I need to change our Mac lab to 100 full but I can't find how to do it. We are running Imac G5's with Tiger. Please help. Thanks in advance

System Preferences>Network>Show:>Ethernet>Ethernet Tab>Configure=Manually, Speed=100, Duplex=Full... Apply.

/temp filesystem is full

I apologise for being a plonker, but I have a problem installing Solaris on a Compaq E500 laptop.
I try to install Solaris 8 X86 version on this laptop following Sun's instructions to the letter until it gets to the FDISK stage then problems occur.
The laptop has an 18GB hard disk and the BIOS does not correctly report the size of the hard disk to FDISK. FDISK says that it is only 8GB big. I have flashed the BIOS with the latest version from Compaq but it still has the same results.
However, FDISK does let me create Solaris partitions and continue the installation. Everything is fine until we get to the section "Copying mini-root", preparing to eject media and restart etc, then the message /temp filesystem is full (something like that) appears several times and the machine then locks. This happens regardless of how big I set the swap partition.
I am making a big assumption here that this is caused by the hard disk size not being correctly reported. Is there anyway around this problem?
Many thanks,
Richard.

Richard,
Two things: First, try booting the "Software 1 of 2" cdrom rather than the "Installation" one. This skips the whole "copying miniroot" business and runs from the cdrom miniroot during the install.
Second; make sure you are using the latest possible DCA floppy if you are using one.
There may still be driver issues you need to address on the E500 itself but these two step should get you going in the right direction.
Don

Hard code spwr network card for 100-full

How can I hard code a spwr network card for 100-full duplex?
thanks,
Clark

The following lines added to /etc/system sets a hme network card to 100 full duplex,
set hme:hme_adv_autoneg_cap=0
set hme:hme_adv_100T4_cap=0
set hme:hme_adv_100fdx_cap=1
set hme:hme_adv_100hdx_cap=0
set hme:hme_adv_10fdx_cap=0
set hme:hme_adv_10hdx_cap=0

PSAPUNDO tablespace shows 100% full over a long time in oracle 11.2

Dear Experts,
In our server PSAPUNDO table space showing 100% full since three weeks and due to this space critical objects found in PSAPUNDO tablespace.
PSAPUNDO has automatic management enable.
Referred Note 1035137 but no clue found.
Kindly suggest on this.
Thanks,
Bharath.

Deepak Kori wrote:
Hi Bharath,
From the information above PSAPUNDO tablespace has autoextend off.
You need to change is to on for all the datafiles or add a new datafile with auto extend ON.
Hope this helps.
Regards,
Deepak Kori
O.o
sorry, why?
clearly you do not know how undo works and the SAP recommendations.
Automatic UNDO management was introduced in Oracle 9i as indicated on the SAP note 600141. Oracle will try to keep the information the "retention_time" minutes.
from Oracle 10g the behaviour will change a little (explained on the note 1035137).
TRY, that is an important word, Unless you specify otherwise when create the tablespace, Oracle will reuse the data if there is no free space. (that is not the default behaviour).<= corrected here, typo
But, going to the OP question. Oracle will use all the free space on the tablespace, there is no problem with that, and, when full, will start reusing the expired segments. In case there are no expired segments then it will reuse the non expired segments. Depending on the size of the tablespace this would happen or not.
therefore, having the PSAPUNDO "full" means nothing, it is normal, expected.
the problem could be if Oracle is reusing NON expired segments.
The recommendation is, for new installations where you do not know the activity, to keep the PSAPUNDO autoextend so it grows to the "normal" size for your business the operation.
then you deactivate the AUTOEXTEND and monitor if need to use non expired undo
Please, check the Oracle documentation for more information:
http://docs.oracle.com/cd/E11882_01/server.112/e25494/undo.htm#ADMIN013
Message was edited by: Fidel Vales

My PSAPTEMP tablespace always show 100% Full or Used

Dear All ,
Using brtools my tablespace PSAPTEMP always show 100% full or Used, I extend 2 Gb size in PSAPTEMP tablespace but still no solution
Please guide me why it always shows 100% used and what is the performance impact on that
Regards
Shahid

Hello Shahid,
> Please guide me why it always shows 100% used and what is the performance impact on that
This is a normal behaviour in case of a temporary tablespace.
If you do sorts, hash-joins, etc. which can not be done in memory-only you allocate extents in the temporary tablespace. If you finish your query, that temporary extents will be marked as free but they still kept allocated. So over the uptime of your database your temporary tablespace will be filled by allocated (but marked as free) extents .. and so you will see it as "full". You can also check the current "real" usage of your temporary tablespace by the view V$TEMPSEG_USAGE:
http://download.oracle.com/docs/cd/B19306_01/server.102/b14237/dynviews_2164.htm#i1421319
The temporary extents will be deallocated by a database shutdown or by a database start, if the database crashed before.
If you don't get any ORA-01652 by your query .. everything is fine.
Regards
Stefan

Network board 10/100 full duplex

I have an issue with my system (solaris 8 sparc):
If I download a local file from a distant machine the file is going at about 100Mb. If I try to upload that same file then it is going at 10Mb.
I checked the cables etc but could not find anything.
ndd reports correct figures (100Mb and full duplex)
# ndd -get /dev/hme lp_autoneg_cap
1
# ndd -get /dev/hme lp_100fdx_cap
1
# ndd -get /dev/hme lp_100hdx_cap
1
# ndd -get /dev/hme lp_100T4_cap
0
# ndd -get /dev/hme lp_10fdx_cap
1
# ndd -get /dev/hme lp_10hdx_cap
1
Where should I look now?

Thank you for your help
With those parameters set, my bandwith is just worse so I came back to the previous configuration.
I tried hmeconfig (found on the web, a small perl script to display everything) and it showed:
hme0: using RJ-45 at 100 Mbps full-duplex
* RJ-45 Hardware supports: auto-neg, 100-full, 100-half, 10-full, 10-half
* Auto-negotiation is: ON
* Enabled modes: 100-full, 100-half, 10-full, 10-half
* Link Partner supports: auto-neg, 100-full, 100-half, 10-full, 10-half
with default values (nothing in /etc/system) and:
hme0: using RJ-45 at 100 Mbps full-duplex
* RJ-45 Hardware supports: auto-neg, 100-full, 100-half, 10-full, 10-half
* Auto-negotiation is: OFF
* Enabled modes: 100-full
* Link Partner supports: UNKNOWN - interface not in auto-negotiation mode
with the following in /etc/system:
set hme:hme_adv_autoneg_cap=0
set hme:hme_adv_100T4_cap=0
set hme:hme_adv_100fdx_cap=1
set hme:hme_adv_100hdx_cap=0
set hme:hme_adv_10fdx_cap=0
set hme:hme_adv_10hdx_cap=0
This sounds normal except that the second one is just slower for ftp exchanges.
I cannot understand this.

Should I disconnect the charger once the battery is 100% full. Or can I use the laptop plugged even when the battery is full., Should I disconnect the charger once the battery is 100% full. Or can I use the laptop plugged even when the battery is full.

Should I disconnect the charger once the battery is 100% full. Or can I use the laptop plugged even when the battery is full., Should I disconnect the charger once the battery is 100% full. Or can I use the laptop plugged even when the battery is full.

You can leave the charger plugged in, as you cannot overcharge...charging automatically stops when your battery is full. Once every 2 ~ 3 months, its a good idea to drain the battery to about ~ 40%, then recharge to full. Otherwise, you can charge whenever you like & leave connected whenever you like.
Try to avoid full battery discharges, Lithium batteries don't like that & frequent full discharge/recharge cycles will eventually kill your battery.

HT201436 I keep getting notifications that my voicemail is 95%/97%/100% full, but I deleted a ton but still at 95%

I keep getting notifications that my voicemail is 95%/97%/100% full, but I deleted a ton but still at 95%

Voicemail is a carrier feature. Contact your carrier.

Oracle is 100% full

I am using ECC6.0.Here I wanted to make path from Production to development.but there is problem regarding client export. in UNIX command it shows Oracle space is 100% full.So what are the steps are to be taken?
please give me a suggesation

Hi,
When the oracle 100% try to delete unwanted files or move some where.
or check if you can delete 'core' files:
find /oracle -name core -exec rm {} \;
Please check SAP note 16513 will help you.
With Regards,
Krishna.

Half full hard drive became 100% full suddenly! HELP

I have a 160gig harddrive in my MBP. I always keep my internal HD as empty and organized as possible. It was only 70 gigs full this morning, I went out to lunch, came back, and suddenly, the hard drive is 100% full!
I have no idea why it did that, after a reboot, I could not reach the desktop after I sign in. So I booted off the snow leopard os disk, and did a check in my disk utility. Says it's OK.
After, I signed on in Safe Mode, checked my backup drive (using Carbon Copy Cloner), and it didn't seem to duplicate, not even full like my internal drive.
Does anyone know what the cause for this may be? Should I just wipe out my drive, and restore my drive from my backup (I'm a little hesitant to do that at the moment, becuase I haven't backed up for a week, and there is still new work I don't want to be wiped out). Thanks for any help in advance!

Use Disk Inventory X or OmniDiskSweeper to find out what made your hard drive full.

/var filesystem 100% full

Similar Messages

Maybe you are looking for